DOC-00032 Rev-A AppNote303 Stabiliti-FaultMonitoring Published
DOC-00032 Rev-A AppNote303 Stabiliti-FaultMonitoring Published
Ideal Power provides a robust means of monitoring and managing Power Converter System (PCS) faults
via the Modbus interface. This Fault Application Note details the specific fault implementation and support
logic found in the Stabiliti™ Series products, Ideal Power’s 3rd generation 30 kW PCS. The PCS constantly
checks internal and external conditions to ensure that the converter is operating properly and that external
conditions are within allowed safe operating ranges. A fault occurs when a condition is encountered that
is outside of a pre-defined safe operating range. This document describes the fault handling information
and data that is provided to users via the Modbus interface along with detailed descriptions of each fault.
Recommended site controller responses to reported faults and PCS operating states is also covered.
This application note should be used in conjunction with other product and safety documentation provided
by Ideal Power (“IPWR”). The intended audience is engineering and lab personnel familiar with high-
voltage/high-power systems, and the general safety issues related to the wiring and use of 3-phase AC;
battery systems; and PV energy sources utilized during the evaluation and qualification testing of power
converters. Additionally, this document assumes that the reader is already familiar with the Stabiliti 30C or
30C3 Modbus Interface, and is comfortable using that interface to configure, monitor and command PCS
operation.
Applies to Stabiliti™ 30C & 30C3 products with the following firmware versions:
FPGA 3.1.845
Communications 3.2.93
CONTENTS
Ideal Power PCS allow for a partly automated fault handling system which broadly operates as follows.
When a fault limit is reached during operation, the PCS will halt operation and set a unique flag indicating
the fault type in an Active Fault register and then start one of three internal reconnect count-down timers.
Once the count-down timer reaches zero, the PCS will reconnect and resume operation as prior to the
fault. Note that the fault must be clear at the end of countdown sequence in order for the PCS resume
operation, and that any fault reoccurrence during the countdown period will restart the countdown
process. If a fault persists indefinitely, the PCS will be stuck in the countdown mode until the fault finally
clears.
Once the fault condition has passed the PCS will clear the Active Fault flag and increment a counter on
the associated Occurred Fault register. The PCS may also be operated in a manual mode which is typically
meant for test or troubleshooting operation. If set to manual operating mode, the PCS will not automatically
attempt to recover after a fault: the external system controller must issue a user_start command via the
Modbus interface.
Each fault is associated with one of the three count-down timers. In particular, faults associated with
IEEE1547 over/under voltage & over/under frequency grid conditions are assigned to reconnect timer #0
which has a factory default of 320 seconds to conform with utility interconnection requirements. Timers #1
and #2 have 20 second factory defaults. Some faults are not transient by definition, such as with a blown
GFDI fuse, and do not self-clear. These select lockdown faults are not assigned to a reconnect timer and
typically require on-site service to clear. For all other cases, an external acknowledgement or reset of the
fault is not required as is sometimes demanded by other equipment employing latching fault states.
There are 54 reported faults, which are all described in the Fault Descriptions section of this document.
The fault severity level determines if the fault is simply recorded, or the converter response is to shut
down.
The severities are as follows and are detailed further in the description of Modbus Register 9.
FAULT DATA
When a fault does occur, registers 24 – 27 show which fault(s) have occurred since the last time that faults
were cleared. In addition, registers 16 – 19 show which faults are currently active. To see more detailed
information about a given fault that has occurred enter the number of the fault in question into register 0,
then check registers 1 – 9 for specific fault details. In the case where a particular fault has occurred more
than once since it was last reset, the values shown in register 1 – 9 represent the most recent occurrence.
Available fault information includes the fault limit, fault value, fault counter, a timestamp, fault selector and
status.
Ideal Power recommends the following practice for an external system controller to monitor faults. Details
of each item are provided in later sections of this document.
1. Monitor the System Status register 298 for a snapshot view of the fault statuses and reconnect timer
activity indicating a fault is active or a fault occurred and a reconnect count-down has begun.
2. Monitor the Active Faults registers 16-19 in real-time and automatically parse the bit fields to identify
the fault ID. Continuously monitor Active Faults, log and timestamp once identified.
3. Monitor the Occurred Faults registers and automatically parse the bit field to identify the fault ID.
When a fault occurs, iteratively query the fault data for all faults that have occurred. Log this fault
data with a timestamp.
4. If the user encounters recurring active faults or repeating fault state oscillation, human intervention
may be required. Consider implementing a system controller timeout to discontinue system start/
stop oscillation as a result of constantly repeating faults. Review the fault listing below for further
explanation of fault causes and possible paths to fault resolution.
5. After a period of time logging faults and data the fault history can be cleared/reset by way of register
12, fault_global_control, in order to start fresh.
