We had a serious, most likely self induced, set of brownouts. (The robot was running well, hard to blame the driver). After we replaced the battery, we discovered that 4 of the 5 encoders on our 'bot (3 Mag encoders and 1 Lamprey) were fried; all 5 are powered off the rio DIO ports. The 1 that lived was ‘closest’ to the rio.
A visit to the DS log showed 52 brownout events.
The hypothesis is that the brownouts caused this problem. A smarter mentor than myself suggested that it could have been an ‘impedance shock’ (I think those were his words, sorry Nick, if I mangled it).
Has anyone else seen this? Is this a reasonable hypothesis?
We’ll be putting current limits on all motors and retesting, but it’d be nice to get affirmation for our theory.
I get what he’s suggesting (turning the 5v off and on too frequently resulted in getting >5v to encoders due to reflections of the inrush current down the wire), but don’t know enough EE to estimate whether that’d actually exceed component ratings. Seems pretty exotic on a gut check.
Usually excess voltage is absorbed by the battery (on the main 12vdc circuit). I don’t know if the regulator relies on that too.
Source of the brownouts is fairly clearly 4 falcon swerve drive motors + 2 kraken shooter motors + drive team having a blast + no current limits. Mebbe current limits should go higher on our priority list next time .
Hm. Sounds like what we did four times yesterday, but we’ve never burnt sensors when stacking brownout events. I’d still check your battery and main power wiring (including roborio power) to see if there was something intermittent, because losing contact to the battery could allow excessive voltage on Vbatt.
More detail on “fried” might help. This sounds unusual. I’d check all your batteries to be sure there isn’t one with backwards wiring on it (+/- reversed).
We didn’t see anything obviously wrong with the wiring, although I can’t say we studied it in great detail. I would think I would see some evidence of that in the DSlog, though, and I don’t.
As far as ‘fried’, the three MAG encoders which normally have a green light had no light at all and were not reporting any value. Plugging a new mag encoder into the same wiring harness worked fine. The lamprey was still showing status lights but was not returning any values.
.