Welcome to A Precision Low-Level DAS/ECG Cardio tachometer Demo

board presentation. The presentation will focus on an interesting application of

analog circuits where they are utilized to amplify and condition the very low
level electrical signals associated with the human cardiac system. Often these
applications involve detecting very small electrical signals and amplifying them
in the presence of very large, potentially interfering signals.
A cardiotachometer demonstration board has been developed for this purpose
and our session today will underscore its capabilities and the difficulties that it
overcomes in the harsh monitoring environment. The cardiotachometer is an
instrument for measuring the rapidity of the heartbeat and can provide the
details of the heart rhythm as it progresses from one beat to the next.
In case you are not familiar with the acronyms DAS/ECG it is appropriate to
explain them. DAS represents Data Acquisition System, which is an
electronics system used to collect information, and condition the information
such that it can be analyzed. For example, collecting and analyzing the
heartbeat or other biophysical characteristics over a period of time.
Electrocardiography, is a non-invasive procedure for recording the electrical
changes in the heart. The record, which is called an electrocardiogram (ECG
or EKG), shows the series of waves that relate to the electrical impulses which
occur during each beat of the heart1.
1 www.healthatoz.com

This is an outline of the subjects that will be touched upon during this
presentation.

Most often the stimulus behind biophysical activity taking place in a living
organism is the result of small electrical changes that occur within muscle and
nerve cells. These electrical changes are the result of biopotential differences.
As the name implies biopotenials are biologically based electrical potentials
acting as minute batteries.
The diagram illustrates the resting potential which remains steady at about
-70mV. But when commanded by the brain, a shift in the biopotential takes
place and moves from -70mV to +20mV when the muscle reaction is
undertaken. The shift amounts to a change of nearly 100mV as the muscle
transitions from a resting state to an action state.

These minute electrical changes within the muscle cells can be electrically
observed through external instrumentation. The heart (myocardium) is a multichambered muscle and its health is central to life itself. Therefore the heart is
often monitored using electrocardiography. The electrocardiograph is the
instrument that detects, signal conditions, records and displays the hearts
activity.
An important point to keep in mind is that even though the biopotential is
strongest at the source, by time it is detected at the body surface it has been
greatly attenuated making biophysical occurrences more difficult to detect and
separate from interfering electrical sources.

Biopotentials are developed from electrochemical gradients established across

cell membranes. These are voltage differences that exist between separated
points in living cells, tissues, and organelles. The potential difference
measured with electrodes between a living cells interior cytoplasm and the
exterior aqueous medium is generally called the membrane potential or resting
potential (ERP). This potential is relatively constant in striated muscle cells with
a potential of about -50 to -100mV. Nerve cells show a similar range2.
Related to these biopotentials are the ionic charge transfers, or currents that
give rise to much of the electrical changes occurring in nerve, muscles and
other electrically active cells3. This current is the direct result of the
electrochemistry associated with ions internal and external to the cell.
The biopotential plot has a rising section depicting depolarization and a falling
section indicating repolarization. Depolarization can simply be though of as the
electrical stimulation of the heart muscle cells. During depolarization the
muscle fibers shorten causing contraction. While during repolarization the
muscle cells relax, lengthen, and return to the resting state4.
2,3

Biopotentials and Ionic currents, Answers.com

Welch Allyn Protocol Clinical Support

The human heart cutaway shown in the diagram exposes the four chambers
the right atrium, right ventricle, left atrium and left ventricle. The function of the
right side of the heart is to deliver deoxygenated blood from the body to the
lungs. The function of the left side of the heart is to deliver oxygenated blood
from the lungs to the body.
The cardiac cycle consists of two phases - the Systole and Diastole. Although
these phases will not be further explored here, the waveform diagram
accompanying the cutaway shows the relative timing and amplitude of the
biophysical signals as the heart components go through a complete cycle.
The individual waves associated with each portion of the hearts function
sequence combine to produce the ECG waveform monitored on the body
surface. The resulting ECG waveform is shown at the bottom of the waveform
diagram.

