Elektor Books - Alex Pozhitkov and BoB Gudgel - Renewable Energy at Home - 2024
Elektor Books - Alex Pozhitkov and BoB Gudgel - Renewable Energy at Home - 2024
books books
books
Renewable Energy
at Home Renewable Energy
A Hands-on Guide to
Crafting Your Own Power Plant at Home
A Hands-on Guide to
● All rights reserved. No part of this book may be reproduced in any material form, including photocopying, or
storing in any medium by electronic means and whether or not transiently or incidentally to some other use of this
publication, without the written permission of the copyright holder except in accordance with the provisions of the
Copyright Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licencing Agency
Ltd., 90 Tottenham Court Road, London, England W1P 9HE. Applications for the copyright holder's permission to
reproduce any part of the publication should be addressed to the publishers.
● Declaration
The author, editor, and publisher have used their best efforts in ensuring the correctness of the information contained
in this book. They do not assume, and hereby disclaim, any liability to any party for any loss or damage caused by
errors or omissions in this book, whether such errors or omissions result from negligence, accident or any other cause.
All the programs given in the book are Copyright of the Author and Elektor International Media. These programs
may only be used for educational purposes. Written permission from the Author or Elektor must be obtained before
any of these programs can be used for commercial purposes.
Elektor is the world's leading source of essential technical information and electronics products for pro engineers,
electronics designers, and the companies seeking to engage them. Each day, our international team develops and delivers
high-quality content - via a variety of media channels (including magazines, video, digital media, and social media) in
several languages - relating to electronics design and DIY electronics. www.elektormagazine.com
●4
Contents
For Whom is This Written? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Learn CAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Data Logger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Book Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Pi-logger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Pyranometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Theoretical Foundation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Garage Powerplant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
On the Roof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Turbine Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Conduits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Lead-acid Batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Program: Voltread . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Program: Voltlog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Hybrid inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
●5
DS18B20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
SUVT Plexiglas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
LTC3108 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
●6
Many books have already been written about various aspects of renewable energy, e.g.,
economic, political, etc. In addition, there are also books for the DIYers claiming to guide
creating an off-grid solar energy powered house. The book you are reading now was written
while we created a garage and a deck "power plants" from scratch. We present actual steps
and considerations relevant to establishing your own energy independence together with
some revealing experiments along the way.
We speak to the experimenters and geeks who would like to play with their renewable en-
ergy setup, learn fundamentals experimentally and ultimately build their own unique solu-
tion. The readers are encouraged to follow the projects described here and produce their
own variations specific to their locations / circumstances. Some of the experiments and
projects may sound like reinvention of a wheel, however we firmly believe that it is okay to
re-invent and make a couple of wheels in someone’s lifetime if it provides the experience
and in-depth understanding of the subject.
Finally, our intention was not to simply rephrase information from various sources to fill up
the pages, but rather create a new hands-on experience that will grow on its own. Where
appropriate, we provide references to published literature and stable (hopefully) web re-
sources for additional information for those who are interested.
●7
Learn CAD
Creating your own renewable energy power plant requires serious attitude and attention
to details. Even if you are not an engineer, you will have to become one, at least partially.
Hence, planning your design will require some CAD work. We are not talking about thou-
sands of dollars CAD software, rather there are wonderful CAD products like DesignSpark
Mechanical (DSM). There are multiple video tutorials bringing you up to speed with CAD.
For any DIY engineer, CAD is simply indispensable. All designs described in this book are
accompanied by DSM files.
Data Logger
We will be evaluating the performance of our solar and wind setups in terms of voltage
and current output. Also, we will monitor the amount of solar light as a function of time in
our specific location. Moderately sophisticated charge controllers will provide data logging,
some of which will even allow online monitoring. Nevertheless, monitoring the wind turbine
performance or measuring the amount of solar energy requires a separate data logger.
There are professional expensive data loggers providing several channels for voltage and
current logging. Here we suggest a cheaper and more creative solution based on Raspberry
Pi single board computer (https://www.raspberrypi.com/). Specifically, we used Raspberry
Pi Zero 2W, which is a very small single board computer with Wi-Fi. Further in the book we
provided details about setting up and deploying the data logger. Before that, learn some
basics of Linux!
Tools
One wise man’s words: "with appropriate tools one could achieve virtually anything". We
highly recommend investing in a set of tools. There is nothing more frustrating than work-
ing with inadequate equipment. Please, do not attempt assembling a project holding parts
in the air or with your feet! Get yourself a drill press, hand drill with a hammer mode,
impact driver, a set of "number" drill bits, tap, die, jigsaw, angle grinder, belt grinder and
a circular saw. All of these can be purchased over time from a home improvement stores
relatively inexpensively.
References
There is a very useful "Pocket Reference" book (1) that provides invaluable information
on fasteners and corresponding hole sizes, ampacity of wires, material properties, etc. It
is best to avoid the choice of materials and fasteners based on feelings, rather than on
●8
well-established standards. These standards came about through trial and error, and in
this case the re-invention of the wheel is not helpful. After all, inadequate fasteners may
corrode or be not strong enough resulting in a collapse of your solar or wind energy setup.
Book Organization
The fist chapter of the book proposes two DIY instruments, which we will use while building
our renewable energy system. These instruments perhaps may be replaced with the off-
the shelf analogs, however building, and using our own may enrich the whole experience.
The following chapter is devoted to photovoltaics, from theory to the actual garage "power
plant". The fourth chapter deals with wind energy and a small wind turbine, which can be
placed on the roof or deck. Another chapter describes curious power sources and suggests
building a microbial cell battery based on rabbit manure. The manure battery is quite weak,
however playing around with such power sources may be inspirational for curious research-
ers. The last chapter discusses technical aspects of wiring and battery recovery.
●9
Abbreviations
● 10
Pi-logger
A data logger can be very helpful to investigate the performance of our power plant, wheth-
er it is a microbial power source, solar panels, or a wind turbine. Specifically, the data log-
ger is used for measuring the amount of solar radiation (i.e., insolation) by a pyranometer,
as discussed in the photovoltaics chapter. Also, the wind turbine presented in the following
chapter can be optimized based on the measurements obtained by the logger. There are
many professional data loggers available on the market, which are capable of measuring
voltage and current across several channels. These solutions are quite expensive and do
not seem to offer a "1-wire" interface, which is used for temperature measurements among
other things. Here we propose a data logger, the "Pi-logger" based on Raspberry Pi Zero
2W single board computer, featuring Wi-Fi as well as several analog and digital inputs. The
schematic of the Pi-logger along with actual implementation is shown in Figure 1.
The Pi-logger runs Linux operating system (OS). Raspberry Pi web pages provide step by
step instructions about the installation and configuration of the OS. Briefly, a microSD card
must be loaded with a bootable Linux ISO image using the Rufus software. There is also
a Raspberry Pi imager software that serves the same purpose. The Linux image may be
obtained from the Ubuntu or Raspberry Pi web sites. Choose the Linux installation without
the desktop because the Pi-logger works with text-based interface, while graphics pose
unnecessary complications. During the installation, it is important to have the keyboard
and monitor connected to see the progress, to create an admin account and provide Wi-Fi
settings. After the installation, the Raspberry Pi will be used "headless", i.e., only using the
terminal software over the network. To make the system up to date, run
● 11
The Raspberry Pi Zero 2W is sandwiched with another board, MCC 118 ADC ("ADC hat",
by Measurement Computing Corporation). Mains AC voltage is conditioned using a sin-
gle-board SMPSU (available at Digi-Key) producing 12 VDC. This voltage is further reduced
to 5 VDC using a linear regulator 7805 outfitted with a heat sink. A current transducer type
CAS 6 from LEM is a magnetic transconductance amplifier based on flux core saturation
that converts current in its primary winding into the output voltage. This voltage is fed into
channel 2 of the ADC. The transducer provides 3 turns of its primary winding that can be
connected in series or in parallel to adjust sensitivity via ampère-turns. In our case, the
● 12
turns are connected in series for maximum sensitivity. Note, the current transducer works
with both DC and AC currents.
The JAE connector of the Pi-logger provides stabilized 12 V power and two inputs, digital
and analog. The digital input goes directly to GPIO 4 of the Raspberry Pi bypassing the ADC
hat, while the analog input is connected to channel 3 of the ADC. This arrangement allows
measuring solar or wind power generation characteristics using the pyranometer or a wind
sensor, respectively. The pyranometer (discussed below) requires 12 V power and produces
digital output over the 1-wire interface. A wind sensor usually also requires 12 V power and
produces analog voltage output proportional to the wind speed. The two BNC connectors
provide input for voltage measurements in the interval of –10 to +10 V. The black (–) and
red (+) spring-loaded contacts provide input for the current measurements.
A rather simple attenuator circuit was added to the Pi-logger to expand the range of meas-
ured voltages, Figure 1 inset. The attenuator’s input resistance is rather low, about 100
k-ohms, which makes the attenuator not an "ideal" instrument. Nevertheless, for our pur-
poses of measuring battery voltages, the attenuator is perfectly adequate.
Calibration
Although the datasheet of CAS 6 current sensor provides a response curve, it is impor-
tant to calibrate current measurements. We wrote a small program voltread (see Appen-
dix) that prompts the operator to input the known calibration value followed by reporting
measured voltage averaged over several measurements. The source code is shown in the
Appendix. The code is to be copied into a voltread.c file and compiled with gcc directly on
the Pi-logger. The reader is encouraged to inspect the code to understand how to commu-
nicate with MCC 118. For CAS6 calibration, the program was invoked as follows, where the
parameter -c is a bit mask selecting channel 2, and the parameter -n indicates averaging
over 10 measurements:
The attenuator was calibrated using the same program. The calibration results of the cur-
rent transducer as well as the voltage attenuator are shown on Figure 2. Note the excep-
tional linearity and reproducibility of the measurements. At zero current, the transducer
produces about 2.5 V, depending on the direction of the current, the voltage will either
increase or decrease from that value.
● 13
Pyranometer
Realistic expectations from our solar power plant (described below) will depend on how
much solar radiation is delivered to our location. The correct instrument for the job is called
a "pyranometer", which measures global irradiance. Global irradiance is the amount of en-
ergy per unit of time and unit of surface area (i.e., flux, W/m2) coming from a hemisphere
centered at the axis of the pyranometer. This type of irradiance includes direct and diffuse
sunlight. In our case the surface of interest is tilted, thus the hemispherical field of view is
essentially the sky dome!
● 14
Figure 3. Classical pyranometer, adopted from ref. 17. A black disk absorbs radiation,
heats up, and reaches a dynamic equilibrium with the heat sink. The thermocouples
produce voltage as an output. An identical disk is placed underneath in the dark for
compensation to account for the heat sink temperature.
There are many electronic pyranometers available that use a photodiode and a calibration
function to convert a photocurrent into the insolation. For us, however, such an approach
is not very interesting, because we want to measure radiation from the first principles.
Let’s build our own DIY pyranometer for the sake of experience with measurements and
understanding the underlying physics behind energy emission/absorption. Also, the DIY
pyranometer will be rather inexpensive! Our pyranometer will be like the classic one but
with slightly different principles. A schematic of the pyranometer is shown in Figure 4.
● 15
In the DIY pyranometer, the solar radiation flows through a window onto a black radiation
absorber. As a result, the temperature of the absorber increases. The temperature will not
increase infinitely, because the absorber radiates energy back through the window into
Space. This process is governed by the Stefan-Boltzmann law. Specifically, at certain tem-
perature, the amount of absorbed radiation will be equal to the amount of energy radiated
back, hence the temperature will become constant:
Eq. 1
Where
For example, e =1 for a black body and near zero (e =0.025) for polished gold [18]. It is
also important to note that good emitters are good absorbers and vice-versa. Hence, we
will not be using polished gold for the absorber, but rather a black painted sheet of alumi-
num.
The actual design of the DIY pyranometer is shown in Figure 5. The heat absorbing element
— a sheet of metal — is situated on two thermally insulating rods, which are placed into
the absorber shell — a black cylinder. The absorber shell and the window form a chamber
in which the radiation absorber only interacts with the environment via radiation but not
thermal conduction. The absorber shell is placed into a plywood box and it is subjected to
● 16
forced airflow to maintain ambient temperature (air). Therefore, the heat absorbing ele-
ment re-emits the absorbed energy through the window as well as into the bottom of the
absorber shell, whose temperature is maintained constant. The window is made of a ¼ in.
thick SUVT Plexiglas, which is transparent to a wide range of light spectrum (see Appendix
for datasheet).
Our instrument needs calibration, so a known source of energy is necessary. Since the ra-
diation absorber converts photons into its internal energy and radiates it back, it does not
matter how energy is supplied. The simplest and the most accurate way is to electrically
warm up the absorber with a tape heater on the back side of the absorber. The electrical
power applied to the heater will produce the calibration flux w = I × V/A, where I – current;
V voltage; A area of the absorber.
Figure 5. A sketch of the DIY pyranometer (left) and its cross section (right). For
calibration, a film heater is placed on the underside of the radiation absorber. See CAD
file pyranometer.rsdoc for dimensions and details.
An interesting question is what size shall we choose for the radiation absorber? First, as
much as possible we want to cover the inner space of the absorber shell, which can be a
4" sewer pipe cap. A reasonably good dimension is 2.75 × 2.75". What about the thick-
ness? Obviously, the sheet must be thick enough to sustain its own weight as well as a
temperature sensor attached to it. The upper limit of thickness is determined by the desired
response rate. Indeed, the time course of the temperature as the absorber approaches the
equilibrium is as follows:
In this equation, the first term is the incoming energy onto the area A of the absorber; as
mentioned above, the second term is the radiation loss from various surfaces of the absorb-
er, which may have different areas and emissivity coefficients (Ai, ei), e.g., one side is black
painted while the other side is bare metal. The rightmost term shows the rate of change
of the temperature multiplied by the heat capacity Cp and the total mass of the absorber,
Ahr, i.e., the product of area, thickness, and density, respectively. For the initial modeling,
it is safe to assume that only two major areas of equal size A (top and bottom) are emitting
the radiation. Hence, the A may be cancelled; the final differential equation is as follows:
● 17
Eq. 3
his equation does not have a quadrature solution; therefore, we will solve it numerical-
ly using Wolfram Mathematica (Wolfram Research Inc). The desktop version of the soft-
ware is rather expensive, but a cloud account is free. In Mathematica, the computational
workspace is called a "notebook". As mentioned at the beginning of the book, the reader
will have to become an engineer to a certain extent; hence learning technical computing
products like Mathematica is highly recommended. One important characteristic of Math-
ematica is its "World knowledge". For instance, density and specific heat of aluminum can
be directly obtained within the notebook by a function call. Also, Mathematica is capable
of computing with physical units, which is very helpful to making sure the results make
physical sense. The numeric solutions for w in the range of 250-1000 W/m2 are shown in
Figure 6. The parameters used in the calculation are as follows: Ta = 293 K; e1 = 1; e2 =
0.9; h = 0.01"; Cp = 904 J/(kg.K); r = 2700 kg/m3. The notebook pyranometer_model.nb
is provided as a supplement.
The time course curve suggests that with thickness of 0.01", the equilibrium will be reached
within approx. 200 s, which is very quick and acceptable for our purposes.
