CA detecting overcurrent erroneously

[E-HP]

I experienced some strange behavior at the end of my ride yesterday, and again at the beginning today. From a stop or from slow speeds, if I twist the throttle, power to the motor is being cutoff, but then would start functioning normally. A first I thought it was a throttle issue, but it only happened a couple of time yesterday. Today, when I rode down to the bike path, it happened again. It resumed functioning, but did it again after a few yards. That happened few times, then was dead for good, seemingly. This is starting off barely on the throttle.

When it happened the last couple of times, I saw the CA watts hit 10kW (9.8kW), before the motor cut off. It wasn't actually pulling that much or I'd be lying on my back, but the CA thought so. I opened my connection box and wiggled the wires around the external shunt, and the JST connector I have between it and the CA. After that I rode about 20 miles without incident.

While wiggling the wires seems to have fixed the issue, I'm at a loss as to what the technical problem is/was. For the CA to think it's seeing 10kW, then voltage across the shunt would have to be higher, not lower, and a poor connection logically would more likely result in the voltage across the shunt to be lower or zero. I'd really like to understand what the possibilities are, since it came on randomly, and I don't want it to happen 20 miles from home. Disconnecting the shunt might work, if I really got stuck, if the issue is related to it, but I'm not even sure of that.

 

[fechter]

The CA is measuring from both sides of the shunt. If there was a bad connection between the shunt itself and the big wires it could have lots of voltage across it. This includes somewhere in the middle of the shunt.

What kind of shunt are you using?

 

[E-HP]

The CA is measuring from both sides of the shunt. If there was a bad connection between the shunt itself and the big wires it could have lots of voltage across it. This includes somewhere in the middle of the shunt.

What kind of shunt are you using?

Here's what I'm usijng:

https://www.amazon.com/gp/product/B08477W6Q9/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

So poor connection to the battery or controller negative/ground wire could cause the spike?

 

[amberwolf]

For an incorrect current (and thus power) reading, the most likely problem is a poor connection between the shunt and the CA. I've had this cause ultra high power readings, which then cause the CA to go into A or W limiting (capitalizing the corresponding letter in the diag screen's lower left side).

In my case it's usually the plug from the CA to the shunt itself; the shunt I'm using is wired directly into the battery-to-controller path, but I"m still using the old 6-pin JST that came on the shunt and sometimes it's connection gets poor. Unplugging and replugging the JST has fixed it every time so far. (except once when the blue wire pulled out of the contact, and I had to recrimp it).

If you are out riding somewhere and don't want to or can't modify the wiring / disconnect stuff, you could change the CA's watt or current limit upwards to beyond the reading you get that triggers the limiting, until you can get home and deal with the problem. And/or change the shunt resistance in the CA so the reading it gets is lower.

 

[E-HP]

For an incorrect current (and thus power) reading, the most likely problem is a poor connection between the shunt and the CA. I've had this cause ultra high power readings, which then cause the CA to go into A or W limiting (capitalizing the corresponding letter in the diag screen's lower left side).

In my case it's usually the plug from the CA to the shunt itself; the shunt I'm using is wired directly into the battery-to-controller path, but I"m still using the old 6-pin JST that came on the shunt and sometimes it's connection gets poor. Unplugging and replugging the JST has fixed it every time so far. (except once when the blue wire pulled out of the contact, and I had to recrimp it).

If you are out riding somewhere and don't want to or can't modify the wiring / disconnect stuff, you could change the CA's watt or current limit upwards to beyond the reading you get that triggers the limiting, until you can get home and deal with the problem. And/or change the shunt resistance in the CA so the reading it gets is lower.

That must be it. I'm using a two pin JST that goes to my homemade CA JST connector (combined with the controller inputs). I definitely wiggled that connector when I opened up the connector box. 9.8kW is way over what I currently have the CA set to limit at (~6kW), so it pretty much cut the power right away. I didn't pay attention to whether A or W came on, but I think it would be W since I'm limiting on watts.

The current limit trick is a great idea to have as a backup. It's been pretty wet out lately, so maybe just spraying around in the connector box with WD40 might be a good idea to prevent any more intermittent connections.

Thanks to you both for the info and providing a direction to look. I'm going to do some more reading on how spikes get created from the poor connections to fill the gap in my logic.

 

[amberwolf]

I fill the JST housings with vaseline (anything dielectric and jellylike works) to keep the water out. Does alright, but in the summer it gets so hot down near the pavement (the shunt/etc is all under the trike) that the stuff melts out of the connector, leaving just a greasy coating (which still does the job).

Tried WD40 but it didn't work very long--we hardly ever get enough rain to worry about, but when we do it's a lot. In between it's usually hot and dry, and I guess (don't know) the oils evaporate or are contaminated with dust/dirt/etc that then attracts the water/moisture/etc.

The same thing happens eventually with the vaseline (or other dielectric jelly), but it takes a lot longer, giving me more time to remember to redo it.

I don't use it on the connectors up top like on the handlebars anymroe, because it will make a mess on anything anywhere near it as it flows along the surfaces, and I'll end up getting it on my hands.


BTW, if you find you can't turn the limit up enough to get past the problem, you can set the CA to BYPASS mode to use the throttle directly, unmodified. Doesn't help for PAS, which still wont' work, and if there are sensitivity issues or other reasons you use the CA to modify throttle, those will rear their heads.