I'm using the AD8280 in a Lithium Polymer Battery Pack Monitoring System as safety device (together with the AD7280A device used as main monitoring device).
While testing the system functionality I found that the trip points of Overvoltage, Undervoltage and Overtemperature have no Hysteresys, in spite of what stated by the datasheet (in the TRIP POINT ERRORS section of Table 1).
Or maybe my understanding of the Hysteresis concept applied to trip points is not correct.
To give an example, the behavior I expected was the following:
1) One (or more) of the monitored cell voltage inputs reaches the Overvoltage trip point (that is the voltage at the OV input);
2) The AVOUTOV output became active;
3) The AVOUTOV stays active until all the monitored voltage inputs drop below the trip point MINUS the stated Hysteresis (at least 40 mV)
The behavior I get is instead:
2) The AVOUTOV output is activated and deactivated continuously, following the slight variations of input voltages;
3) Further increasing of input voltages leads to a more and more stable activation of the AVOUTOV output;
4) When one (or more) of the monitored cell voltage is more than 60mV greater than the Trip Point the AVOUTOV becomes steady active;
5) Decreasing of cell voltages leads again to AVOUTOV blinking;
6) Dropping all the monitored cell voltages back below the overvoltage trip point leads to a stable deactivation of the AVOUTOV output;
Undervoltage and overtemperature alarms show the same behavior.
I've checked out the OV, UV and OT voltage levels and they are totally stable as they are generated by means of resistive voltage dividers starting from the REF voltage.
Am I missing something or is it a misbehavior?
Thank You and regards,
I have moved your question to a more relevant community. Someone here will be capable of assisting you with this question.
you ou have the correct understanding of what hysteresis should do. Is it possible that there is a large amount of noise on the inputs to the AD8280? Perhaps higher frequency noise that does not show up on a meter But would show up on a scope? Are you using the deglitch function?
Thank You for your support.
I'm testing the board using a National Instrument PXI System, cell voltages are generated with an analog output module (PXIe-4322) that is pretty precise and stable: the maximum absolute error on the generated cell voltage is 1.7 mV.
There are no source of noise near the test area, I've always a scope connected to be sure and double check the voltage levels.
I'm not using the deglitch function because this way the self-test operation takes the smaller time possible, but as far as I know it shouldn't have any consequence on the Hysteresis function.
I have to rectify an incorrect information I gave previously: the Hysteresis is correctly applied by the AD8280 while detecting the overtemperature condition. Only the Overvoltage and Undervoltage show the misbehavior.
In the meantime I've tested some more devices and they all behave the same way.
Would it be possible for you to share a scope screen shot of the input voltage? Are you monitoring the cell voltages differentially with a scope? Are the cell voltages generated by 6 different outputs from the PXI board or are you using some kind of divider to generate the cell voltages?
We do test the hysteresis function on every part during ATE testing so I am leaning toward your observations being setup related.