What about the heater power? Given the desired calibration energy flux of approx. 1000
W/m2, an easily available film heater of 2 × 2" (2.581.10-3 m2) would have to be at least
2.6 W.
Pyranometer construction
Dimensions and the detailed 3D model of the pyranometer are provided in the DesignSpark
Mechanical CAD file pyranometer.rsdoc. We begin the construction process from the radia-
tion absorber, Figure 7 shows all the steps involved. First, cut a 2.75 × 2.75 × 0.01" sheet
of aluminum. Second, make a heat conductive clamp (brass) to hold the digital temper-
ature transducer DS18B20 (purchased from Mouser; datasheet in Appendix). Unlike the
● 18
traditional pyranometer that uses thermopiles and voltage output, the digital output has
two main advantages. First, given that the temperature digitizing is taking place within the
DS18B20 device itself, we do not need to worry about the noise in the transmitting cables.
Second, there is a plethora of thermometers compatible with DS18B20, which features "1-
wire" interface including our Pi-logger.
The clamp can be made from a 0.25" wide strip of 0.016" thick brass (available at hardware
/ hobby stores) that is pressed into a wooden die to form a u-shaped profile (Figure 7A).
The die is a block of wood of reasonable dimensions, e.g., 2 × 1 × 1", having a hole drilled
in its center with a drill #9 (0.196") and cut in half across the hole. Use a shank of the
drill #20 (0.161") as a punch and apply pressure with a vise. By the way, such clamp may
be handy to other devices in TO-92 package. Attach the temperature transducer with the
clamp and a pair of #4 screws and nuts. On the other side of the radiation absorber, place a
20 W poliimide adhesive heater (Kapton), size 2 × 2". Finally, paint the radiation absorbing
side with black spray paint and glue the absorber onto the two 0.25" diameter polystyrene
rods held in the absorber shell by friction. The absorber shell is a 4" sewer pipe end cap. In
the middle of the cap, place another DS18B20 to measure the shell’s temperature. The cap,
rods, aluminum sheet and the heater may be obtained from McMaster-Carr.
Figure 7: Steps for making the radiation absorber. A: sheet of aluminum, temperature
transducer DS18B20, brass mounting clamp; B: radiation absorber with temperature
transducer attached; C: Kapton calibration heater adhesively attached on the side
opposite to the thermometer; D: black painted radiation absorber and the shell into which
the absorber is to be mounted on the polystyrene posts.
● 19
Finally, the absorber-shell assembly is inserted into a box (plywood or plastic) covered with
a back wall outfitted with a fan and two "exhaust" copper pipes. The box should be paint-
ed white to minimize absorption of solar radiation. The window is made of a SUVT acrylic
(Figure 8). Route all the wiring to the outside using tie posts and / or a pin connector, e.g.,
JAE, yellow and grey respectively in Figure 8.
Figure 8: Final assembly step of the pyranometer. Note the copper pipes for
venting air, which are to be inserted inwards of the pyranometer box.
● 20
Calibration
After the pyranometer is assembled, it is time to calibrate it and assess if our physical mod-
el is correct. In fact, we will do two calibrations: one is to calibrate the heater and two is
to calibrate the measured temperature versus applied known heat flux. The importance of
calibrating the heater is to be able to use just the voltage as a measure of provided energy,
rather than measuring both voltage and current. We record characteristics of the heater,
which is powered from a variable transformer as follows, Table 1.
… … … …
Table 1: Voltage, current and heat flux of the 4 in2 film heater.
As well-known from the Joule–Lenz law, heat output is directly proportional to the square
of applied voltage. Graphically we see a perfect straight line, as expected.
● 21
Knowing the heater characteristics, we can now finally calibrate the pyranometer. We will
simultaneously measure the temperatures of the radiation absorber and the temperature
of the shell. Using the Pi-logger it is easy to measure temperature of both thermometers
simultaneously rather frequently.
As we mentioned earlier, the reader will have to become an engineer. Therefore, learning
Linux shell and setting up Raspberry Pi would be very helpful. To enable Raspberry Pi to
read the DS18B20 thermometers, the data line (Figure 9) must be connected to GPIO 4
and pulled up to +5 V with a 4.7 k-ohm resistor. Grounds of the pyranometer and Pi are
connected as well. Config.txt is to be modified by adding the following line:
dtoverlay=w1-gpio;
It is important to make sure there are not more than 80 lines in the that file! Restart. Issue
the following commands:
Shut down. Connect the pyranometer. Turn the Pi-logger on. Notice in the directory /sys/
bus/w1/devices/ a subdirectory looking like 28-0000000aabc2. This is the internal ID of the
thermometer. Since the pyranometer has two thermometers, two of such subdirectories
will appear. For all subsequent restarts of the Pi-logger, there is no need for more modprobe
commands. The following code makes 230 measurements about every 5 seconds:
● 22
The data will be stored in templog.tsv and will be visible at the standard output (thanks to
the tee command). The records will look like this:
The first number is the number of seconds since 1970-01-01 00:00:00 UTC (quite a big
number!); the second is the trial number, e.g., 4; the third is the temperature of the energy
absorber in oC multiplied by 1000; the fourth is the temperature of the shell.
Let us apply several voltages to the internal heater of the pyranometer to investigate
the heating and cooling characteristics of the pyranometer. A critical characteristic of our
instrument is the response time, which we estimated earlier (Eq. 3, Figure 6). Actual re-
sponse time is shown on Figure 11.
As one can see, after about 200 s, the temperature increase begins significantly slowing
down. But unlike our model, the transition is not as sharp as calculated. This is the differ-
ence between theory and reality. It is not very practical to wait for the equilibrium to be
reached, especially in the field, because the position of the Sun is changing throughout the
day and clouds may obscure the sunshine for some time.
Instead of waiting for the equilibrium, we can use the dynamic behavior to instantaneously
measure the energy flux by rewriting Eq. 3 as follows (Eq. 4):
Eq. 4
● 23
Where a and b are fitting parameters. The temperature derivative at a certain time t is
easy to measure by fitting a straight line through 4 points forward; the slope of this line is
the derivative (Figure 11 inset). is measured directly. Fitting for a and b is done using the
least squares method. Once the parameters are determined, the energy flux is calculated
by solving Eq. 4 for w. In our calibration experiments we took about 20 equally spaced
time points from three temperature curves (Figure 11) corresponding to fluxes 696.1,
0.0, 424.2 W/m2, performed fitting, calculated a and b. See Appendix for the Mathe-
matica notebook pyranometer_calib.nb and templog.tsv. Based on the fitted parameters,
a=0.000427035, b=8.55887*10-11, we back calculated w in all time points and for all flux-
es, including 424.2 W/m2. This test resulted in excellent agreement between calculated and
expected energy fluxes, Figure 12, even in the case of 424.2 W/m2, which was not used in
"training" of the model. This result also indicates that our model accurately captures the
physics of the process.
Thus, the calibrated pyranometer may be used to record the irradiance as a function of time
during the day, in different seasons and at various cloud coverage.
Figure 12: Back calculation of the energy flux w based on parameters a=0.000427035,
b=8.55887*10-11 and measured T(t), T(t)a, and dT(t)/dt .
● 24
Theoretical Foundation
The purpose of this section is to provide a springboard for understanding the processes
taking place in PV. The physical model of the PV foundation is quantum mechanics, which
formally describes the processes and provides necessary calculations. Interested DIYers
are advised to take a course of quantum mechanics, either at school or from multiple free
resources, e.g., MIT Open Courseware that was available at the time of writing. Our ga-
rage power plant will not require deep understanding of quantum mechanics though. Also,
PV Education resource (https://www.pveducation.org/) provides a detailed application of
quantum mechanics and solid-state physics to PV.
Photovoltaic effect
This effect was first discovered by a French scientist E. Becquerel, in 1839 who first de-
tected a generation of photo-current in an electrochemical cell (11). The cell was made of
an acidic solution, 2 platinum plates and a membrane separating the cathodic and anodic
zones. Illumination of one of the plates produced an electrical current. Becquerel also no-
ticed an effect of the light color on the strength of the generated current.
Historically, the findings of the PV effect led to another one, the photoelectric effect, dis-
covered by H. Hertz during his experiments in radio wave transmission, further investigated
by W. Hallwachs and A. Stoletov. Specifically, Hallwachs (15) charged a zinc plate attached
to an electroscope having golden foil leaves as charge indicators. Upon illumination of
the plate with an arc lamp, magnesium flame or a stearin candle, different effects were
observed. The electroscope connected to the positively charged plate showed very slow
discharge (several minutes) when illuminated with an arc lamp or magnesium. To the con-
trary, negative charge quickly (seconds) disappeared when illuminated with the same light
sources. Candlelight failed to produce any charge dissipation of either positive or negatively
charged plates. Hallwachs concluded that it was mostly the ultraviolet light that caused the
discharge. A theoretical explanation of the photoelectric effect was given by A. Einstein.
The photoelectric effect revealed a fundamental property of photons that the energy of
photons is dependent on their frequency, rather than the intensity of light. The equation
reads E = hν, where E is energy, h – Planck’s constant and ν – frequency. This leads to the
following: you may have a very intense light source, but if the frequency of photons is not
high enough, no photoelectric effect will be observed.
Bandgap
The photoelectric effect brings us to the basic principle behind the light-induced electricity
generation by a solar panel, which is essentially a large collection of connected photodi-
odes. The energy of photons is absorbed by the electrons that brings them to another en-
ergy level allowing them to cross over the bandgap. The bandgap is the energy difference
of the electron states corresponding to conducting and non-conducting electrons, respec-
tively. The conducting electrons form a so called "electron gas", while the non-conducting
electrons are tightly bound to the nuclei. In metals, the bandgap is very small compared to
insulators, while the bandgap in semiconductors is in between. Various PV materials have
characteristic bandgaps.
● 25
The importance of the bandgap for the PV materials is that it directly determines the effi-
ciency of solar energy conversion. Efficiency of light conversion is defined as a power ratio
Wout/Win produced electrical energy over the incoming energy of photons per unit of time.
Figure 13 shows a theoretical calculation of a single-junction silicon-based PV element. One
of the curves on the left panel has large dips; this curve corresponds to the solar spectrum
at sea level. (More recent measurements of the solar spectrum are provided in Gueymard
et al., 14). The spectrum is the photon flux intensity i.e., the number of photons per unit
cm2 per sec., as a function of the photon energy. The yellow shaded area is the total
number of photons with energies above the bandgap (e.g., 1.35 eV for silicon). Now let
us switch to the right panel, whereby the Y coordinate of each point on the "100%" curve
represents the area of the yellow shading as a function of a bandgap energy value. The blue
shaded area under that curve is the total energy delivered by the photons with energies
above the bandgap. Unfortunately, there is an intrinsic loss due to radiative recombination.
The PV element not only absorbs, but also emits light. The rate of the radiative emission
exponentially increases with the voltage developed across the load (determined by the laws
of thermodynamics and statistical mechanics). The emission results in the equivalent of
"radiative current", which subtracts from the current through load. At the maximum power
point (MPP, discussed below), the effective curve (formerly 100%) shifts to the left, which
is labeled "W" in Figure 13, right panel.
Figure 13: Efficiency of a single-junction PV element with a bandgap Eg=1.35 eV. Left
panel, curve with dips: solar spectrum in the Northern Hemisphere, sea level. Right
panel: Y coordinate of each point of the "100%" curve is the area under the solar
spectrum curve (yellow shading) as a function of a given bandgap energy. Blue shading
indicates total solar energy flux with photons having energies above the Eg value. See
text for details. Adopted from Henry 1980 (12).
● 26
All is left is the white rectangle corresponding to 31% of the solar energy that can be con-
verted into electricity. It is important to understand for the material with some other value
of the bandgap, the white shaded rectangle will be different. If several junctions are sand-
wiched together, efficiency may be drastically increased, however such PV panels do not
seem to be commercially available at present. Moreover, such cells would require a concen-
trator (lens or mirror), which is not practically feasible at home. A recent review in the Sci-
ence magazine provides a survey of existing PV materials, bandgaps and efficiencies (13).
Solar panels
For our garage power plant, we chose Renogy RNG-100D-SS monocrystalline 100-watt
solar panels. It is important to understand that 100 watts are produced in standard test
conditions, i.e.,
• MPP
• irradiance 1000W/m2
• temperature 25 oC
• AM 1.5
MPP is a condition on the I-V curve, at which the solar panel delivers maximum power to
the load. Specifically, MPP for Renogy RNG-100D-SS is shown in Figure 14. What is AM?
Figure 14: MPP corresponds to the maximum on the power curve (green).
Adopted from Renogy RNG-100D-SS datasheet.
● 27
RNG-100D-SS
100W Monocrystalline Solar Panel
Key Features Potential Uses
Sleek design and a durable frame, the Renogy 100 The Renogy 100 Watt Monocrystalline Panel can be
Watt 12 Volt Monocrystalline Panel provides you used in various off-grid applications that include 12
with the highest efficiency per area and is the perfect and 24 volts arrays, water pumping systems, signal-
item for off-grid applications. ing systems and other off-grid applications.
● 28
RNG-100D-SS
100W Monocrystalline Solar Panel
Electrical Data Mechanical Data
Maximum Power at STC* 100 W Solar Cell Type Monocrystalline (6.1 x 4 in)
Optimum Operating Voltage (Vmp) 17.9 V Number of Cells 33 (3 x 11)
Optimum Operating Current (Imp) 5.72 A Dimensions 42.2 x 19.6 x 1.38in (1074 x 498 x 35mm)
Open Circuit Voltage (Voc) 21.6 V Weight 14.3 lbs (6.5 kg)
Short Circuit Current (Isc) 6.24 A Front Glass Tempered Glass 0.13 in (3.2 mm)
Cell Efficiency 21.0% Frame Anodized Aluminium Alloy
Maximum System Voltage 600 VDC UL Connectors Solar Connectors
Maximum Series Fuse Rating 15 A Fire Rating Class C
Power (W)
Voltage (V)
*All specifications and data described in this data sheet are tested under Standard Test Conditions (STC - Irradiance: 1000W/m2 , Temperature: 25 º C,
Air Mass: 1.5) and may deviate marginally from actual values. Renogy and any of its affiliates has reserved the right to make any modifications to the
information on this data sheet without notice. It is our goal to supply our customers with the most recent information regarding our products. These data
sheets can be found in the downloads section of our website, www.renogy.com
● 29
(14). A diagram in Figure 15 represents these standardized conditions, resulting in the path
length through the atmosphere of 1.5 times longer relative to zenith. The importance of the
atmosphere is that part of the light will be scattered. The extent of scattering is dependent
on wavelength, the shorter the wavelength the higher the scattering. Remember the dawn
and dusk colors? The blue portion of light is scattered as it travels through the long path
in the atmosphere, as it occurs at sunrise and sunset. The transmitted light through the
thickness of the atmosphere follows the Beer-Lambert’s law,
where I is the light intensity (W/m2) as it is absorbed / scattered in the medium (e.g., air),
at the distance L from the point where the intensity is I0. a is the "extinction coefficient".
The extinction coefficient is dependent on the medium and the wavelength of the light.
Figure 15: Standardized light path calculation in North America at the average latitude
of 37O. Average sun elevation above the horizon is 41.81O, resulting in 48.19O angle
between zenith and the light path.