The Cardiac Conduction System is the name given to the hearts electrical
conduction system. It controls the contraction of the heart. The SA node is
often referred to as the hearts pacemaker. It generates the electrical impulse
and sets the pace of the heart.
The Bundle of HIS is a thick bundle of nerves that transmits the electrical
impulses from the AV node to the Purkinje fibers. These fibers distribute the
electrical impulses to the individual heart muscle cells5.
Each wave and interval appear on the ECG display as the result of a particular
electrical function of the heart6. These individual functions are observed on the
ECG display and labeled as P,Q,R,S,T and U, corresponding to the particular
heart interval. Cardiologist assess the functionality and gross condition of the
heart muscle from these different segments of the ECG waveform.
5

Welch Allyn Protocol Clinical Support

The electrodes are transducers that detect the minute ionic currents
associated with the biopotenials. They can be thought of as an ion to electron
converter. This conversion allows the electrical currents to be amplified and
conditioned by external circuitry. The DAS/ECG board that will be described is
designed to perform these external functions.
The electrode is composed of silver (Ag) with a silver chloride (AgCl) surface.
When placed against the skin chloride is exchanged from the skin to the
electrode, and silver is exchanged from the electrode to the skin. In doing so
there is a free two-way exchange of ions, so no double layer is formed at the
surfaces.

For ECG applications three or more electrodes are placed on the body. The
diagram shows one of the most commonly used connections between the
body and ECG equipment. One electrode is placed on each arm, while a third
is placed on the right leg.
The arm electrodes are intended to detect the minute differential biopotentials
associated with the hearts activity. The third electrode, connected to the right
leg, provides a common mode drive voltage.
This third electrode serves two purposes; first, it may be used to impose a
common DC level on the patient. An example would the +2.5V shown in the
diagram which provides DC biasing, to the two differential sensors. And
second, it provides common-mode signal feedback to aid in common-mode
noise cancellation. The latter is very important because common-mode noise
may be hundreds to thousands of times greater than the detected ECG
biopotentials.
From the arm electrodes, the tiny differential signals are coupled to an
instrumentation amplifier (INA) for the first level of amplification.

The ECG Einthoven triangle dates back to the earliest days of

electrocardiography and provides the basis for electrode placement. The
equilateral triangle is formed by raising the arms and positioning the points on
the limbs equidistant. Either leg may be used for a lead connection and the
other leg then becomes the reference to which the other limbs are referenced.
The lead vectors associated with Einthovens lead system are conventionally
found based on the assumption that the heart is located at the center of a
infinite, homogenous volume conductor (at the center of a homogeneous
sphere representing the torso). With these assumptions, the voltages
measured by the three limb leads are proportional to the projections of the
electric heart vector on the sides of the lead vector triangle7. Einthovens Law
provides the voltage relationships between the leads.
With time this was perfected into the more commonly used connections today,
which may include as many as 12 electrodes. This allows the heart
biopotential activity to be monitored through many different planes.
7

buttler.cc.tut.fi

When the ECG electrode is physically contacted with the body a complex
electrical model is created. The model includes the body biopotential and
resistance, skin contact resistances and a parallel resistance and capacitance
associated with the probe. The right-hand diagram shows how each of these
subcircuits interconnect to create an overall equivalent circuit.
The electrode itself can be modeled as a 1F capacitor in parallel with a 10k
resistor. The 1F capacitor in conjunction with the 1k skin resistor inserts a
simple RC, low-pass filter function in the ECG path to the amplifier. Its cutoff
frequency is:
fC = 1/(2RC)
For the values shown fC is 159Hz. Although this may appear to be a low cutoff
frequency it is sufficient to pass the frequency components associated with the
ECG. For example, with a heartbeat rate of 60bpm, the fundamental frequency
is 1Hz. Even the fast R-wave potion with a duration of about 0.03 seconds at
60bpm, has a fundamental frequency of about 33Hz. But because this is a
quickly ramping up and down pulse, a greater harmonic bandwidth is needed.
The 159Hz satisfies the requirement for even shorter R-waves.
The bandwidth limited electrode/skin interface helps reduce the circuits
response to unwanted higher frequency electrical interference.