Note: Many faults are transient in nature, and the triggering condition will pass quickly after the PCS
has halted operation. As a result, the system controller may not capture every Active Fault flag as such
transient faults occur. The only means to identify such faults is to query the Occurred Fault registers, and
log fault counts. However, the user should recognize that an externally induced fault state may appear to
clear but may be immediately triggered again, during or after the countdown process, which will reset the
countdown state. This may result in an oscillating fault state, as described above.
The Stabiliti™ PCS includes a Black Box Fault Recorder system that automatically captures and saves
operating data around the time of a fault. When a user is reporting failures, persistent or repetitive faults to
Ideal Power we require that the applicable Black Box fault log also be sent in order to better support our
customers.
The Black Box data files are accessible through the PCS Ethernet connection over HTTP port 80. They
are located in the ‘faults’ folder in the root directory of the PCS webserver as shown in the second image
below. One FLT file is generated per day to capture any faults occurring that day and is named with the
date in the form <MM-DD-YY.FLT>. Note that FLT files are binary formatted and only readable by Ideal
Power.
The FLT files may also be downloaded manually from the webserver with a web browser at the PCS IP
address as shown in the first image below.
The LCD front panel display of the PCS includes fault indicators and information.
• Ground Fault Indicator ‘G’ – When this fault indicator is solid color red, there is a Ground Fault within
the system. This indicator is dedicated to Fault #0 and Fault #1 as described below.
• Fault Indicator ‘F’ – When this fault indicator is solid color red, there is an active fault or a fault that
has resulted in a lockdown of the system.
Fault information is provided in the 3rd grouping of text down from the top of the display. The most recent
fault is indicated first. For faults that occurred simultaneously, the fault with lowest ID number will be
displayed. Second, active Fault Sets 0-3 are displayed in hexadecimal format along with an indicator of
how many faults are currently active. An example image is provided below along with its interpretation.
FIGURE 2
SAMPLE FRONT
PANEL FAULT
DISPLAY
Note that this set of four faults are the standard response to a loss of AC supply (utility grid).
REGISTER MAP
Below is a detailed description of the Modbus holding registers that pertain to fault status and fault data
reporting. In each case a chart shows the usage of the 16 bits for each Modbus Register. Also refer to the
Modbus Map provided by IPWR for your version of firmware. Access levels include Read & Write (RW),
Read-Only (R) and Write Only (WO) which indicates the PCS will clear a command after it is written.
Access: RW
Default: NA
Format: Unsigned 16-bit value, 7-bit integer
Limits: 0 to 63
Description: The fault index is written by a user to select which fault information they would like to be
provided in the following registers 1 – 9.
Access: R
Default: NA
Format: Signed 16-bit integer
Limits: NA
Description: If the fault is associated with a value going beyond a given limit, the limit that was exceeded
is shown here.
Access: R
Default: NA
Format: Signed 16-bit integer
Limits: NA
Description: The actual measured value that exceeded the fault limit and triggered the fault flag.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description: The number of times a fault of this type has occurred since the last time that faults were
reset.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description: Bits 15:0 of the 32-bit UNIX timestamp of the most recent occurrence of this fault type.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description: Bits 31:16 of the 32-bit UNIX timestamp of the most recent occurrence of this fault type.
Access: R
Default: NA
Format: Unsigned 16-bit value, bit field
Limits: NA
Description: In some cases, such as line over-voltage, a fault may have multiple sources. In those cases,
this register will represent which source caused the fault. See the Fault Descriptions section of this
document for the meaning of the selector for a given fault.
Access: R
Default: NA
Format: Unsigned 16-bit value, bit field
Limits: NA
Description:
b0-b2: The severity of the fault.
0x000 = Info: increments the fault counter only.
0x001 = Alert: increments the fault counter only.
0x010 = Alarm: fault is logged.
0x011 = Abort 0: fault is logged and unit is stopped. Reconnect timer 0 is used for restart.
Access: WO
Default: 0x0000
Format: Unsigned 16-bit value, bit field
Limits: 0 to 1
Description: Write a 1 to reset (clear) all fault counters and history. Register self-clears, will always read
back 0.
Access: R
Default: NA
Format: Unsigned 16-bit value, bit field
Limits: NA
Description:
b0 – b15: 1 = fault currently active, 0 = fault not currently active.
Address 16 is for faults 0 -15.
Address 17 is for faults 16 - 31.
Address 18 is for faults 32 - 47.
Address 19 is for faults 48 - 63.
Faults 53 through 63 are currently not defined and reserved for future use.
Access: R
Default: NA
Format: Unsigned 16-bit value, bit field
Limits: NA
Description:
b0 – b15: 1 = Fault occurred since last reset, 0 = fault has not occurred since last reset.
Address 24 is for faults 0 -15.
Address 25 is for faults 16 - 31.
Address 26 is for faults 32 - 47.
Address 27 is for faults 48 - 63.
Faults 53 through 63 are currently not defined and reserved for future use.
Access: RW
Default: 320
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_set_0 register defines soft reconnect time after a disconnect triggered by a fault of
severity level Abort 0. This soft reconnect time is only used when the PCS in auto-control mode following
an Abort 0 fault clearance. This register is in units of 1 second. Changing the duration of reconnect timer
#0 may invalidate your utility interconnection agreement.