Light management
Interestingly, the same law plays an important role in the way the solar panel absorbs
light for conversion into electricity. According to PV Education (https://www.pveducation.
org/), the extinction coefficient of the monocrystalline silicon at 550 nm (peak of the so-
lar spectrum at AM 1.5) is 10 times larger than that of the amorphous silicon. Hence, the
● 30
monocrystalline silicon absorbs more photons than its amorphous counterpart over the
same depth from the surface into the material. Also, the surface of the solar panel is rough-
ened to create ridges and valleys enabling multiple light interactions through reflection
therein. Finally, the surface is coated with an anti-reflective coating sending the light back
to the surface (Figure 16). These approaches are called "light management" that intend to
increase L — the optical path — in the Beer-Lambert’s equation. The visible effect of the
light management is that the panel looks dark due to effective absorption of the light.
Garage Powerplant
As we finished with the theoretical foundation, it is time to create our own powerplant! We
will use Renogy RNG-100D-SS monocrystalline 100-W solar panels, which are very light
and easy to handle. Renogy provides short (about 4 inch) mounting brackets, which work
well on an "ideal" surface — flat and smooth. However, if the roof is covered with tiles,
one needs a different type of mounting. We designed a bracket based on a long aluminum
z-bar. The long bracket allows locating the best tethering position, i.e., the one with small-
est gap, on the uneven tiles (Figure 17).
● 31
Figure 17: Solar panels mounted on an uneven surface with z-bar brackets.
Figure 18 shows all necessary steps to attach the mounting brackets. First, drill 1/8" in.
pilot holes in the panel frame and z-bars. Second, use drill #9 to enlarge the z-bar holes.
Attach the bracket to the panel with #8–1/2 galvanized sheet metal screws. The sheet
metal screws cut their thread in the 1/8 in. holes of the panel frame.
● 32
● 33
On the Roof
While working on the roof, use appropriate fall protection gear. Usually, the fall protection
is a thick nylon rope with a carabiner on one end and a lanyard with a sliding grip. The car-
abiner is attached to a strong anchor on the roof, while the lanyard has a carabiner as well,
which connects to the harness. There is also an energy absorbing element that decelerates
the body during the fall thus minimizing the whiplash.
● 34
SPECIFICATIONS
Tested & Certified as Compliant per ANSI Z359.1-07, ANSI A 10.32-12 Regulations for Fall Protection
Universal
Harness
3.5 Gallon
Storage
Bucket
50 ft. Reusable
Lifeline Temper
Assembly Anchor
with
nails/screws
The equipment is designed for use by persons with a combined weight (clothing, tools, etc.) of no more than 310 lbs.
No more than one personal protective system may be connected at one time.
www.gibraltarbuildingproducts.com
Place the panel with attached mounting brackets onto the roof in a desired location. Figure
20 illustrates the steps involved. The mounting brackets provide several holes for locating
the optimal fastening points (as an example, see red markings on Figure 20). Drill holes
with a 5/32 masonry bit and a hammer drill. Fill the holes with a waterproofing material, for
example, Gardner-Gibson Black Jack Neoprene Super Flash Cement. Fasten the panels on
● 35
the roof with screws of sufficient length, providing the screw penetration through all layers
of the roof (e.g., #10–3-1/2 Hillman multi-material screw).
Figure 20: Placement of PV panels on the roof. Top to bottom: optimal holes located,
marked red; holes drilled, waterproof cement added; panels screwed on. Observe black
waterproofing cement around the screws.
● 36
It turns out that finding an optimal configuration is not as simple as you may think. A re-
view from 2018 (16) covers the issue extensively. The takeaway from this analysis is that a
PV array configured as one string (series) will have several maximums on the power-volt-
age curve, unlike the one depicted on Figure 14. Such curve poses a serious difficulty for
MPP tracking because the algorithm will likely get stuck in one of the maximums. Malathy
et al. (16) provide references for interesting solutions that use neural networks and heuris-
tics to search for the maximum, which is currently more academic than practical. A parallel
configuration of the PV array will have only one maximum, same as on Figure 14, however
as discussed above, high current poses a disadvantage. Apparently, the ideal configuration
ensures homogenous distribution of the shaded panels across the array, see Figure 21 bot-
tom. In this case, only one peak on the P-V curve is expected, hence the charge controller
will find MPP without problems.
● 37
Figure 21: Parallel, series and parallel strings connection configuration of a 9-panel PV
array. Grey shading indicates partial shading of the PV array. Bypass diode is connected
in parallel with a PV element allowing the current to pass thru in case of partial shading.
When the diode is active, it is marked red.
Here is the specific shading pattern of the PV array of our garage powerplant (Figure 22).
The study was done on Aug 28-29. On these days in Los Angeles, CA the sunrise was at
6:25, sunset at 19:23. Interestingly, the eastern field of the roof is shaded for almost 5
hours after the sunrise. This is because of a big oak tree on the property that provides
a nice shade for the main house and casts some onto the garage. We opted out of the
chainsaw solution in favor of tolerating some power loss. The shading pattern suggests
an optimal configuration of 3 strings 3 panels each. The east-field panels should be mixed
● 38
in with west-field panels to ensure that in the morning and in the afternoon each of the 3
strings has just 1 panel shaded. Better yet, one of the panels from the east field may be
moved to the west.
After all connections have been made, rout a cable rated for outdoors, for instance SOOW
(oil- and water-resistant insulation, 600 V), through the "service cap" into the building (Fig-
ure 23). If all 9 panels work at the MPP, the 3 strings in parallel will produce 12 A. For this
current, the cable should carry #14 or #12 AWG wires. As mentioned in the Introduction,
refer to the Pocket Ref (1) for determining the ampacity of the wires and insulation codes.
● 39
Inside of the building, set up a charge controller, inverter, and a load center with circuit
breakers. A schematic of a possible arrangement is shown on Figure 24. S1-S3 are the
overcurrent protection circuit breakers. In our case, we established a 5-kW system with the
MN5048DIY charger-inverter (see Appendix for datasheet and manual). Importantly, this
inverter can take DC and AC input from the grid or a generator. The inverter must also be
connected to the grid, such that when the batteries discharge below a certain threshold, the
grid would take over to provide power to the garage as well as charging the batteries. Here
comes the issue of sizing our power plant: the battery capacity should be sufficient to hold
overnight without switching over to the grid. Yet, the battery capacity must be matched
with the PV output to be able to charge them during the day.
The circuit breakers S1 and S2 are rated for DC current, 20 A and 125 A, respectively. Note,
such breakers could not be found at a regular Home Improvement store at the time of writ-
ing. However, there are plenty of inexpensive online options. Also, these breakers will not
fit standard house subpanels, therefore a box with a DIN rail is required. In contrast, S3 is
a regular AC breaker rated for 50 to 60 A, which should be installed in a separate panel to
distribute usable 120 VAC.
● 40
The finalized PV system is shown on Figure 26. The "PV inlet" box is where the PC through
the wall is where the PV cables are coming into the building through the service cap and
its conduit.
The box where the PV power comes into the building is a perfect place for adjusting the
configuration if multiple arrays exist on the roof. For example, as discussed above, some of
the arrays may be connected in series, while others are in parallel. The last chapter Con-
duits and Batteries provides a good starting point for understanding and choosing conduits,
boxes, and conduit bodies.
● 41
Figure 26: Finished photovoltaic system configured to generate and distribute usable
power at 110 VAC. Grounding wire (GND) leads to the physical ground outside of the
building. Note a small device designated as EQ — an optional battery equalizer.
● 42
Let’s use the pyranometer to measure insolation near the solar panels. The pyranometer
was placed on a large plywood sheet, which was secured on the roof. The plywood provided
a flat base for the pyranometer. The Pi-logger was used to collect the data throughout the
day using a command that called for a script to run in the background:
#!/bin/bash
counter=0
while true; do
for i in {1..10000}; do sleep 3; echo -e `date +%s`’\t’`cat /sys/bus/w1/
devices/28-0000000aabc2/temperature`’\t’`cat /sys/bus/w1/devices/28-0000000cb89a/
temperature`; done > templog$counter.tsv
counter=$(( counter + 1 ))
done
The obtained temperature measurements were converted into the energy flux according
to the pyranometer calibration, Figure 27. Interestingly, for almost the whole morning the
panels did not receive any significant sunshine. This is due to the shadow cast by a big tree
in our yard, which we are not planning to cut!
How much electrical energy, watthours, can we expect to get from our panels given the
measured insolation? To answer this question, we need to
Using numerical integration in Mathematica and the solar panel datasheets, we estimate
about 5 kWh. This amount of energy should be sufficient for the four 12 V 80 Ah batteries,
which require 3.8 kWh. In fact, the charger-inverter consistently reported batteries fully
charged in the afternoons of the sunny days in August 2023.
● 43
Charging regime
Initially we planned our power plant to be 100% off-grid and did not connect the hybrid
inverter to the grid at all. After the system was put into production, we were surprised to
see the batteries seemingly lose their capacity rather quickly. Without any significant load,
after a couple of months, each morning the system sounded an alarm indicating voltage
below 44 V and the inverter output at zero. Assessing the battery conditions by measuring
electrolyte density and capacity revealed significant degradation. In the last chapter of the
book, we discuss how to assess battery capacity and how to revive a lead-acid battery,
while it is still possible.
A presumed reason for the battery degradation was allowing batteries to deep discharge
and daily charge / discharge cycles. Anecdotal evidence from a friend who is running a sim-
ilar size off-grid system with a lithium battery set revealed a progressive loss of capacity
after about 6 months of production use. The battery was discharged 50% nightly and re-
charged fully the following day. After 6 months, the battery lost about 25% of its capacity.
To save the batteries, we connected the hybrid inverter to the grid with the intent to switch
to the grid when the batteries discharge to a specific level. The following parameters were
set on the hybrid charger / inverter:
- Inverter priority mode: the inverter switches to the grid when the battery voltage
is below a threshold.
- PV charging priority: charging predominantly from the solar. When the system is
switched to the grid, then charge from the grid as well.
● 44
- Battery undervoltage alarm, 48.4 V. It turned out that the battery / grid threshold
and the undervoltage alarm settings were linked together. (The reader’s equip-
ment may be different).
- Battery undervoltage recovery point, 52.0 V. As the voltage reaches the recovery
point, the system goes off grid.
The last parameter is critical, because at night we do not want to charge the batteries
quickly from the grid with high current. If we did allow rapid charge from the grid, in the
night the system would switch off the grid and return to using the batteries. After that, the
batteries would discharge again making the system switch back to the grid. Such cycling
would result in battery wearing out. Note, this parameter only affects AC charging; during
the daytime there is no restriction on the PV charging current.
● 45
A small "backyard" turbine in a windy area may be a great source of energy, if properly
used. Interestingly, a cursory look at the customer comments on Amazon reveals that the
turbines receive mixed reviews. The main criticism is that upon actual measurements, a
turbine produces insufficient power. Why is this happening? This is because for a given
wind speed, the performance of the turbine nonlinearly depends on its RPM, which in turn
depends on the load. Let us dive into some theory to understand this "inconvenient" phe-
nomenon so that we could properly harvest energy from the wind.
Eq. 5
The turbine blades of the lengths r cover a disk-shaped area of πr2, such that the total mass
of air coming through this area per second is ρπr2v, where r is air density. This is because
in one second the air extends behind the blades into a "solid cylinder" of length v and base
area πr2. The volume of the cylinder is πr2v. Putting the expressions together, we can obtain
the total amount of energy delivered by the wind per second, i.e., power (Eq. 6):
Eq. 6
We can already see that the bigger the blades, the faster the wind and the denser the air,
the more energy is carried. Unfortunately, the turbine cannot extract the total power due
to several factors, i.e., aerodynamic efficiency, mechanical loss, and Joule (heat) loss in
the windings. The aerodynamic efficiency is quite nontrivial: it originates from lift, drag
and wake turbulence. The aerodynamics of the blades determines a fraction of wind power
extractable by the blades [19]. This fraction is called a "power coefficient", which depends
on the "tip speed ratio" (TSR). TSR is an important concept to understand, as it measures
how fast the blade tips are moving with respect to the wind (Eq. 7).
Eq. 7
where f is the number of revolutions per second. Interestingly the power coefficient is a
curve with a maximum, and it cannot exceed 0.59 (the Betz Limit). Figure 28 shows an ex-
ample of the power coefficient for some blades. Note, it does not even reach the Betz Limit.
● 46
We can appreciate the fact that the extraction of energy from wind is complicated by the
interplay of various losses and the power coefficient. Therefore, each turbine needs to be
characterized experimentally, e.g., see ref 20. Earlier we have conducted similar experi-
mental research with a Chinook 200 wind turbine (MidNite Solar Inc.). The electrical char-
acteristics of the stator were determined by turning the turbine with a motor on a special
stand. The aerodynamic characteristics were measured by driving the turbine on a truck
at various speeds to generate a range of wind speeds [21]. In that study, the power coef-
ficient curve was reproduced and, importantly, the voltage of the optimal energy transfer
was determined (Figure 29).
● 47
Figure 29: Experimental characterization of the Chinook 200 wind turbine. Top: power
coefficient dependence on TSR; middle and bottom — optimization of loading "battery"
voltage for the 24 and 12 V stators, respectively. Figures adopted from Ref. 21.
● 48
The peak of the power-loading voltage curve is readily understandable. Indeed, at a low
voltage, the current is high, the opposing magnetic field is high, and the turbine turns
slowly. As it turns slowly, a large portion of air mass passes between the blades without
delivering any energy. On the other hand, if the voltage is too high, the turbine must reach
high RPM to be able to transfer charge to the batteries. At high RPM, the blades are sub-
jected to a significant drag.
Turbine Setup
The Chinook 200 turbine characteristics are as follows: rotor diameter 3.28’ (1 m), rat-
ed output 200 W at 31.3 mph (14 m/s) wind speed, weight 14 lbs, start-up wind speed
7.8 mph (3.5 m/s). The turbine has a swivel mounting mechanism for a tubular mast. This
is a critical part of the turbine that requires attention. Specifically, the generated current
is passed through a slip ring mechanism that connects stationary tube and wires with the
swiveling turbine, Figure 30. We advise inspecting the slip ring mechanism of your turbine
to make sure it is lubricated. We found that Magnalube grease (Magnalube LLC, USA) works
very well, especially in the location with saltwater mist (Ocean coast). Interestingly, this
grease is not marked as electrically conductive, nevertheless no conductivity issues were
encountered. If after some time the turbine does not swivel well, it is time to inspect the
slip rings!
Figure 30: Slip ring mechanism at the base of the turbine. The Magnalube green grease is
used for lubricating the contacts. Also notice a 3-phase bridge rectifier on the left.
● 49
After ensuring lubrication, it is time to mount the turbine on a mast. One possible option
for the mast is a pair of thick walled 5-ft long 1-1/2" galvanized conduit pipes, ID 1.62",
OD 1.9" (from McMaster-Carr). The pipes are connected via a tee conduit body, through
which the wires exit. In a real-life test, the arrangement being described here withstood
wind gusts of 60 mph, which occurred in August 2022 in the San Diego, CA area. Figure
31 shows the mounted turbine along with several accessories, which will be discussed ac-
cordingly.
Figure 31: Top, the Chinook turbine mounted onto a mast made of two Rigid conduit
pipes joined by a tee; grey junction box houses a brake switch. Bottom left, inside the
weatherproof garden box containing the battery, load control circuit, diversion load, and
inverter. Bottom right, weatherproof power outlet mounted on the side of the box.