10

This is comparison of the fundamental frequency and bandwidth requirements

for monitoring blood pressure in the head and an ECG. The blood pressure
waveform has a period that coincides with the R pulses of the ECG, but note
the smoothness of the waveform as compared to the ECG waveform.
Therefore, the bandwidth requirements are much less for a blood pressure
monitoring application.

11

Here is an example of a normal ECG chart recoding for a heartbeat of 62bpm.

The rate can be determined from the rate of R wave occurrences. The
P,Q,R,S,T,and U portions of the ECG are labeled for convenience. A 1mV
calibration pulse is posted for comparison and has an amplitude of 500uV per
vertical division. Note that the R wave pulse has an amplitude about equal to
1mV, while the others are much smaller. Any electrical interference can easily
mask these important portions of the waveform.
The drift in the baseline is normal and can be due to the long charging time
constant of AC coupled circuits and/or the subtle changes in the electrode halfcell potentials associated with the ionic charge transfers (current).

12

These displays provide examples of irregular ECG tracings caused by both

internal and external factors. Muscle shaking is an example of an irregularity
caused by internal muscle tremors, referred to as a somatic tremor. The
gradual baseline drift discussed in the previous slide is due to charging of the
high-pass, coupling circuit and/or changes in the ionic current levels. So this
characteristic is connected with the equipment rather an internal bodily
function.
Sixty hertz AC pick-up is the result of induced electric field energy present in
the vicinity of the ECG equipment; often received by the electrodes or
electrode leads. Not only 60Hz, but any induced frequency such as RF can
disturb the ECG adding noise to the baseline.
Short-term DC instability may be an indication of an issue with the ECG
equipment.

13

The DAS/ECG demo board functions as a self-contained heart-rate monitor

providing a visual, audible, and digital indication of heart rate. The three ECG
electrodes are built in and conveniently accessed off one end of the board. If
necessary, external leads and contacts may be connected to the board as
well.
The demo board contacts provide the input for the differential ECG signals via
the right and left thumbs. Common-mode drive is accessed via a finger
electrode under the board. Since the board is only being used to detect heart
rate and not a detailed ECG pattern, precise Einthoven electrode connection
are not required.

A variety of different sensors may be directly interfaced to the board making

possible other types of medical-related and non-medical measurements.

14

The biopotentials detected at the body surface by the ECG are highly
attenuated relative to their point of origination. Often, the amplitude is on the
order of a few hundred microvolts (V). Other body signals such as brain
waves may have amplitudes a fraction of this level.
Very high voltage gain (V/V) is required to bring these minute signals to a level
where signal processing may be reliably applied. This is accomplished through
the use of high performance instrumentation and operational amplifiers on the
demo board. Additionally, on-board circuitry is provided so that the amplifiers
may be configured for sensor interfacing and filtering functions. These will be
discussed in more detail a little later.

Once the low-level signals are amplified the output is applied to the cardiotach
circuit. The amplified waveform is passed through a 150V peak-to-peak
threshold detector. If the amplitude of the waveform is sufficient, it will trigger a
one-shot multivibrator. The one-shot output may be counted, used to pulse an
LED, to key a 1kHz burst oscillator.
The DAS/ECG board also provides a probe point where the amplified ECG
waveform may be observed with an oscilloscope.

15

Moving to the next level of circuit complexity reveals the IC building blocks
used in the demo board:
1. U1, U2, U3 Input instrumentation amplifier and gain stages.
2. U4, U5, U6 Peak-to-peak detector and monostable multivibrator circuit.
3. U7 Low dropout regulator supplies +5V to power the circuitry.
4. U8 Auto power down circuit which is especially useful when using battery
power.
5. U9 An uncommitted op-amp useful for providing sensor interface.
6. U10 Provides a stable +2.5V reference voltage for mid-scale commonmode biasing.
7. U11 An optional socket for the OPT101 Monolithic Photodiode/SingleSupply Transimpedance Amplifier.