Access: RW
Default: 20
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_set_1 register defines soft reconnect time after a disconnect triggered by a fault of
severity level Abort 1. This soft reconnect time is only used when the PCS in auto-control mode following
an Abort 1 fault clearance. This register is in units of 1 second. Ideal Power recommends this set-point not
be set less than 3 seconds.
Access: RW
Default: 20
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_set_2 register defines soft reconnect time after a disconnect triggered by a fault of
severity level Abort 2. This reconnect time is only used when the PCS in auto-control mode following an
Abort 1 fault clearance. This register is in units of 1 second. Ideal Power recommends this set-point not be
set less than 3 seconds.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_stat_0 register reads back the current state of the reconnect_timer_0 timer in the
PCS. This reconnect timer defines soft reconnect time after a disconnect triggered by a fault of severity
level Abort 0. This reconnect time is only used when the PCS in auto-control mode following an Abort 0
fault clearance. This register is in units of 1 second.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_stat_1 register reads back the current state of the reconnect_timer_1 timer in the
PCS. This reconnect timer defines soft reconnect time after a disconnect triggered by a fault of severity
level Abort 1. This reconnect time is only used when the PCS in auto-control mode following an Abort 1
fault clearance. This register is in units of 1 second.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description:
The reconnect_timer_stat_2 register reads back the current state of the reconnect_timer_2 timer in the
PCS. This reconnect timer defines soft reconnect time after a disconnect triggered by a fault of severity
level Abort 2. This soft reconnect time is only used when the PCS in auto-control mode following an Abort 2
fault clearance. This register is in units of 1 second.
Access: R
Default: NA
Format: Unsigned 16-bit value, bit field
Limits: NA
Description:
Provides information on fault activity and the reconnect timers statuses as follows:
b1: 1 = Reconnect timer #0 is active, an Abort 0 category fault occurred.
b2: 1 = Reconnect timer #1 is active, an Abort 1 category fault occurred.
b3: 1 = Reconnect timer #2 is active, an Abort 2 category fault occurred.
b11: 1 = Lockdown is active, a Lockdown category fault is active or occurred.
b12: 1 = PCS fault of severity level Abort 0 is active.
b13: 1 = PCS fault of severity level Abort 1 is active.
b14: 1 = Fault of severity level Abort 2 is active.
b15: 1 = GFDI fault or IMI fault detected.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description:
The gfdi_header register reads back the installed firmware version identifier for the GFDI/IMI fault
detection PCS sub-system.
Access: R
Default: NA
Format: Unsigned 16-bit hexadecimal value
Limits: NA
Description:
The imi_data register reads back the computed impedance detected between the DC-COM node and earth
ground. This impedance is calculated once every 30 secs. If the computed impedance is below a built-in
threshold of 4kΩ, the PCS system asserts an IMI fault.
Access: R
Default: NA
Format: Unsigned 16-bit integer
Limits: NA
Description:
The gfdi_data register reads back a non-zero value if the PCS ground fault fuse is intact, or closed. If the
fuse opens (blows) due to a system DC ground fault, the GFDI fault detection sub-system will write a value
of 0 to the gfdi_data register indicating detection of the ground fault.
FAULT DESCRIPTIONS
The faults that can occur in the Stabiliti Series are listed here along with a brief description.
Faults in the Abort 0 category are associated with utility interconnection trip limits and the AC input should
be investigated for troubleshooting.
Faults in the Abort 1 category are associated with external PCS faults. To troubleshoot, review the fault
description and investigate the applicable external condition.
Faults in the Abort 2 category are associated with internal PCS faults. Typically, there are no actions
an operator may take to troubleshoot these faults. Such faults may occur occasionally. If such a fault
begins to occur consistently under a particular condition, please contact [email protected] for
troubleshooting assistance.
When contacting Ideal Power for fault handling support, please provide all the available fault
information provided by the PCS through the Modbus interface and Black Box Recorder as described in
this document.
Note: A tabulated listing of all the faults may be found at the end of this document in the Quick Reference
section.
Hexadecimal Reference
Decimal 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Hex 0 1 2 3 4 5 6 7 8 9 A B C D E F
Binary 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 1111
This document is provided “as is” and Ideal Power Inc. (IPWR) makes no representations warranties, expressed or implied, with respect to theinformation
contained herein. IPWR has made reasonable efforts to ensure the accuracy of information herein at the time of publication; however,information is constantly
evolving and IPWR does not purport the information provided is correct, comprehensive or exhaustive. This document is for informational purposes only; you
should not act upon information without consulting IPWR or its authorized distributors.
© Copyright 2017, Ideal Power Inc. All rights reserved. No parts of this document may be reproduced in any form without the express written permission of
IPWR. Ideal Power Inc., and the Ideal Power logo are trademarks of Ideal Power Inc. All other trademarks and service marks belong to their respective owners.