● 50
The electrical arrangement is shown in Figure 32. The Chinook 200 turbine provides a built-
in full-wave rectification (Figure 30); other turbines may be single or 3-phase AC genera-
tors. The pulses must be smoothed by a large capacitance. The resulting DC is managed
by a PWM load control circuit. The output of the load control circuit is fed into the battery,
which is shunted by a PWM diversion load. The diversion load measures battery voltage
and drains excess of charge (via a resistor), when the voltage exceeds a set limit. Charge
draining is necessary for protecting the battery from overcharging and keeping the turbine
loaded. One cannot simply disconnect the battery, because an unloaded turbine would spin
up too fast causing mechanical damage. There are times when the wind or wind gusts are
too strong, and it is advisable to disable the turbine. To this end, a switch "Brake" is added
to the design that short circuits the turbine. A shorted turbine spins very slowly due to the
opposing strong magnetic field.
Figure 32: Overall electrical arrangement of the wind turbine setup. From left to right:
turbine AC generator, bridge rectifier (built into the turbine), brake switch, load control
circuit, diversion load, battery, inverter.
The purpose of load control circuit is to maintain optimal wind energy harvesting by reach-
ing maximum on the Power coefficient – TSR curve, as discussed above, see Figure 28, Fig-
ure 29 and ref. 21. The earlier experiments used programmable DC load, which maintained
set input voltage and converted energy into heat. The load control circuit is essentially a
buck converter [22] that maintains input rather than output voltage. The other very impor-
tant role of the lad control circuit is to provide flexibility with respect to the voltage of the
batteries. The same turbine may charge 12 or 24 V batteries, because it "sees" a specific
voltage set by the load control circuit regardless of the batteries.
Figure 33 shows the schematic of the load control circuit as well as an oscillogram of some
interesting test points. Full-wave rectified turbine voltage is stored in C1 – a large electro-
lytic capacitor bank, 3× 270-µF, 200-V rated. The turbine voltage is monitored by the gate
driver U1, which has sufficient hysteresis, i.e., turning on at 2.1 to 3.0 V, turning off at 0.8
to 1.5 V. To measure relatively high turbine voltage and to maintain the control window of
about 1.5 V, we used a series of Zener diodes, whereby individual diodes may be shorted.
Zener diodes effectively subtract a large voltage offset while the reminder appears on R3.
Such an arrangement does not require a voltage divider, which would significantly widen
the voltage control window. The L2R4C4 network acts as a low pass filter eliminating noise
from the turbine voltage for precise measurements. As soon as the turbine voltage reaches
the set level, the gate driver U1 turns on, two parallel FETs Q1, Q2 begin conducting and
the current through the battery begins rising. The turbine and C1 voltage begin dropping
● 51
and continue to do so until about 1.5 V below the set level, see the oscillogram at the
bottom of Figure 33. After that, the process starts over. The values of C1 and L1 were cal-
culated such that the switching frequency is rather low, hundreds of Hz. The gate driver is
powered from a linear regulator 7812, which is in turn powered by a voltage clipper built
around Q3. The linear regulator allows max 40 V at the input, which may be too low if we
decide to operate the turbine at voltages higher than 40 V. The clipper is a common collec-
tor topology that uses a high voltage transistor KSP44, which can sustain 400 V. The clipper
limits the voltage around 20 V.
● 52
Figure 33: Load control circuit. Top, the "McDonald’s" sign indicates full-wave rectified
turbine output. DIP switches shorten selected Zener diodes to maintain a desired voltage.
Bottom left, an oscillogram of the C1 voltage during the turbine rotation. Bottom right,
voltage measured across the battery over ~ 1 hour.
● 53
After deploying the turbine into production, optimization steps can be taken. Specifically,
voltage setting of the load control circuit can be adjusted to obtain maximum performance
of the turbine. Using the Pi-logger, we conducted short-term measurements of the battery
voltage using the voltlog program (see code in the Appendix).
The program averages (-a) voltages over 50000 measurements (-n) at the sampling rate
50000 samples/s (-r). Hence, the output occurs every second. To stop the program, a sig-
nal must be sent to the process id, for example:
Figure 33 bottom right shows measurements at two different load control circuit settings,
20 and 24 V. It seems that at 20 V, more energy transfer occurred. Also, at 24 V, the tur-
bine rotated considerably faster, and a whistling noise was heard. Whistling is a sign of
turbulence, which leads to power loss. In all fairness, 1 hour observation per trial is not
enough for proper optimization, because the wind speed was likely quite different in the two
observation periods. One would have to monitor the performance for a week or so, which
is easy to do thanks to the Pi-logger.
Special thanks to my friend and a colleague Michael D. Morris for offering his deck as a
platform for the wind generator.
● 54
There have been proposed several interesting batteries based on microbial transformation
of organics, also known as microbial fuel cells. These batteries were based on sediment
(2), urine (3, 4), yeast and grape juice (6), and others. These systems rely on the electron
transfer from organic matter to oxygen (in the air) mediated by microbes. The electron
transfer process is vital for the functioning of any living organism, including the microbes.
The circuit consuming the power from the microbial fuel cell de facto becomes a part of a
microbial electron transfer process (respiration). It was shown that such prototype systems
were capable of powering microcontroller-based devices. Imagine turning your compost
into a power source in addition to feeding your plants! Successful energy harvesting circuits
from the microbial fuel cells have been created (e.g., 5). It is important to distinguish the
microbial fuel cells from another "natural" battery such as a potato with iron and copper
electrodes. The potato battery is closely related to the original Volta element, whereby two
dissimilar metals are immersed into an electrolyte. The potato is that semi-solid electro-
lyte; but in fact, no potato organic matter participates in the electrochemistry of the energy
generation.
For our first experience with renewable energy, let us discuss a microbial fuel cell battery
based on rabbit manure. The manure was obtained from our pet rabbit Mr. Flop (Figure 34).
Interestingly, it was shown that the rabbit manure could be used to produce biogas due
to its rich microbial composition and nutrients (7). The rabbit manure contains cellulose,
hemi-cellulose and lignin, which can be digested in our microbial fuel cell by the microbes
that are present in the said manure. (Other types of manure will likely work too, because
of similar composition, ref. 8.) The manure battery is inspired by the sediment battery (2)
with significant modifications. Specifically, unlike the sediment battery, which used sodium
acetate as fuel, we will be using manure and hay without any other additional fuels!
Figure 34: Mr. Flop is the "manure engineer" for the microbial fuel cell.
● 55
Figure 35 illustrates the steps for making the manure battery. First, prepare 4 gallons (~15
liters) of electrolyte by dissolving 35.4 g of sea salt "instant ocean" in tap water. Assuming
that sea salt is mostly NaCl, it makes a 40 mM NaCl solution. The NaCl concentration is
chosen based on the work describing the sediment fuel cell (2). Additional electrolytes in
the sea salt are likely beneficial to the manure microbes. The other required materials are
plastic pail, stainless steel mesh, solid copper wires, terminals, perforated plastic sheet.
Using a roll or a sheet of stainless-steel mesh, cut out appropriately sized cathode and
anode electrodes. The size must be such that the electrodes easily fit into the pail covering
most of the pail’s diameter. Attach a solid copper wire to each electrode using a bolt and a
nut. Place one electrode, the anode, on the bottom of the pail. Cover the anode with rabbit
manure and hay (biofuel). Cover the biofuel with a perforated plastic disc to prevent hay
and other substances from floating up. Pour the electrolyte; place the other electrode, the
cathode, such that it is semi-immersed into the electrolyte and exposed to air. Use foam
cubes as flotation devices to suspend the cathode. Lastly, short-circuit cathode and anode
at the external terminals to enable the "bio-current" to flow while the microbes grow. It
is important to understand that although we also have dissimilar metals, such as stain-
less steel and copper, they do not contribute to the power generation. This is because the
dissimilar metals in our case are in a direct contact, hence any voltage generated at the
junction is automatically shorted.
● 56
Figure 35: Manure battery. Step 1: stainless steel mesh anode placed on the
bottom of a plastic pail. Step 2: biofuel (manure and hay) is placed on top of the anode.
Step 3: a perforated plastic disc placed on top of the biofuel. Step 4: electrolyte
is poured, cathode is situated on the foam floaters.
You will need some patience waiting until the microbial community is actively digesting the
biofuel. Consistent with the referenced publications, it will take about a month and a half
before any significant output emerges! It is important to keep the short circuit between
the anode and cathode while the battery is maturing. This is due to the ecosystem evolving
and positively selecting the bacteria that can participate in the electrochemical process with
current flowing through the external circuit. If the current is not permitted to flow, some
other bacteria may grow which are unable to give off the electrons from the biofuel to the
stainless-steel anode.
Let’s now measure the output of our manure battery. Interesting values are the short-cir-
cuit current and the open circuit voltage. The former indicates the amount of material being
digested per second. The latter tells us about the nature of chemical species involved in the
● 57
process. Measuring the short-circuit current is very easy, because its value is stable during
the measurement process: the battery is already in the short circuit state, we just replace
the jumper wire with a micro amperemeter. Measuring open-circuit voltage is more difficult.
It turned out that the voltage keeps increasing over time during the measurement, pre-
sumably due to the accumulation of unreduced oxygen. Therefore, we arbitrarily decided to
record the voltage after 5 min of open circuit state. Figure 36 shows the current and voltage
measurements over a two-month period.
An interesting question to investigate is how much manure and hay to put into the battery?
Presumably if we put too much organics, the bacteria may not reach the anode. Hence,
there must be an optimal ratio between the amount of the organics and the anode/organics
contact area. In addition, the sodium chloride concentration may need to be optimized. Too
little salt will harm the bacteria and reduce conductivity of the electrolyte. Too much salt will
harm the bacteria by osmotically withdrawing the water.
What can we do with such a low but steady power? One option could be to connect several
of such batteries in series to produce usable voltage. Another option is to use an ultra-low
voltage energy harvesting circuit based on the LTC3108 integrated circuit (Analog Devices;
see datasheet extract reproduced in the Appendix). The device is a pulsed DC-DC converter
that operates with voltages as low as 20 mV. An additional important component required
is a 1:100 transformer.
● 58
Figure 36: Output of the manure battery. Top, short-circuit current, microamperes;
bottom, open circuit voltage after 5 min.
● 59
In this chapter we will discuss some boring but critical technical details of our DIY renewa-
ble energy power plant. The wiring must be housed in the conduits and the batteries must
be taken care of. We do not discuss lithium batteries here because currently they are very
"smart". In fact, the large capacity lithium batteries come with a built-in battery man-
agement system, which takes care of balancing cells, preventing overcharge, etc. Unlike
the lead-acid battery, there is not much that can be done if a lithium battery is degraded.
Therefore, for the experimentalist, a lead-acid battery is more interesting!
Conduits
The wiring associated with the PV of the garage power plant, or the deck wind generator
is hidden in metallic conduits. The connections are made in the junction boxes. One could
choose a non-metallic option with PVC conduits and boxes, but we prefer metal. Among the
metallic ones, there are several types of conduits: Electrical Metal Tubing (EMT), Rigid and
Flex. Figure 37 shows EMT and Flex conduits as well as the fittings relevant to our project.
The EMT is a thin-walled tube, which can be made of galvanized or stainless steel. The Rigid
uses much thicker steel; hence it is more expensive but more durable as well. Flex is a met-
al hose that may be covered with a plastic sheath, in which case it may be used outdoors.
Figure 37: Conduits and fittings. A – two tubes elbow connector, indoors use. B – raintight
tube to box compression connector, in/outdoor use. C – Flex to box connector,
indoors use. D – tee conduit body to join 3 tubes. E – set screw tube to box connector,
indoors use. F – cable to box screw clamp, indoor use. G – EMT conduit. H – plastic-
covered Flex conduit.
● 60
A good starting point for the wiring is the building inlet for the PV or wind power. Place a
junction box at this place. Establish the conduit runs in the building and use conduit bodies
for turns. The wires of appropriate gauge may be solid or stranded.
A steel "fish tape" is very handy for pulling a bundle of wires through a conduit. Insert a fish
tape into one end of the conduit at the junction box or a conduit body, push it forward until
it shows up on the other end, such as another box. Tie a "pulling line", i.e., a nylon rope
to the hook of the fish tape and pull everything back. Detach the pulling line from the fish
tape. At this point, the pulling line is going through the conduit, and it is dangling on both
ends. Tie the end of the bundle of wires to the pulling line using electric tape. Carefully pull
on the rope to draw the wires through the conduit.
Lead-acid Batteries
The battery capacity can be assessed with Pi-logger. For example, Figure 38 shows charg-
ing and discharging dynamics of the batteries when we suspected diminished battery ca-
pacity. The current is characteristic of the entire bank. The voltage was measured only on
one battery, however given the battery equalizer, all other batteries had a similar voltage.
Apparently, in the "dark time", i.e., from about 17:00 until 10:00 next day (~17 h), the
batteries were discharged at 1 A reaching a complete discharge. This gives us about 17 Ah,
instead of the expected 80 Ah.
● 61
Figure 38: An example of a charging and discharging cycle of the battery bank. The
voltage measured on one of the batteries, the current is characteristic of the entire bank.
Using the "old school" non-sealed lead-acid batteries has the advantage of diagnosing and
potentially recovering the batteries. Figure 39 shows our "battery clinic" where we measure
the electrolyte (sulfuric acid) density and replace the electrolyte if necessary.
The charging process of a 12 V lead-acid battery involves three plus one stages. The bulk
stage keeps the charging current (high) constant until battery reaches 14.4 to 14.8 V. The
current will reduce while maintaining the fixed voltage leading to the absorption stage. At
● 62
the absorption, the voltage is reduced to 14.2 to 14.4 V to top off the charge (about 30%).
The current further reduces as the battery becomes fully charged. Finally, the third float
stage begins. The voltage at this stage is reduced to 13.2 to 13.8 V passing a relatively
small current compensating for the battery self-discharge. After several hours of floating,
the electrolyte density will reach a certain value indicative of a full charge. Approximately
once a month it is useful to run the equalization charge, at which the voltage is held around
15.5 to 16 volts. Equalization helps to prevent the buildup of stratification, where acid
concentration is higher at the bottom of the battery. During that time one can hear the bat-
teries "boil" producing hydrogen. No smoking or barbequing around the lead acid batteries!
Despite charging and waiting, the density of the electrolyte in our batteries would not in-
crease. We decided to replace the electrolyte. The electrolyte and the hydrometer can be
purchased in an automotive store rather cheaply. Usually, the hydrometer indicates the
state of charge along with the value of the electrolyte density. Every cell of the battery
must be evaluated in terms of the electrolyte density. Generally, electrolyte is likely to be
replaced in all cells.
Wearing the proper attire of a "battery doctor", i.e., gloves, safety glasses, and a coat, open
the cells with a screwdriver prying off the plugs. Aspirate some quantity of the fluid, assess
the density, and return the fluid back to the cell. Keeping the cells open, carefully tilt the
battery on a solid support (e.g., cinder block) and tilt to pour out the electrolyte. Rock the
battery back and forth to remove the electrolyte completely. Bring the battery close to the
electrolyte container and fill up the cells using a hose attached to the electrolyte container.
Fill the fluid until it shows up just below the opening of the cells. Cap the cells and bring the
battery back to service. Dispose of the drained electrolyte according to the local hazmat
code.
● 63
Figure 39: Diagnostics and revival of lead-acid batteries. Top left, the battery clinic set
up: container to receive old electrolyte; cinder block; new electrolyte; a screwdriver;
hydrometer. Top right: measuring the electrolyte density. Bottom left: pouring out old
electrolyte. Bottom right: filling up cells with the new electrolyte.