16

Here the analog front-end has been separated from the remaining circuits. A
precision, rail-to-rail INA326 instrumentation amplifier is at the front end
providing low offset (<100uV), a minimum CMRR of 100dB (114dB typ.) and
an adjustable gain from less than 0.1V/V to >10,000V/V. The INA326 gain is
set to -5V/V in this example.
The INA326 is followed by an OPA335 auto-zeroing operational amplifier that
features a maximum voltage offset of 5V, a voltage offset drift of 0.05V/C
and maximum operating current of 285A. Here the OPA335 is set to an
inverting gain of -480V/V. A first-order, low-pass filter may be configured within
the stage by the addition and selection of a feedback capacitor.

Since the board is powered by a single supply, it is necessary to establish a

mid-scale voltage. That is accomplished by connecting the +2.5V reference
voltage as a common-mode voltage to both the INA326 and OPA335.
The overall gain is the product of the individual gains of the two stages;
(-5V/V)(-480V/V), or 2400V/V. A 1mVP-P input is amplified to 4.8VP-P, centered
about +2.5V. The high common-mode rejection of the INA326 rejects the 60Hz
and other common-mode interference picked up by the electrodes. Likewise,
common-mode DC voltage is rejected by the amplifier.

17

Just the front-end portion of the INA326 is shown illustrating how the right-leg
DC drive voltage is developed and controlled. The INA326 gain set resistor, R G
is split into two equal resistors. Any DC common mode voltage present at the
two inputs will shift the DC level at the resistor junction. This voltage is
buffered by A1, and then applied to A2 which has an inverting gain of minus
19.5V/V. The inversion is important because it will be used to counter a DC
common-mode, electrode potential on the electrodes. A +2.5V common mode
voltage is applied to A2s non-inverting input via a resistive divider. The +2.5V
voltage is the mid-scale voltage level for all the analog circuitry.
A2 will amplify the difference in voltage applied to its two inputs and in turn
drive the common-mode potential applied to the right leg until it is equal to the
+2.5V reference voltage. This auto-zero feature keeps the DC level constant
which is necessary for a stable ECG display baseline.

18

This very busy circuit portion of the DAS/ECG circuit diagram provides the
remainder of the analog front-end circuit. The INA326 circuit includes a
provision for DC or AC coupling. AC coupling removes the DC electrode offset.
This offset is taken care of using a DC restorer circuit that will be discussed in
the next slide. The AC high-pass frequency response is selected at 0.05Hz,
0.5Hz, or 2.0Hz using a resistor-jumper provision.
The INA326 is followed by of a OPA2335, gain stage. The gain is set by
selecting an input resistor via a jumper. Additionally, a low pass filter function is
provided by this stage. Its cutoff frequency is set by connecting the appropriate
capacitor into the feedback path with another jumper.

19

The INA326 output voltage may be referenced to a voltage applied to the

reference pin, pin 5. If 0V is applied to the non-inverting input, then the output
will be referenced to zero volts and the swing can move up from 0V. If the
reference pin is set to +2.5V, then the output can swing above and below
+2.5V within the output bounds. This reference voltage is sometimes referred
to as a pedestal voltage, because it raises the output up from ground (0V).
The integrator shown in the schematic is referenced to +2.5V on the noninverting input. At DC the integrators gain is very large and any deviation from
+2.5V seen at the inverting input as the result of a common-mode DC
voltage on the INAs inputs - will result in a large DC voltage at the output. This
DC voltage is then applied to the INA326 reference input in such a manner as
to drive the INAs output back to +2.5V.
As the frequency is increased the gain of the integrator rapidly falls off. Thus,
AC signals having a frequency above the integrators -3dB cutoff frequency
have virtually no affect on the reference voltage applied to the INA.
The net result is a DC restorer circuit that compensates for a DC commonmode voltage, such as may be present with the electrodes. It also provides a
high-pass transfer characteristic with a cutoff frequency that is a function of the
integrator RC constant. This results in a circuit equivalent to a capacitively
coupled amplifier, but without any capacitors directly in the signal path. High
quality, high capacitance capacitors are often large and costly and are avoided
using this technique.