● 64
Literature
3. Walter XA, Gajda I, Forbes S., et al. Scaling-up of a novel, simplified MFC stack
basedWalter_16 on a self-stratifying urine column. BIOTECHNOLOGY FOR BIO-
FUELS, 2016, Vol 9, art. no 93.
12. Henry CH. Limiting efficiencies of idealHenry_80 single and multiple energy
gap terrestrial solar cells. Journal of Applied Physics, 1980, 51, 4494.
● 65
17. Riedijk FR, Huijsing JH. A smart balanced thermal pyranometer using a sig-
ma-delta A-to-D converter for direct communicationRiedijk_93 with microcon-
trollers. Sensors and Actuators A: Physical, 1993, 37–38, 16-25.
21. Pozhitkov A. Taming YourPozitkov_19 Wind Turbine. Circuit Cellar. 2019, #347
● 66
Appendix
Program: Voltread
#include "daqhats_utils.h"
#include <daqhats/daqhats.h>
#include <unistd.h>
void prompt()
{
fprintf(stderr, "\nX (q to stop): ");
}
int opt;
while ((opt = getopt(argc, argv, "c:r:n:")) != -1) {
switch (opt) {
case ‘c’:
channelmask = atoi(optarg);
break;
case ‘n’:
numread = atoi(optarg);
break;
case ‘r’:
scanrate = atof(optarg);
break;
default: /* ‘?’ */
fprintf(stderr, "Usage: %s [-c channel_mask] [-r sampling_rate] [-n
samples_to_read]\n Defaults: channel_mask=1, samplig_rate=MAX, samples_to_
read=10;", argv[0]);
● 67
return -1;
}
}
convert_options_to_string(options, display_string);
fprintf(stderr, " Options: %s\n", display_string);
if(scanrate == 0.0)
scanrate = 100000.0/num_channels;
double actual_scan_rate = 0.0;
mcc118_a_in_scan_actual_rate(num_channels, scanrate, &actual_scan_rate);
fprintf(stderr, "\nRequested %f and actual %f scan rates.\nPress enter to
stop...\n", scanrate, actual_scan_rate);
● 68
uint32_t matrixdims[2];
prompt();
double x;
while(scanf(«%lf», &x)!=0)
{
uint16_t status;
uint32_t samples_per_channel=0;
matrixdims[1] = num_channels;
do{
result = mcc118_a_in_scan_read(address, &status, numread, -1, read_buf,
buffer_size, &samples_per_channel);
STOP_ON_ERROR(result);
matrixdims[0] = samples_per_channel;
//average per column
printf("\n%lf\t",x);
for(int col = 0; col < num_channels && samples_per_channel > 0; col++)
{
double colaverage = 0;
for(int row = 0; row < samples_per_channel; row++)
colaverage += *MatrixElement(row, col, read_buf, matrixdims);
colaverage /= samples_per_channel;
printf("%lf\t", colaverage);
}
} while(status & STATUS_RUNNING);
result = mcc118_a_in_scan_cleanup(address);
● 69
STOP_ON_ERROR(result);
prompt();
}
stop:
result = mcc118_close(address);
print_error(result);
free(read_buf);
return 0;
}
Program: Voltlog
#include "daqhats_utils.h"
#include <daqhats/daqhats.h>
#include <unistd.h>
#include <signal.h>
int gContinue=1;
int opt;
while ((opt = getopt(argc, argv, "c:r:n:a")) != -1) {
switch (opt) {
● 70
case ‘c’:
channelmask = atoi(optarg);
break;
case ‘n’:
numread = atoi(optarg);
break;
case ‘r’:
scanrate = atof(optarg);
break;
case ‘a’:
averaging = 1;
break;
default: /* ‘?’ */
fprintf(stderr, "Usage: %s [-c channel_mask] [-r sampling_rate] [-n
samples_per_batch] [-a]\n Defaults: channel_mask=1, samplig_rate=90000/number_of_
channels per sec., samples_per_batch=10, averaging per batch off", argv[0]);
return -1;
}
}
● 71
return -1;
}
convert_options_to_string(options, display_string);
fprintf(stderr, " Options: %s\n", display_string);
if(scanrate == 0.0)
scanrate = MAX_SCANRATE/num_channels;
double actual_scan_rate = 0.0;
mcc118_a_in_scan_actual_rate(num_channels, scanrate, &actual_scan_rate);
fprintf(stderr, "\nRequested %f and actual %f scan rates.\n", scanrate,
actual_scan_rate);
uint32_t matrixdims[2];
matrixdims[1] = num_channels;
uint16_t status;
uint32_t samples_per_channel=0;
while(gContinue)
{
result = mcc118_a_in_scan_start(address, channelmask, numread, scanrate,
options); // fixed nuber of samples
STOP_ON_ERROR(result);
matrixdims[0] = samples_per_channel;
if(averaging)
{
//average per column
printf("\n");
for(int col = 0; col < num_channels && samples_per_channel > 0; col++)
{
double colaverage = 0;
for(int row = 0; row < samples_per_channel; row++)
colaverage += *MatrixElement(row, col, read_buf, matrixdims);
colaverage /= samples_per_channel;
printf("%lf\t", colaverage);
}
} else {
for(int row = 0; row < samples_per_channel; row++)
{
● 72
printf("\n");
for(int col = 0; col < num_channels && samples_per_channel > 0;
col++)
printf("%lf\t",*MatrixElement(row, col, read_buf, matrixdims));
}
}
do{
result = mcc118_a_in_scan_read(address, &status, numread, -1, read_buf,
buffer_size, &samples_per_channel);
STOP_ON_ERROR(result);
result = mcc118_a_in_scan_cleanup(address);
STOP_ON_ERROR(result);
}
stop:
result = mcc118_close(address);
print_error(result);
free(read_buf);
fprintf(stderr, "\nLogging ended\n");
return 0;
}
● 73
Hybrid inverter
MN5048DIY
User Manual
The figure below shows the system application scenario of this product.
A complete system consists of the following parts:
1. PV module: Converts light energy into DC power, and charges the battery
through the onboard MPPT solar charge controller, or to directly drive the inverter
making AC power to drive the loads. (battery-less mode)
3. Battery: Provided to ensure normal power supply to the system loads when solar
energy is insufficient and the AC Main is not connected.
5. MNS Hybrid Inverter: The energy conversion unit of the whole system.
Specific system wiring method depends on the actual application scenario.
● 74
MN5048DIY
User Manual
Parameter Parameter
Settings Description
no. name
Exit setting
00 [00] ESC Exit the setup menu
menu
● 75
MN5048DIY
User Manual
Parameter Parameter
Settings Description
no. name
PV priority charging ; when not PV is unavailable,
[06] CSO
charging from the mains is started.
● 76
MN5048DIY
User Manual
Parameter Parameter
Settings Description
no. name
Ternary lithium battery; which is adjustable.
[08] N13/N14
● 77
MN5048DIY
User Manual
Parameter Parameter
Settings Description
no. name
Battery discharge limit voltage; when the
Battery battery voltage is lower than the point, the
15 discharge [15] 40V default output is turned off immediately ; t he setting
limit voltage range is 40V~ 52V, with a step of 0.4V. It is valid
for user - defined battery and lithium battery.
[16] DIS Equalizing charge is disabled
Battery
Equalizing charge is enabled, only valid for
16 equalization
[16] ENA default vented lead-acid battery and sealed lead- acid
enable
battery
Battery Equalizing charge voltage; setting range :
[17] 58.4V
17 equalization 48V~ 58.4V, with a step of 0.4V; valid for vented
default
voltage lead - acid battery and sealed lead - acid battery
Equalizing charge time ; setting range :
Battery
5min~900min, with a step of 5 minutes; valid
18 equalized [18] 120 default
for vented lead-acid battery and sealed lead -
time
acid battery
Equalizing charge delay ; setting range :
Battery
5min~900min, with a step of 5 minutes ; valid
19 equalized [19] 120 default
for vented lead - acid battery and sealed lead -
time out
acid battery
Battery Equalizing charge derating time, 0~30days,
20 equalization [20] 30 default with a step of 1 day ; valid for vented lead -acid
interval battery and sealed lead -acid battery
Battery [21] DIS Stop equalizing charge immediately.
21 equalization
[21] ENA default Start equalizing charge immediately.
immediately
[22] DIS default Power saving mode disabled .
When the power saving mode is enabled, if there
Power saving is no load or less than 50W , the inverter
22
mode [22] ENA output is turned of f after a delay for a certain
period of time. When the load is more than
50W, the inverter automatically restarts.
Automatic restart when overload is disabled. If
Restart when
23 [23] DIS an overload occurs and the output is turned off,
over load
the unit will not restart .
● 78
MN5048DIY
User Manual
Parameter Parameter
Settings Description
no. name
Auto matic restart when over temperature is
disabled. If an over - temperature shutdown
[24] DIS
occurs, machine will not restart to turn the output
Restart when
on .
24 over
Automatic restart when over temperature is
temperature
enabled. If an over-temperature shutdown occurs,
[24] ENA default
the mac hine will restart when the temperature
drops.
[25] DIS Alarm is disabled
25 Alarm enable
[25] ENA default Alarm is enabled
Alarm beep is disabled when the status of the
Beeps while [26] DIS
main input source changes
26 primary source
Alarm beep is enabled when the status of the
is interrupted [26] ENA default
main input source changes
It is disabled to automatically switch to the Mains
[27] DIS
Bypass output when the inverter is overloaded.
27
when over load It is enabled to aut omatically switch to the Mains
[27] ENA default
when the inverter is overloaded.
Max AC charger
28 [28] 40A default Max AC charger current. Setting range: 0~40A;
current
Supply for industrial frequency transformer
[29] DIS default
(disabled )
29 Split Phase
Supply for industrial frequency transformer
[29] ENA
(enabled )
Max PV charger
36 [36] 40A default Max PV charger current. Setting range: 0~40A
current
Battery fully After the battery is fully charged, it needs to be
37 charged [37] 52V default lower than this set voltage before it can be
recovery point recharged
● 79
MN5048DIY
User Manual
4.3 Battery type parameters
36~60V
Absorption charge voltage 57.6V 56.8V 58.4V
(Adjustable)
36~60V
Float charge voltage 55.2V 55.2V 55.2V
(Adjustable)
36~60V
Undervoltage alarm voltage 44V 44V 44V
(Adjustable)
36~60V
Discharge limit voltage 40V 40V 40V
(Adjustable)
1~30s
Over - discharge delay time 5s 5s 5s
(Adjustable)
0~600
120
Equalizing charge duration 120 minutes - minutes
minutes
(Adjustable)
0~250
Equalizing charge interval 30 days - 30 days days
(Adjustable)
10~600
120
Absorption charge duration 120 minutes 120 minutes minutes
minutes
(Adjustable)
● 80
MN5048DIY
User Manual
Low voltage
disconnection
38.8V 42V 48.8V 45.6V 42V 36~60V
(Adjus table) (Adjustable) (Adjustable) (Adjustable) (Adjustable) (Adjustable)
voltage
Over -
discharge
30s 30s 30s 30s 30s 1~30s
(Adjustable) (Adjustable) (Adjustable) (Adjustable) (Adjustable) (Adjustable)
delay time
Absorption charge
120 120 120 120 120 10~600
duration minutes minutes minu tes minutes minutes minutes
(Adjustable) (Adjustable) (Adjustable) (Adjustable) (Adjustable) (Adjustable)
● 81
MN5048DIY
User Manual
8.Technical Parameter
Model MN5048DIY
AC mode
Rated input
voltage 110/120Vac
Input voltage
range ( 90Vac-140Vac)
47±0.3Hz~55±0.3Hz(50Hz);
Frequency range
57±0.3Hz~65±0.3Hz(60Hz);
Overload/short-
circuit protection Breaker
Power factor 1
● 82
MN5048DIY
User Manual
Specification of
bypass breaker 63A
Power saving
mode Load ≤ 50W
AC charge
Frequency
Battery typerange Lead acid or lithium battery
Maximum charge 40A
current
Charge current ±5Adc
error
Charge voltage
range 40–60Vdc
Breaker
specification 63A
Overcharge Turn off charge after 1min alarm
protection
Solar charge
Maximum PV open
-circuit voltage 500Vdc
PV operation
voltage range 120-500Vdc
MPPT voltage
range 120-450Vdc
Battery voltage
range 40-60Vdc
Maximum
output power 5000W
Charge current
range of solar 0-80A
energy(settable)
Charge short-
circuit protection Blown fuse
● 83
MN5048DIY
User Manual
Authentication specification
Specification
CE(IEC62109)、RoHs
authentication
EMC authentication
grade En61000
Operation
temperaturerange -15°Cto55°C
Storage
temperaturerange -25°C~60°C
Noise ≤60dB
Dimension(L*W*D) 426mm*322mm*126mm
Weight(kg) 11.6
● 84
DS18B20
PRELIMINARY
DS18B20
Programmable Resolution
1-Wire® Digital Thermometer
www.dalsemi.com
VDD
750 ms (max.) DQ 4 5 GND
User-definable, nonvolatile temperature alarm
settings DS18B20Z
Alarm search command identifies and 8-Pin SOIC (150 mil)
addresses devices whose temperature is
outside of programmed limits (temperature PIN DESCRIPTION
alarm condition) GND - Ground
Applications include thermostatic controls, DQ - Data In/Out
industrial systems, consumer products, VDD - Power Supply Voltage
thermometers, or any thermally sensitive NC - No Connect
system
DESCRIPTION
The DS18B20 Digital Thermometer provides 9 to 12-bit (configurable) temperature readings which
indicate the temperature of the device.
Information is sent to/from the DS18B20 over a 1-Wire interface, so that only one wire (and ground)
needs to be connected from a central microprocessor to a DS18B20. Power for reading, writing, and
performing temperature conversions can be derived from the data line itself with no need for an external
power source.
Because each DS18B20 contains a unique silicon serial number, multiple DS18B20s can exist on the
same 1-Wire bus. This allows for placing temperature sensors in many different places. Applications
where this feature is useful include HVAC environmental controls, sensing temperatures inside buildings,
equipment or machinery, and process monitoring and control.
1 of 27 050400
● 85
SUVT Plexiglas
● 86
LTC3108
LTC3108
Ultralow Voltage Step-Up
Converter and Power Manager
FEATURES DESCRIPTION
n Operates from Inputs of 20mV The LTC®3108 is a highly integrated DC/DC converter
n Complete Energy Harvesting Power ideal for harvesting and managing surplus energy from
Management System extremely low input voltage sources such as TEGs (ther-
n Selectable V
OUT of 2.35V, 3.3V, 4.1V or 5V moelectric generators), thermopiles and small solar cells.
n LDO: 2.2V at 3mA The step-up topology operates from input voltages as
n Logic Controlled Output low as 20mV. The LTC3108 is functionally equivalent to
n Reserve Energy Output the LTC3108-1 except for its unique fixed VOUT options.
n Power Good Indicator
Using a small step-up transformer, the LTC3108 provides
n Uses Compact Step-Up Transformers
a complete power management solution for wireless
n Small 12-Lead (3mm × 4mm) DFN or 16-Lead
sensing and data acquisition. The 2.2V LDO powers an
SSOP Packages
external microprocessor, while the main output is pro-
APPLICATIONS grammed to one of four fixed voltages to power a wire-
less transmitter or sensors. The power good indicator
n Remote Sensors and Radio Power signals that the main output voltage is within regulation.
n Surplus Heat Energy Harvesting A second output can be enabled by the host. A storage
n HVAC Systems capacitor provides power when the input voltage source
n Industrial Wireless Sensing is unavailable. Extremely low quiescent current and high
n Automatic Metering efficiency design ensure the fastest possible charge times
n Building Automation of the output reservoir capacitor.
n Predictive Maintenance
The LTC3108 is available in a small, thermally enhanced
All registered trademarks and trademarks are the property of their respective owners.