20

The output from the amplifier section may be sampled and processed by
external circuitry, or the onboard facilities provided on the DAS/ECG demo
board may be utilized.
The amplified ECG waveform is passed to a differentiator and peak-to-peak
detector that produces pulses at the heartbeat rate. These pulses trigger a
one-shot multivibrator which stretches the pulses to a uniform time duration.
The stretched pulses from the one-shot are then used to key a 1kHz burst
oscillator for a time period that corresponds to the one-shot pulse duration.
The burst oscillator has audible tone that is available through the speaker.
These pulses may also be used to flash an LED as a visual indictor of BPM, or
be counted by a BPM meter.

21

The first section of U4 is connected as an absolute value amplifier, producing a

positive going replication of the positive or negative going ECG wave. U4s
second section is a peak detector where the peak value of the ECG waveform
is stored on C11 (0.1F). The circuit has a lower threshold of about 150V.
U5s first section buffers the peak detector output, while the second section
amplifies and squares up the waveform. The input signal is amplified to a
level such that the second stage output runs rail-to-rail, nearly 0 to 5V. This is
ideal for triggering the first TLC556 section, which is configured as a 100ms,
one-shot.
The TLC556s second section is arranged as a 1kHz, astable multivibrator,
keyed by the preceding one-shot stage.

22

The DAS/ECG board may be powered by either a 9V alkaline battery, or an

external supply. Current varies from <1mA to about 15mA depending on the
boards operating mode.
A TPS71550, 5V low-drop out (LDO) regulator provides the supply voltage for
the board. Its +5V regulator output is supplied to the TLV3491 comparator
supply pin. An RC timing circuit is connected from the +5V supply line to the
inverting input, while a +4.5V voltage is developed at the non-inverting input.
The voltage at the RC node is initially zero and eventually charges to a voltage
exceeding +4.5V, in about 40 minutes. When that voltage is below +4.5V the
comparator output is high, at about +4.8V. Then supply current is delivered to
the DAS/ECG circuits through the comparators output stage. Once the
comparator RC input voltage crosses +4.5V, the output voltage drops to nearly
zero volts removing power from the DAS/ECG circuits.
A small amount of hysteresis is added to the comparator to improve noise
immunity at the threshold.

23

The DAS/ECG demo board has a number of features that make it easy to use for
testing circuit ideas and experimentation. In addition to the EGC cardiotachometer
application, it may be used for other portable applications where high voltage gain
and high common-mode rejection are required.

24

This image displays the top side of the DAS/ECG cardiotachometer board.
The left arm (LA) and right arm (RA) electrodes are located on the end of the
board, while the right finger drive electrode is placed underneath the board.

25

This shows some of the user selectable functions on the board. The gain and
low-pass and high-pass cut-off frequencies care established using jumpers
and can be changed as needed. There is an ON/OFF switch and start switch
for the 40 minute, power ON timer function. The speaker, LED and supply
connections are also shown.

26

Heres a more detailed layout showing the location of the analog circuits and
the tachometer circuits that follow them.

27

The back side of the ECG/DAS board contains the important common-mode
drive pad. This is typically biased at +2.5V when powered by a +5V supply. It
is important from the standpoint that it sources complementary phase, AC
common-mode signals back to the body. These add to the AC common-mode
signals on the body and help in the cancellation of these unwanted signals.
The image also shows the back side of the pin sockets that are used for wires
connections to the board and the +9V battery holder.
A brief set of instructions for the cardiotachometer use are provided on the
board, in the upper right-hand corner.

28

Here, John Brown the DAS/ECG demo board developer, demonstrates how
the board is held while in the standing position. The key to obtaining a good
cardiotachometer result is to gently grasp the electrode pads as shown while
holding the board steady. The board is easier to steady and maintain an even
contact while sitting, so do so if possible.

29

The cardiotachometer amplifier circuit is capable of detecting a biopotential of

about 200uVp, in the presence common-mode AC interference with an
amplitude of about 2Vp. Therefore, it is equally suitable for other applications
where very small signals may be buried amongst large common-mode signals.
Certainly other biomedical monitoring applications fall into this category, but
also analytical and scientific instrumentation, industrial monitoring, and some
automotive and industrial sensor applications as well.

30

Some other applications will be explored now to show the versatility of the
DAS/ECG design. This application will demonstrate how a bridge transducer
can be directly interfaced with the DAS/ECG board.
A puffing tube in conjunction with a bridge transducer will be used to detect a
change in gas pressure. Puffing tubes find application in industrial gas lines
and valves where the gas pressure and flow characteristics require
monitoring. Medical uses include applications where the tube serves to direct
the breath pressure of a user to the bridge transducer. The magnitude of the
breath pressure can then be used to control a medical assist apparatus such
as a wheelchair.

31

Silicon Microsystems manufactures a thin film pressure bridge transducer that

interfaces with a air lines, such as the puffing tube. The bridge connects
directly to each INA326 differential input. Current for the bridge transducer
may be supplied by the DAS/ECG, on-board +5V reference.
The transducers sensitivity in this application results in a differential voltage of
about 0.16 to 2.4mVp-p. A nominal value of 1.5mVp-p is used for illustrative
purposes. The gain is set to 2000V/V and this produces an output voltage of
2.5VDC 1.5VP for a range of 1.0 to 4.0V. If the differential voltage measured
2.4mVp-p, then the output range would span from 0.2V to 4.8V. The 2.5V
center voltage is from the pedestal voltage applied to the INA326 reference
pin.
The bridge transducer may have an offset, or imbalance between the two
sides as great as 50mV. Any input common-mode DC voltage and bridge
offset voltage will be auto-zeroed by U3 as previously discussed.

32

The ECG/DAS board bridge sensor input is shown coupled to a mechanical

pressure gauge in the puffing pressure bridge application. The output phase
between the mechanical gauge and the DAS/ECG board are set the same so
that both result in an upscale reading. The sensitivity of the mechanical gauge
is established at 5mm Hg for a 0.0375psi pressure change, while the bridge
produces a 0.75mVpk change for the same input.
As mentioned, the DAS/ECG board gain has been set to 2000V/V. This is
adequate for the bridge sensor output range. The DAS/ECG board has been
set with a bandwidth of 2Hz to 17Hz in this application.

33

Heres the actual oscilloscope display for the DAS/ECG board output with a
simulated puffing input (upper trace). The input puffing rate is a much faster
0.2s than a human can deliver, but illustrates the ability of the board to detect
and amplifier the bridge sensor output even at this higher rate. The output
swings approximately 1.5VP-P, and is centered about the +2.5V pedestal
voltage.
The lower trace indicates that the burst oscillator is being activated and it
provides pulses. The pulses can be counted and used to arrive at the puffing
rate.

34

This is pressure bridge application where DC or very low frequency

signals require monitoring. In this example the pressure change
associated within a squeezing a tube will be observed and measured.
The DAS/ECG will now be configured to provide DC coupling - versus
the AC coupling used in the previous applications. Now the bridge offset
must be taken into account to assure the DAS/ECG board output does
not saturate.

35

If the DAS/ECG board is configured for DC coupling any offset associated with
the bridge will be amplified by the very high circuit gain. That could result in an
voltage level that would exceed the amplifiers minimum or maximum output
level. Therefore, one must be cognizant of a sensors DC offset and the
direction it will drive the output.