12-lead (3mm × 4mm) DFN package and a 16-lead
SSOP package.
TYPICAL APPLICATION
Wireless Remote Sensor Application Powered From a Peltier Cell
VOUT Charge Time
1nF
1:100 5V
C1 VSTORE 1000
+
+ + 0.1F
VOUT = 3.3V
THERMOELECTRIC LTC3108 COUT = 470µF
220µF 330pF 6.3V
GENERATOR
C2 VOUT2 100
PGOOD
PGD µP
20mV TO 500mV 2.2V
SW VLDO
TIME (sec)
2.2µF SENSORS 10
3.3V
VS2 VOUT
+ RF LINK
470µF 1
1:100 Ratio
VS1 VOUT2_EN
1:50 Ratio
VAUX GND 3108 TA01a
1:20 Ratio
0
0 50 100 150 200 250 300 350 400
1µF
VIN (mV)
3108 TA01b
Rev. D
● 87
LTC3108
ABSOLUTE MAXIMUM RATINGS (Note 1)
PIN CONFIGURATION
TOP VIEW
TOP VIEW
GND 1 16 GND
VAUX 1 12 SW
VAUX 2 15 SW
VSTORE 2 11 C2
VSTORE 3 14 C2
VOUT 3 13 10 C1
VOUT 4 13 C1
GND
VOUT2 4 9 VOUT2_EN VOUT2 5 12 VOUT2_EN
VLDO 5 8 VS1 VLDO 6 11 VS1
PGD 6 7 VS2 PGD 7 10 VS2
GND 8 9 GND
DE PACKAGE
12-LEAD (4mm × 3mm) PLASTIC DFN
GN PACKAGE
TJMAX = 125°C, θJA = 43°C/W 16-LEAD PLASTIC SSOP NARROW
EXPOSED PAD (PIN 13) IS GND, MUST BE SOLDERED TO PCB (NOTE 4)
TJMAX = 125°C, θJA = 110°C/W
ORDER INFORMATION
LEAD FREE FINISH TAPE AND REEL PART MARKING* PACKAGE DESCRIPTION TEMPERATURE RANGE
LTC3108EDE#PBF LTC3108EDE#TRPBF 3108 12-Lead (4mm × 3mm) Plastic DFN –40°C to 125°C
LTC3108IDE#PBF LTC3108IDE#TRPBF 3108 12-Lead (4mm × 3mm) Plastic DFN –40°C to 125°C
LTC3108EGN#PBF LTC3108EGN#TRPBF 3108 16-Lead Plastic SSOP –40°C to 125°C
LTC3108IGN#PBF LTC3108IGN#TRPBF 3108 16-Lead Plastic SSOP –40°C to 125°C
Contact the factory for parts specified with wider operating temperature ranges. *The temperature grade is identified by a label on the shipping container.
Tape and reel specifications. Some packages are available in 500 unit reels through designated sales channels with #TRMPBF suffix.
ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the full operating
junction temperature range, otherwise specifications are for TA = 25°C (Note 2). VAUX = 5V, unless otherwise noted.
PARAMETER CONDITIONS MIN TYP MAX UNITS
Minimum Start-Up Voltage Using 1:100 Transformer Turns Ratio, VAUX = 0V 20 50 mV
No-Load Input Current Using 1:100 Transformer Turns Ratio; VIN = 20mV, VOUT2_ 3 mA
EN = 0V; All Outputs Charged and in Regulation
Input Voltage Range Using 1:100 Transformer Turns Ratio l VSTARTUP 500 mV
Rev. D
● 88
LTC3108
ELECTRICAL CHARACTERISTICS The l denotes the specifications which apply over the full operating
junction temperature range, otherwise specifications are for TA = 25°C (Note 2). VAUX = 5V, unless otherwise noted.
PARAMETER CONDITIONS MIN TYP MAX UNITS
Output Voltage VS1 = VS2 = GND l 2.30 2.350 2.40 V
VS1 = VAUX, VS2 = GND l 3.234 3.300 3.366 V
VS1 = GND, VS2 = VAUX l 4.018 4.100 4.182 V
VS1 = VS2 = VAUX l 4.90 5.000 5.10 V
VOUT Quiescent Current VOUT = 3.3V, VOUT2_EN = 0V 0.2 µA
VAUX Quiescent Current No Load, All Outputs Charged 6 9 µA
LDO Output Voltage 0.5mA Load l 2.134 2.2 2.266 V
LDO Load Regulation For 0mA to 2mA Load 0.5 1 %
LDO Line Regulation For VAUX from 2.5V to 5V 0.05 0.2 %
LDO Dropout Voltage ILDO = 2mA 100 200 mV
LDO Current Limit VLDO = 0V l 4 11 mA
VOUT Current Limit VOUT = 0V l 2.8 4.5 7.5 mA
VSTORE Current Limit VSTORE = 0V l 2.8 4.5 7.5 mA
VAUX Clamp Voltage Current into VAUX = 5mA l 5 5.25 5.60 V
VSTORE Leakage Current VSTORE = 5V 0.1 0.3 µA
VOUT2 Leakage Current VOUT2 = 0V, VOUT2_EN = 0V 0.1 µA
VS1, VS2 Threshold Voltage l 0.4 0.85 1.2 V
VS1, VS2 Input Current VS1 = VS2 = 5V 0.01 0.1 µA
PGOOD Threshold (Rising) Measured Relative to the VOUT Voltage –7.5 %
PGOOD Threshold (Falling) Measured Relative to the VOUT Voltage –9 %
PGOOD VOL Sink Current = 100µA 0.15 0.3 V
PGOOD VOH Source Current = 0 2.1 2.2 2.3 V
PGOOD Pull-Up Resistance 1 MΩ
VOUT2_EN Threshold Voltage VOUT2_EN Rising l 0.4 1 1.3 V
VOUT2_EN Pull-Down Resistance 5 MΩ
VOUT2 Turn-On Time 5 µs
VOUT2 Turn-Off Time (Note 3) 0.15 µs
VOUT2 Current Limit VOUT = 3.3V l 0.15 0.3 0.45 A
VOUT2 Current Limit Response Time (Note 3) 350 ns
VOUT2 P-Channel MOSFET On-Resistance VOUT = 3.3V (Note 3) 1.3 Ω
N-Channel MOSFET On-Resistance C2 = 5V (Note 3) 0.5 Ω
Note 1: Stresses beyond those listed under Absolute Maximum Ratings temperature (TJ) is calculated from the ambient temperature (TA) and
may cause permanent damage to the device. Exposure to any Absolute power dissipation (PD) according to the formula: TJ = TA + (PD • θJA°C/W),
Maximum Rating condition for extended periods may affect device where θJA is the package thermal impedance.
reliability and lifetime. Note 3: Specification is guaranteed by design and not 100% tested in
Note 2: The LTC3108 is tested under pulsed load conditions such that TJ ≈ production.
TA. The LTC3108E is guaranteed to meet specifications from 0°C to 85°C Note 4: Failure to solder the exposed backside of the package to the PC
junction temperature. Specifications over the –40°C to 125°C operating board ground plane will result in a thermal resistance much higher than
junction temperature range are assured by design, characterization and 43°C/W.
correlation with statistical process controls. The LTC3108I is guaranteed Note 5: The absolute maximum rating is a DC rating. Under certain
over the full –40°C to 125°C operating junction temperature range. conditions in the applications shown, the peak AC voltage on the C2 pin
Note that the maximum ambient temperature is determined by specific may exceed ±8V. This behavior is normal and acceptable because the
operating conditions in conjunction with board layout, the rated thermal current into the pin is limited by the impedance of the coupling capacitor.
package thermal resistance and other environmental factors. The junction
Rev. D
● 89
LTC3108
TYPICAL PERFORMANCE CHARACTERISTICS TA = 25°C, unless otherwise noted.
EFFICIENCY (%)
2500 50
IVOUT (µA)
IIN (mA)
500 10
1 0 0
10 100 1000 0 100 200 300 400 500
VIN (mV) VIN (mV)
3108 G00
3108 G01
EFFICIENCY (%)
EFFICIENCY (%)
800 40
1600 IVOUT 40
(VOUT = 4.5V)
600 30
IVOUT 1200 30
(VOUT = 4.5V)
400 20
800 20
200 10 400 10
0 0 0 0
0 100 200 300 400 500 0 100 200 300 400 500
VIN (mV) VIN (mV)
3108 G03 3108 G02
7
6
IVOUT (µA)
5 100
1:50 RATIO
4
3
10 1Ω
2 1:100 RATIO
2Ω
1 5Ω
10Ω
0 0
0 100 200 300 400 500 0 100 200 300 400 500 600 700 800
VIN (mV) VIN OPEN-CIRCUIT (mV)
3108 G05
3108 G04
Rev. D
● 90
LTC3108
TYPICAL PERFORMANCE CHARACTERISTICS TA = 25°C, unless otherwise noted.
IVOUT vs VIN and Source Resistance, IVOUT vs VIN and Source Resistance,
1:50 Ratio 1:100 Ratio
10000 1000
C1 = 4.7nF C1 = 1nF
1000
IVOUT (µA)
IVOUT (µA)
100 100
10 1Ω 1Ω
2Ω 2Ω
5Ω 5Ω
10Ω 10Ω
0 10
0 100 200 300 400 500 600 700 800 0 100 200 300 400 500
VIN OPEN-CIRCUIT (mV) VIN OPEN-CIRCUIT (mV)
3108 G06 3108 G07
100 2V/DIV
SW PIN
50mV/
10 DIV
1:50 RATIO
1:100 RATIO 3108 G09
0.14
DROP IN VLDO (%)
0.12
–0.50 0.10
0.08
0.06
–0.75
0.04
0.02
–1.00 0.00
0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4
LDO LOAD (mA) LDO LOAD (mA)
3108 G10 3108 G11
Rev. D
● 91
LTC3108
TYPICAL PERFORMANCE CHARACTERISTICS TA = 25°C, unless otherwise noted.
VLDO
20mV/
20mV/DIV
DIV
ILDO
5mA/DIV
3108 G14
3108 G15
100ms/DIV 200µs/DIV
0mA TO 3mA LOAD STEP
CLDO = 2.2µF
Rev. D
● 92
LTC3108
PIN FUNCTIONS (DFN/SSOP)
VAUX (Pin 1/Pin 2): Output of the Internal Rectifier Circuit VS1 (Pin 8/Pin 11): VOUT Select Pin 1. Connect this pin
and VCC for the IC. Bypass VAUX with at least 1µF of to ground or VAUX to program the output voltage (see
capacitance. An active shunt regulator clamps VAUX to Table 1).
5.25V (typical).
VOUT2_EN (Pin 9/Pin 12): Enable Input for VOUT2. VOUT2
VSTORE (Pin 2/Pin 3): Output for the Storage Capacitor will be enabled when this pin is driven high. There is an
or Battery. A large capacitor may be connected from this internal 5M pull-down resistor on this pin. If not used,
pin to GND for powering the system in the event the input this pin can be left open or grounded.
voltage is lost. It will be charged up to the maximum VAUX
C1 (Pin 10/Pin 13): Input to the Charge Pump and
clamp voltage. If not used, this pin should be left open Rectifier Circuit. Connect a capacitor from this pin to the
or tied to VAUX. secondary winding of the step-up transformer.
VOUT (Pin 3/Pin 4): Main Output of the Converter. The
C2 (Pin 11/Pin 14): Input to the N-Channel Gate Drive
voltage at this pin is regulated to the voltage selected by Circuit. Connect a capacitor from this pin to the secondary
VS1 and VS2 (see Table 1). Connect this pin to an energy winding of the step-up transformer.
storage capacitor or to a rechargeable battery.
SW (Pin 12/Pin 15): Drain of the Internal N-Channel
VOUT2 (Pin 4/Pin 5): Switched Output of the Converter.
Switch. Connect this pin to the primary winding of the
Connect this pin to a switched load. This output is open transformer.
until VOUT2_EN is driven high, then it is connected to
VOUT through a 1.3Ω P-channel switch. If not used, this GND (Pins 1, 8, 9, 16) SSOP Only: Ground
pin should be left open or tied to VOUT. The peak current GND (Exposed Pad Pin 13) DFN Only: Ground. The DFN
in this output is limited to 0.3A typical. exposed pad must be soldered to the PCB ground plane.
VLDO (Pin 5/Pin 6): Output of the 2.2V LDO. Connect a It serves as the ground connection, and as a means of
2.2µF or larger ceramic capacitor from this pin to GND. conducting heat away from the die.
If not used, this pin should be tied to VAUX. Table 1. Regulated Voltage Using Pins VS1 and VS2
PGD (Pin 6/Pin 7): Power Good Output. When VOUT is VS2 VS1 VOUT
within 7.5% of its programmed value, PGD will be pulled GND GND 2.35V
up to VLDO through a 1MΩ resistor. If VOUT drops 9% GND VAUX 3.3V
below its programmed value PGD will go low. This pin VAUX GND 4.1V
can sink up to 100µA. VAUX VAUX 5V
VS2 (Pin 7/Pin 10): VOUT Select Pin 2. Connect this pin
to ground or VAUX to program the output voltage (see
Table 1).
Rev. D
● 93
LTC3108
BLOCK DIAGRAM
LTC3108
VOUT2
1.3Ω
ILIM VOUT2
VOUT2_EN
OFF ON
1.2V
SYNC RECTIFY REFERENCE VREF 5M
C1 VOUT
VIN 1:100 C1 VOUT
C2 +
C2 VS1
–
SW SW VOUT CHARGE
VOUT
CONTROL VS2
VSTORE PROGRAM
0.5Ω VREF
VLDO
1M
PGD
–
PGOOD
+
VAUX VSTORE
VBEST LDO
VOUT VREF
1µF CSTORE
GND (SSOP) EXPOSED PAD (DFN) VLDO
3108 BD
2.2V
2.2µF
The LTC3108 is designed to use a small external step-up average power draw is very low, but there may be peri-
transformer to create an ultralow input voltage step-up odic pulses of higher load current required. This is typi-
DC/DC converter and power manager. It is ideally suited cal of wireless sensor applications, where the quiescent
for low power wireless sensors and other applications power draw is extremely low most of the time, except for
in which surplus energy harvesting is used to generate transmit bursts when circuitry is powered up to make
system power because traditional battery power is incon- measurements and transmit data.
venient or impractical. The LTC3108 can also be used to trickle charge a standard
The LTC3108 is designed to manage the charging and capacitor, supercapacitor or rechargeable battery, using
regulation of multiple outputs in a system in which the energy harvested from a Peltier or photovoltaic cell.
Rev. D
● 94
LTC3108
OPERATION
Oscillator Synchronous Rectifiers
The LTC3108 utilizes a MOSFET switch to form a resonant Once VAUX exceeds 2V, synchronous rectifiers in parallel
step-up oscillator using an external step-up transformer with each of the internal diodes take over the job of rectify-
and a small coupling capacitor. This allows it to boost ing the input voltage, improving efficiency.
input voltages as low as 20mV high enough to provide
multiple regulated output voltages for powering other cir- Low Dropout Linear Regulator (LDO)
cuits. The frequency of oscillation is determined by the The LTC3108 includes a low current LDO to provide a
inductance of the transformer secondary winding and is regulated 2.2V output for powering low power processors
typically in the range of 10kHz to 100kHz. For input volt- or other low power ICs. The LDO is powered by the higher
ages as low as 20mV, a primary-secondary turns ratio of of VAUX or VOUT. This enables it to become active as soon
about 1:100 is recommended. For higher input voltages, as VAUX has charged to 2.3V, while the VOUT storage
this ratio can be lower. See the Applications Information capacitor is still charging. In the event of a step load on
section for more information on selecting the transformer. the LDO output, current can come from the main VOUT
capacitor if VAUX drops below VOUT. The LDO requires
Charge Pump and Rectifier a 2.2µF ceramic capacitor for stability. Larger capacitor
The AC voltage produced on the secondary winding of values can be used without limitation, but will increase
the transformer is boosted and rectified using an external the time it takes for all the outputs to charge up. The LDO
charge pump capacitor (from the secondary winding to output is current limited to 4mA minimum.
pin C1) and the rectifiers internal to the LTC3108. The
VOUT
rectifier circuit feeds current into the VAUX pin, provid-
ing charge to the external VAUX capacitor and the other The main output voltage on VOUT is charged from the
outputs. VAUX supply, and is user programmed to one of four
regulated voltages using the voltage select pins VS1 and
VAUX VS2, according to Table 2. Although the logic threshold
The active circuits within the LTC3108 are powered from voltage for VS1 and VS2 is 0.85V typical, it is recom-
mended that they be tied to ground or VAUX.
VAUX, which should be bypassed with a 1µF capacitor.
Larger capacitor values are recommended when using Table 2. Regulated Voltage Using Pins VS1 and VS2
turns ratios of 1:50 or 1:20 (refer to the Typical Application VS2 VS1 VOUT
examples). Once VAUX exceeds 2.5V, the main VOUT is GND GND 2.35V
allowed to start charging. GND VAUX 3.3V
VAUX GND 4.1V
An internal shunt regulator limits the maximum voltage
VAUX VAUX 5V
on VAUX to 5.25V typical. It shunts to GND any excess
current into VAUX when there is no load on the converter
When the output voltage drops slightly below the regu-
or the input source is generating more power than is
lated value, the charging current will be enabled as long
required by the load.
as VAUX is greater than 2.5V. Once VOUT has reached the
Voltage Reference proper value, the charging current is turned off.
The LTC3108 includes a precision, micropower reference, The internal programmable resistor divider sets VOUT,
for accurate regulated output voltages. This reference eliminating the need for very high value external resis-
becomes active as soon as VAUX exceeds 2V. tors that are susceptible to board leakage.
Rev. D
● 95
LTC3108
OPERATION
In a typical application, a storage capacitor (typically a few The VOUT2 enable input has a typical threshold of 1V
hundred microfarads) is connected to VOUT. As soon as with 100mV of hysteresis, making it logic-compatible. If
VAUX exceeds 2.5V, the VOUT capacitor will be allowed to VOUT2_EN (which has an internal pull-down resistor) is
charge up to its regulated voltage. The current available low, VOUT2 will be off. Driving VOUT2_EN high will turn on
to charge the capacitor will depend on the input voltage the VOUT2 output.
and transformer turns ratio, but is limited to about 4.5mA Note that while VOUT2_EN is high, the current limiting cir-
typical. cuitry for VOUT2 draws an extra 8µA of quiescent current
from VOUT. This added current draw has a negligible effect
PGOOD
on the application and capacitor sizing, since the load on
A power good comparator monitors the VOUT voltage. the VOUT2 output, when enabled, is likely to be orders of
The PGD pin is an open-drain output with a weak pull-up magnitude higher than 8µA.
(1MΩ) to the LDO voltage. Once VOUT has charged to
within 7.5% of its regulated voltage, the PGD output will VSTORE
go high. If VOUT drops more than 9% from its regulated The VSTORE output can be used to charge a large stor-
voltage, PGD will go low. The PGD output is designed age capacitor or rechargeable battery after VOUT has
to drive a microprocessor or other chip I/O and is not reached regulation. Once VOUT has reached regulation, the
intended to drive a higher current load such as an LED. VSTORE output will be allowed to charge up to the VAUX
Pulling PGD up externally to a voltage greater than VLDO voltage. The storage element on VSTORE can be used
will cause a small current to be sourced into VLDO. PGD to power the system in the event that the input source
can be pulled low in a wire-OR configuration with other is lost, or is unable to provide the current demanded by
circuitry. the VOUT, VOUT2 and LDO outputs. If VAUX drops below
VSTORE, the LTC3108 will automatically draw current
VOUT2
from the storage element. Note that it may take a long
VOUT2 is an output that can be turned on and off by the time to charge a large capacitor, depending on the input
host, using the VOUT2_EN pin. When enabled, VOUT2 is energy available and the loading on VOUT and VLDO.
connected to VOUT through a 1.3Ω P-channel MOSFET Since the maximum current from VSTORE is limited to a
switch. This output, controlled by a host processor, can few milliamps, it can safely be used to trickle-charge NiCd
be used to power external circuits such as sensors and or NiMH rechargeable batteries for energy storage when
amplifiers, that do not have a low power sleep or shut- the input voltage is lost. Note that the VSTORE capacitor
down capability. VOUT2 can be used to power these cir- cannot supply large pulse currents to VOUT . Any pulse
cuits only when they are needed. load on VOUT must be handled by the VOUT capacitor.
Minimizing the amount of decoupling capacitance on
VOUT2 will allow it to be switched on and off faster, allow- Short-Circuit Protection
ing shorter burst times and, therefore, smaller duty cycles All outputs of the LTC3108 are current limited to protect
in pulsed applications such as a wireless sensor/transmit- against short-circuits to ground.
ter. A small VOUT2 capacitor will also minimize the energy
that will be wasted in charging the capacitor every time Output Voltage Sequencing
VOUT2 is enabled.
A timing diagram showing the typical charging and
VOUT2 has a soft-start time of about 5µs to limit capacitor voltage sequencing of the outputs is shown in Figure 1.
charging current and minimize glitching of the main out- Note: time not to scale.
put when VOUT2 is enabled. It also has a current limiting
circuit that limits the peak current to 0.3A typical.
Rev. D
● 96
LTC3108
OPERATION
5.0
VSTORE (V)
2.5
0
3.0
2.0
PGD (V)
1.0
0
5.0
VOLTAGE (V)
0
3.0
2.0
VLDO (V)
1.0
0
0
0 10 20 30 40 50 60 70 80
TIME (ms)
3108 F01a
Figure 1. Output Voltage Sequencing with VOUT Programmed for 3.3V (Time Not to Scale)
Rev. D
● 97
LTC3108
APPLICATIONS INFORMATION
Introduction Refer to the IIN vs VIN curves in the Typical Performance
The LTC3108 is designed to gather energy from very Characteristics section to see what input current is
required from the source for a given input voltage.
low input voltage sources and convert it to usable output
voltages to power microprocessors, wireless transmitters For a given transformer turns ratio, there is a maximum
and analog sensors. Such applications typically require recommended input voltage to avoid excessively high
much more peak power, and at higher voltages, than secondary voltages and power dissipation in the shunt
the input voltage source can produce. The LTC3108 is regulator. It is recommended that the maximum input volt-
designed to accumulate and manage energy over a long age times the turns ratio be less than 50.
period of time to enable short power bursts for acquiring
Note that a low ESR bulk decoupling capacitor will usu-
and transmitting data. The bursts must occur at a low
ally be required across the input source to prevent large
enough duty cycle such that the total output energy dur-
voltage droop and ripple caused by the source’s ESR
ing the burst does not exceed the average source power
and the peak primary switching current (which can reach
integrated over the accumulation time between bursts.
hundreds of milliamps). The time constant of the filter
For many applications, this time between bursts could be
capacitor and the ESR of the voltage source should be
seconds, minutes or hours.
much longer than the period of the resonant switching
The PGD signal can be used to enable a sleeping micro- frequency.
processor or other circuitry when VOUT reaches regula-
tion, indicating that enough energy is available for a burst. Peltier Cell (Thermoelectric Generator)
1000 100
TEG: 30mm
127 COUPLES VOC
TEG MAXIMUM POUT —IDEAL (mW)
R = 2Ω
TEG VOPEN_CIRCUIT (mV)
100 10
MAX POUT
(IDEAL)
10 1
1 0.1
1 10 100
dT (°C)
3108 F02
● 98
LTC3108
APPLICATIONS INFORMATION
temperature differential between the plates. When used in source resistance will always provide more output current
this manner, a Peltier cell is referred to as a thermoelectric capability by providing a higher input voltage under load.
generator (TEG).
Peltier Cell (TEG) Suppliers
The low voltage capability of the LTC3108 design allows
it to operate from a TEG with temperature differentials as Peltier cells are available in a wide range of sizes and
low as 1°C, making it ideal for harvesting energy in appli- power capabilities, from less than 10mm square to over
cations in which a temperature difference exists between 50mm square. They are typically 2mm to 5mm in height.
two surfaces or between a surface and the ambient tem- A list of Peltier cell manufacturers is given in Table 3.
perature. The internal resistance (ESR) of most cells is Table 3. Peltier Cell Manufacturers
in the range of 1Ω to 5Ω, allowing for reasonable power CUI, Inc.
transfer. The curves in Figure 2 show the open-circuit www.cui.com (Distributor)
Fujitaka
output voltage and maximum power transfer for a typi- www.fujitaka.com/pub/peltier/english/thermoelectric_power.html
cal Peltier cell (with an ESR of 2Ω) over a 20°C range of Ferrotec
temperature differential. www.ferrotec.com/products/thermal/modules
Kryotherm
TEG Load Matching www.kryothermusa.com
Laird Technologies
The LTC3108 was designed to present a minimum input www.lairdtech.com
resistance (load) in the range of 2Ω to 10Ω, depending Marlow Industries
www.marlow.com
on input voltage and transformer turns ratio (as shown Micropelt
in the Typical Performance Characteristics curves). For www.micropelt.com
a given turns ratio, as the input voltage drops, the input Nextreme
resistance increases. This feature allows the LTC3108 to www.nextreme.com
TE Technology
optimize power transfer from sources with a few ohms of www.tetech.com/Peltier-Thermoelectric-Cooler-Modules.html
source resistance, such as a typical TEG. Note that a lower Tellurex
www.tellurex.com
Rev. D
● 99
LTC3108
APPLICATIONS INFORMATION
Thermopile Generator these applications the C2 and SW pins are not used and
can be grounded or left open.
Thermopile generators (also called powerpile generators)
are made up of a number of series-connected thermo- Examples of such input sources would be piezoelectric
couples enclosed in a metal tube. They are commonly transducers, vibration energy harvesters, low current
used in gas burner applications to generate a DC output generators, a stack of low current solar cells or a 60Hz
of hundreds of millivolts when exposed to the high tem- AC input.
perature of a flame. Typical examples are the Honeywell A series resistance of at least 100Ω/V should be used
CQ200 and Q313. These devices have an internal series to limit the maximum current into the VAUX shunt
resistance of less than 3Ω, and can generate as much regulator.
as 750mV open-circuit at their highest rated tempera-
ture. For applications in which the temperature rise is too
high for a solid-state thermoelectric device, a thermopile COMPONENT SELECTION
can be used as an energy source to power the LTC3108.
Step-Up Transformer
Because of the higher output voltages possible with a
thermopile generator, a lower transformer turns ratio can The step-up transformer turns ratio will determine how
be used (typically 1:20, depending on the application). low the input voltage can be for the converter to start.
Using a 1:100 ratio can yield start-up voltages as low as
Photovoltaic Cell 20mV. Other factors that affect performance are the DC
The LTC3108 converter can also operate from a single resistance of the transformer windings and the inductance
photovoltaic cell (also known as a PV or solar cell) at of the windings. Higher DC resistance will result in lower
light levels too low for other low input voltage boost con- efficiency. The secondary winding inductance will deter-
verters to operate. However, many variables will affect mine the resonant frequency of the oscillator, according
the performance in these applications. Light levels can to the following formula.
vary over several orders of magnitude and depend on 1
lighting conditions (the type of lighting and indoor versus Frequency = Hz
2 • π • L(sec) • C
outdoor). Different types of light (sunlight, incandescent,
fluorescent) also have different color spectra, and will Where L is the inductance of the transformer secondary
produce different output power levels depending on which winding and C is the load capacitance on the secondary
type of photovoltaic cell is being used (monocrystalline, winding. This is comprised of the input capacitance at pin
polycrystalline or thin-film). Therefore, the photovoltaic C2, typically 30pF, in parallel with the transformer sec-
cell must be chosen for the type and amount of light avail- ondary winding’s shunt capacitance. The recommended
able. Note that the short-circuit output current from the resonant frequency is in the range of 10kHz to 100kHz.
cell must be at least a few milliamps in order to power See Table 5 for some recommended transformers.
the LTC3108 converter
Table 5. Recommended Transformers
Non-Boost Applications VENDOR PART NUMBER
Coilcraft LPR6235-752SML (1:100 Ratio)
The LTC3108 can also be used as an energy harvester www.coilcraft.com LPR6235-253PML (1:20 Ratio)
and power manager for input sources that do not require LPR6235-123QML (1:50 Ratio)
boosting. In these applications the step-up transformer Würth 74488540070 (1:100 Ratio)
can be eliminated. www.we-online 74488540120 (1:50 Ratio)
74488540250 (1:20 Ratio)
Any source whose peak voltage exceeds 2.5V AC or 5V
DC can be connected to the C1 input through a current-
limiting resistor where it will be rectified/peak detected. In
Rev. D
● 100
LTC3108
APPLICATIONS INFORMATION
C1 Capacitor Using External Charge Pump Rectifiers
The charge pump capacitor that is connected from the The synchronous charge pump rectifiers in the LTC3108
transformer’s secondary winding to the C1 pin has an (connected to the C1 pin) are optimized for operation from
effect on converter input resistance and maximum output very low input voltage sources, using typical transformer
current capability. Generally, a minimum value of 1nF is step-up ratios between 1:100 and 1:50, and typical C1
recommended when operating from very low input volt- charge pump capacitor values less than 10nF.
ages using a transformer with a ratio of 1:100. Too large a
Operation from higher input voltage sources (typically
capacitor value can compromise performance when oper-
250mV or greater, under load), allows the use of lower
ating at low input voltage or with high resistance sources.
For higher input voltages and lower turns ratios, the value transformer step-up ratios (such as 1:20 and 1:10) and
of the C1 capacitor can be increased for higher output larger C1 capacitor values to provide higher output current
current capability. Refer to the Typical Applications sche- capability from the LTC3108. However, due to the result-
matic examples for the recommended value for a given ing increase in rectifier currents and resonant oscillator
turns ratio. frequency in these applications, the use of external charge
pump rectifiers is recommended for optimal performance.