36

The auto-zero feature has been disabled and the INA326 reference pin is
connected to zero volts. Notice that the overall gain has been reduced
substantially from its previous AC setting of 2000V/V, down to 100V/V. The
bridge offset is so large that the gain has to be limited to this much lower
value. This is to prevent the offset from driving the output into the positive
output rail.
For this example the bridge offset is 43mV and when multiplied 100x the
output is about +4.3V, placing the output close to the positive rail. However,
the bridge phase has been selected such that when the squeezing pressure
is applied the bridge resistances shift in the direction that moves the output
downward and away from the positive rail.

37

This image depicts how the squeezing tube is connected to the bridge and
also some of the DAS/ECG board settings for DC operation. The same
pressure bridge transducer is used here as earlier, but the bridge circuit
connections have been changed to assure the amplifiers operate within their
linear range. Gain resistors have been changed and the low-pass bandwidth
jumper set as needed.
Bridge bias is provided by the on-board TPS71550 LDO regulator. It has been
observed that the particular bridge used for this example resulted in a
differential offset of 43mV. The device is specified with a maximum offset of
50mV. If the offset was as high as 50mV, then the output would be up against
the rail. An alternative to lowering the gain would be to reduce the voltage
applied to the bridge.
A resistive divider in located on the board and divides the +5V down to +2.5V.
Since U2, the dual OPA2335 (or OPA2336) is not used in this application, it
can easily be configured as a unity-gain buffer. The output is then used to bias
the bridge, but note that doing so does reduce the bridge output by 50%.

38

The output response of the DC coupled DAS/ECG board during a squeeze is

displayed in this oscilloscope image. The upper trace is the response with a
+5V bridge excitation, while directly below it is the response with a +2.5V
bridge excitation. Notice that the amplitude change during the squeezing event
is about half with +2.5V excitation as compared to that with +5V excitation.
This is as expected.
Also observe that the event had a duration of about 5 seconds. This translates
to a frequency of about 0.2Hz. This is still within the boards AC passband
when the high-pass filter is set to a cut-off frequency such as 0.05Hz. Setting
the board for DC coupling may be the best option for use at even lower
frequencies.
Some examples of low frequency uses are geophysical, mechanical and
industrial process control applications.

39

Heres an interesting AC coupled application for the DAS/ECG board where

relative blood pressure may be optically detected and monitored. The circuit
configuration is that of a plethysmograph; an instrument used for measuring
changes in volume within an organ, body members or the whole body.
An LED is positioned so that its light output is directed through the finger. A
sensitive photodiode or a combined photodiode/transimpedance amplifier such
as the OPT101 is located on the other side of the finger. Fluctuations in the
blood volume within the finger changes the transmission path between the
LED light source and that reaching the photodiode. The blood volume
coincides with the pressure and the DAS/ECG board provides a relative
indication of the pressure.
Notice the connection of the photo diode and the 3 series-connected, 499k
resistors across the photodiode. The cathode end of the diode is referenced to
+2.5V. This same common-mode voltage appears at both of the INA326 inputs
through the resistors. When light shines on the photodiode, photo generated
current flows through the diode and through the 3 resistors.
One 499k resistor is connected directly across the INA326 differential inputs.
As the photo generated current changes in response to the blood volume
fluctuations a differential voltage is created across the resistor and is amplified
by the DAS/ECG board amplifiers.

40

The DAS/ECG board is shown outfitted with the monolithic OPT101

photodiode/transimpedance amplifier. The three, 499k bias resistors have
been added to the board. An overall gain of 6kV/V is used with the application
and the bandwidth has been set from 2Hz to 17Hz.

41

This oscilloscope display provides the output traces when the DAS/ECG is
connected in the plethysmograph application. The upper trace tracks the
changing blood volume within the finger indicating the blood pressure level. A
700mVP-P output amplitude results when the overall gain is set to 6kV/V.
The middle two traces are the 1st timers input and output pulses. The output
pulse corresponds with the peak blood pressure. This pulse is used to key the
output burst oscillator.

42

In summary, the DAS/ECG board is useful for demonstrating the ability of highperformance analog circuits in low signal level, front-end applications. The
boards versatility allows one to experiment, evaluate and optimize circuit
performance in medical and non-medical sensor applications.

43