Squegging In applications where the step-up ratio is 1:20 or less, and
Certain types of oscillators, including transformer-cou- the C1 capacitor is 10nF or greater, the C1 pin should be
pled oscillators such as the resonant oscillator of the grounded and two external rectifiers (such as 1N4148 or
LTC3108, can exhibit a phenomenon called squegging. 1N914 diodes) should be used. These are available as dual
This term refers to a condition that can occur which diodes in a single package. Avoid the use of Schottky rec-
blocks or stops the oscillation for a period of time much tifiers, as their lower forward voltage drop increases the
longer than the period of oscillation, resulting in bursts of minimum start-up voltage. See the Typical Applications
oscillation. An example of this is the blocking oscillator, schematics for an example.
which is designed to squegg to produce bursts of oscilla-
tion. Squegging is also encountered in RF oscillators and VOUT and VSTORE Capacitor
regenerative receivers. For pulsed load applications, the VOUT capacitor should
In the case of the LTC3108, squegging can occur when a be sized to provide the necessary current when the load
charge builds up on the C2 gate coupling capacitor, such is pulsed on. The capacitor value required will be dictated
that the DC bias point shifts and oscillation is extinguished by the load current, the duration of the load pulse, and
for a certain period of time, until the charge on the capaci- the amount of voltage droop the circuit can tolerate. The
tor bleeds off, allowing oscillation to resume. It is difficult capacitor must be rated for whatever voltage has been
to predict when and if squegging will occur in a given selected for VOUT by VS1 and VS2.
application. While squegging is not harmful, it reduces the ILOAD(mA) • tPULSE (ms)
COUT (µF) ≥
average output current capability of the LTC3108. VOUT (V)
Squegging can easily be avoided by the addition of a Note that there must be enough energy available from
bleeder resistor in parallel with the coupling capacitor on the input voltage source for VOUT to recharge the capacitor
the C2 pin. Resistor values in the range of 100k to 1MΩ during the interval between load pulses (to be discussed
are sufficient to eliminate squegging without having any in the next example). Reducing the duty cycle of the load
negative impact on performance. For the 330pF capacitor pulse will allow operation with less input energy.
used for C2 in most applications, a 499k bleeder resistor
is recommended. See the Typical Applications schematics The VSTORE capacitor may be of very large value (thou-
for an example. sands of microfarads or even Farads), to provide holdup
Rev. D
● 101
LTC3108
APPLICATIONS INFORMATION
at times when the input power may be lost. Note that this Due to the very low input voltage the circuit may operate
capacitor can charge all the way to 5.25V (regardless of from, the connections to VIN, the primary of the trans-
the settings for VOUT), so ensure that the holdup capacitor former and the SW and GND pins of the LTC3108 should
has a working voltage rating of at least 5.5V at the tem- be designed to minimize voltage drop from stray resis-
perature for which it will be used. The VSTORE capacitor tance and able to carry currents as high as 500mA. Any
can be sized using the following: small voltage drop in the primary winding conduction path
will lower efficiency and increase capacitor charge time.
CSTORE ≥
[6µA + IQ + ILDO+ (IBURST • t • f)] • TSTORE
5.25 − VOUT Also, due to the low charge currents available at the out-
puts of the LTC3108, any sources of leakage current on
Where 6µA is the quiescent current of the LTC3108, IQ is the output voltage pins must be minimized. An example
the load on VOUT in between bursts, ILDO is the load on board layout is shown in Figure 3.
the LDO between bursts, IBURST is the total load during
VIN
the burst, t is the duration of the burst, f is the frequency
of the bursts, TSTORE is the storage time required and
VOUT is the output voltage required. Note that for a pro-
grammed output voltage of 5V, the VSTORE capacitor
cannot provide any beneficial storage time.
To minimize losses and capacitor charge time, all capaci-
tors used for VOUT and VSTORE should be low leakage.
VAUX SW
See Table 6 for recommended storage capacitors. VSTORE
1 12
C2
2 11
VOUT
Storage capacitors requiring voltage balancing are not VOUT
VOUT2
3 10
C1
VOUT2_EN
recommended due to the current draw of the balancing VOUT2 VLDO
4 9
VS1
resistors. VLDO PGD
5 8
VS2
PGOOD 6 7
Rev. D
● 102
LTC3108
APPLICATIONS INFORMATION
maximum rate at which the load pulses can occur for a Therefore, in this application example, the circuit can sup-
given output current from the LTC3108 will also be shown. port a 1ms transmit burst every 1.5 seconds.
In this example, VOUT is set to 3.3V, and the maximum It can be determined that for systems that only need to
allowed voltage droop during a transmit burst is 10%, or transmit every few seconds (or minutes or hours), the
0.33V. The duration of a transmit burst is 1ms, with a total average charge current required is extremely small, as
average current requirement of 40mA during the burst. long as the sleep current is low. Even if the available
Given these factors, the minimum required capacitance charge current in the example above was only 10µA and
on VOUT is: the sleep current was only 5µA, it could still transmit a
40mA • 1ms burst every ten seconds.
COUT (µF) ≥ = 121µF
0.33V The following formula enables the user to calculate the
time it will take to charge the LDO output capacitor and
Note that this equation neglects the effect of capacitor
the VOUT capacitor the first time, from 0V. Here again,
ESR on output voltage droop. For most ceramic or low
ESR tantalum capacitors, the ESR will have a negligible the charge current available from the LTC3108 must be
effect at these load currents. known. For this calculation, it is assumed that the LDO
output capacitor is 2.2µF.
A standard value of 150µF or larger could be used for
COUT in this case. Note that the load current is the total 2.2V • 2.2µF
tLDO =
current draw on VOUT, VOUT2 and VLDO, since the cur- ICHG − ILDO
rent for all of these outputs must come from VOUT during
a burst. Current contribution from the holdup capacitor If there were 50µA of charge current available and a 5µA
on VSTORE is not considered, since it may not be able load on the LDO (when the processor is sleeping), the
to recharge between bursts. Also, it is assumed that the time for the LDO to reach regulation would be 107ms.
charge current from the LTC3108 is negligible compared If VOUT were programmed to 3.3V and the VOUT capacitor
to the magnitude of the load current during the burst. was 150µF, the time for VOUT to reach regulation would be:
To calculate the maximum rate at which load bursts 3.3V • 150µF
can occur, determine how much charge current is t VOUT = + tLDO
ICHG − IVOUT − ILDO
available from the LTC3108 VOUT pin given the input
voltage source being used. This number is best found If there were 50µA of charge current available and 5µA of
empirically, since there are many factors affecting the load on VOUT, the time for VOUT to reach regulation after
efficiency of the converter. Also determine what the total the initial application of power would be 12.5 seconds.
load current is on VOUT during the sleep state (between
bursts). Note that this must include any losses, such as Design Example 2
storage capacitor leakage. In many pulsed load applications, the duration, magnitude
Assume, for instance, that the charge current from the and frequency of the load current bursts are known and
LTC3108 is 50µA and the total current drawn on VOUT in fixed. In these cases, the average charge current required
the sleep state is 17µA, including capacitor leakage. In from the LTC3108 to support the average load must be
addition, use the value of 150µF for the VOUT capacitor. calculated, which can be easily done by the following:
The maximum transmit rate (neglecting the duration of IBURST • t
the transmit burst, which is typically very short) is then ICHG ≥ IQ +
T
given by:
150µF • 0.33V Where IQ is the sleep current on VOUT required by the
t= = 1.5sec or fMAX = 0.666Hz external circuitry in between bursts (including cap leak-
(50µA − 17µA) age), IBURST is the total load current during the burst, t is
Rev. D
● 103
LTC3108
APPLICATIONS INFORMATION
the duration of the burst and T is the period of the transmit hour. It can be determined that the sleep current of 5µA
burst rate (essentially the time between bursts). is the dominant factor because the transmit duty cycle is
In this example, IQ = 5µA, IBURST = 100mA, t = 5ms and so small (0.00014%). Note that for a VOUT of 3.3V, the
T = one hour. The average charge current required from average power required by this application is only 17µW
the LTC3108 would be: (not including converter losses).
100mA • 0.005sec Note that the charge current available from the LTC3108
ICHG ≥ 5µA + = 5.14µA has no effect on the sizing of the VOUT capacitor (if it is
3600 sec
assumed that the load current during a burst is much
Therefore, if the LTC3108 has an input voltage that allows larger than the charge current), and the VOUT capacitor
it to supply a charge current greater than 5.14µA, the has no effect on the maximum allowed burst rate.
application can support 100mA bursts lasting 5ms every
TYPICAL APPLICATIONS
VS1 VOUT2_EN
VAUX GND
OFF ON
T1: COILCRAFT LPR6235-752SML
*COUT VALUE DEPENDENT ON
THE MAGNITUDE AND DURATION 1µF
OF THE LOAD PULSE 3108 TA02
Rev. D
● 104
LTC3108
TYPICAL APPLICATIONS
Li-Ion Battery Charger and LDO Powered by a Solar Cell
T1 0.01µF
1:20
C1 VSTORE
+ +
SOLAR CELL* 220µF VOUT2
330pF
– LTC3108
C2 PGD
2.2V
SW VLDO VLDO
4.1V
VOUT VOUT
VS2 2.2µF
Li-Ion
* 2" DIAMETER MONOCRYSTALLINE CELL VS1 VOUT2_EN
LIGHT LEVEL ≥ 900 LUX
VAUX GND
T1: COILCRAFT LPR6235-253PML
4.7µF
3108 TA03
HONEYWELL T1 4.7nF
CQ200 1:50
THERMOPILE C1 VSTORE
+
220µF VOUT2
330pF
LTC3108
C2 PGD PGOOD
2.2V
SW VLDO VLDO
2.35V
VOUT VOUT
VS2
+ 150mF 2.2µF
T1: COILCRAFT LPR6235-123QML VS1 VOUT2_EN 2.5V
VAUX GND
CAP-XX GZ115F
2.2µF
3108 TA04
DC Input Energy Harvester and Power Manager AC Input Energy Harvester and Power Manager
Rev. D
● 105
LTC3108
PACKAGE DESCRIPTION
GN Package
16-Lead Plastic SSOP (Narrow .150 Inch)
(Reference LTC DWG # 05-08-1641 Rev B)
.015 ±.004
× 45° .0532 – .0688 .004 – .0098
(0.38 ±0.10)
(1.35 – 1.75) (0.102 – 0.249)
.007 – .0098
0° – 8° TYP
(0.178 – 0.249)
DE/UE Package
12-Lead Plastic DFN (4mm × 3mm)
(Reference LTC DWG # 05-08-1695 Rev D)
Rev. D
● 106
LTC3108
REVISION HISTORY
REV DATE DESCRIPTION PAGE NUMBER
A 04/10 Updated front page text and Typical Appliction 1
Updated Absolute Maximum Ratings and Order Information sections 2
Updated Electrical Characteristics 3
Added graph (3108 G00) to Typical Performance Characteristics 4
Updated Block Diagram 8
Text added to Operation section 9
Changes to Applications Information section 12-18
Updated Typical Applications 18, 19, 22
Updated Related Parts 22
B 06/13 Added vendor information to Table 5 14
C 08/13 Changed Würth transformer part numbers 14
D 03/19 Electrical Characteristics. LDO Dropout Voltage Temperature Range changed from Over Temperature to TA = 25°C, 3
VOUT, Current Limit Max & VSTORE Current Limit Max changed from 7V to 7.5V; VAUX Clamp Voltage Max changed
from 5.55V to 5.6V
Rev. D
Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog
Devices for its use, nor for any infringements of patents or other rights of third parties that may result from its use. Specifications
subject to change without notice. No license For more by
is granted information www.analog.com
implication or otherwise under any patent or patent rights of Analog Devices. 21
● 107
LTC3108
TYPICAL APPLICATION
Dual TEG Energy Harvester Operates from Temperature Differentials of Either Polarity
1nF
1:100 5V
HOT C1 VSTORE
+ +
THERMOELECTRIC CSTORE
VOUT2 VOUT2
GENERATOR 330pF
LTC3108
COLD C2 PGD PGOOD
2.2V
SW VLDO VLDO
3.3V
VS2 VOUT VOUT
LPR6235-752SML VS1 VOUT2_EN
+ 2.2µF
COUT
VAUX GND
OFF ON
1µF
VAUX
1nF
1:100
COLD C1 VSTORE
+
THERMOELECTRIC VOUT2
GENERATOR 330pF
LTC3108
HOT C2 PGD
SW VLDO
VS2 VOUT
LPR6235-752SML VS1 VOUT2_EN
VAUX GND
3108 TA07
RELATED PARTS
PART NUMBER DESCRIPTION COMMENTS
LTC1041 Bang-Bang Controller VIN: 2.8V to 16V; IQ = 1µA; SO-8 Package
LTC1389 Nanopower Precision Shunt Voltage Reference VOUT(MIN) = 1.25V; IQ = 0.8µA; SO-8 Package
LT1672/LT1673/ Single-/Dual-/Quad-Precision 2µA Rail-to-Rail Op Amps SO-8, SO-14 and MSOP-8 Packages
LT1674
LT3009 3µA IQ, 20mA Linear Regulator VIN: 1.6V to 20V; VOUT(MIN): 0.6V to Adj, 1.2V, 1.5V, 1.8V, 2.5V, 3.3V,
5V to Fixed; IQ = 3µA; ISD < 1µA; 2mm × 2mm DFN-8 and SC70 Packages
LTC3108-1 Ultralow Voltage Step-Up Converter and Power Manager VIN: 0.02V to 1V; VOUT = 2.5V, 3V, 3.7V, 4.5V Fixed; IQ = 6µA;
3mm × 4mm DFN-12 and SSOP-16 Packages
LTC3525L-3/ 400mA (ISW), Synchronous Step-Up DC/DC Converter VIN: 0.7V to 4V; VOUT(MIN) = 5VMAX; IQ = 7µA; ISD < 1µA; SC70 Package
LTC3525L-3.3/ with Output Disconnect
LTC3525L-5
LTC3588-1 Piezoelectric Energy Generator with Integrated High VIN: 2.7V to 20V; VOUT(MIN): Fixed to 1.8V, 2.5V, 3.3V, 3.6V; IQ = 0.95µA;
Efficiency Buck Converter 3mm × 3mm DFN-10 and MSOP-10E Packages
LTC3642 45V, 50mA Synchronous MicroPower Buck Converter VIN: 4.5V to 45V, 60VMAX; VOUT(MIN): 0.8V to Adj, 3.3V Fixed, 5V Fixed;
IQ = 12µA; ISD < 1µA; 3mm × 3mm DFN-8 and MSOP-8E Packages
LTC6656 850mA Precision Reference Series Low Dropout Precision
LT8410/LT8410-1 MicroPower 25mA/8mA Low Noise Boost Converter VIN: 2.6V to 16V; VOUT(MIN) = 40VMAX; IQ = 8.5µA; ISD < 1µA;
with Integrated Schottky Diode and Output Disconnect 2mm × 2mm DFN-8 Package
LTC4070 Micropower Shunt Li-Ion Charge Controls Charging with µA Source
Rev. D
22
03/19
www.analog.com
For more information www.analog.com ANALOG DEVICES, INC. 2010-2019
● 108
Index
A T
ADC hat 12 tip speed ratio 46
B W
Bandgap 25 Wolfram Mathematica 18
Beer-Lambert’s law 30
Betz Limit 46
C
CAD software 8
current transducer 12
D
diversion load 51
DS18B20 22
E
Electrical Metal Tubing 60
electrolyte 63
energy harvesting circuit 58
F
fall protection 34
fuel cells 55
G
Global irradiance 14
M
maximum power 27
P
photoelectric effect 25
Photovoltaic effect 25
Pi-logger 11
power coefficient 46
R
Raspberry Pi Zero 2W 8
S
Stefan-Boltzmann law 16
● 109
Renewable Energy
at Home Renewable Energy
A Hands-on Guide to
Crafting Your Own Power Plant at Home
A Hands-on Guide to