Randomly pulsing, ghost regen , crazy speed data

[Mark Stosberg]

My e-bike has a number of problems which may or may not be related:

• Most troubling, the assist will randomly cut-out and recover while throttle is being held down. Often the assist is "out" for just a second two, creating a random pulsing feeling. This happens even when the voltage is good (38V for a 36V battery), and is not related to a "low battery" condition.
• The CycleAnalyst will randomly report crazy inaccurate speed readings, from 0mph to 600 Mph. The sensors around the fruit wheel appear fine. The electrical data appears stable even while the speed goes crazy.
• Even when the bike is stationary, the "regen" screen report that the battery is rapidly being regenerated.

I'm riding a Yuba Mundo with:

• Nine Continent direct drive front wheel
• Cycle Analyst (2.2?)
• 35A Infineon controller
• 15Ah LiFePO4 controller

... I already suspect that the CA calibration may not be optimal. Even when the battery is new, it appears to be "empty" (31.5V) at only 7 AH instead of something closer to 15 Ah. Because I have still gotten a reasonable amount of range out of the battery by that point, I don't believe the battery is "broken".

What are possible problems that I might be having with my bike? Thanks!

Mark

 

[icecube57]

Email them directly I suspect there is a resistor that has gone bad for the display but your voltage is too low for that to happen. This normally happens with people with 100v setups on older Cycle Analyst.

 

[NeilP]

If the CA is new, and seems to work correct, then maybe something to do with, as you suggest, the calibration.

If you have not already got the manual for it, go to the e-bikes.ca website and download it.

There is a screen in SETUP|ADVANCED thsat allow adjustment of the dsplayed voltage as compared to the actual voltage.

The default figures are in the book, so you can always go back to orignal values if your changes make things worse.

Try a ZERO Current reset. again in the SETUP|ADVANCED menu

Work on the theory that if it is already damaged, then ou have nothing to loose. The changes I mention can't harm anything.

The pulsing sounds like it is the Low Voltage Cut out or speed limit settings kicking in, maybe because the voltage sensing is wrong

 

[jonathanm]

Sounds like the problems are going on in the CA. maybe an intermittent short or something not set up right.

Disconnect the CA and see if you get the pulsing without it, this will confirm if the CA is the source of the problem.

Then check you have it set up right.

 

[amberwolf]

[*] Most troubling, the assist will randomly cut-out and recover while throttle is being held down. Often the assist is "out" for just a second two, creating a random pulsing feeling. This happens even when the voltage is good (38V for a 36V battery), and is not related to a "low battery" condition.

[*] The CycleAnalyst will randomly report crazy inaccurate speed readings, from 0mph to 600 Mph. The sensors around the fruit wheel appear fine. The electrical data appears stable even while the speed goes crazy.

[*] Even when the bike is stationary, the "regen" screen report that the battery is rapidly being regenerated. [/list]

Are all these problems sudden/new, or existing from the beginning of setting up this bike?

If the former, somethign has probably either gotten damp/wet or partly disconnected. If the latter, a bad connection could still be the problem but is probably not as visible as the former might be.


Is the CA a DP (connected to 6-pin plug on controller) or SA (standalone with separate shunt and speed sensor)? I presume SA based on your mention of the wheel sensor.

If the former, a bad connection at that 6-pin plug could cause the CA erratic speed reading problems.

If the latter, the speed sensor could be defective or a problem could exist in the cable from it to the CA.

Is the CA connected for current or speed limiting? (this requires extra connectors beyond either the 6-pin controller plug and/or a separate shunt and wheel speed sensor)

If so, it could be causing the cutout issues due to bad data from a bad connection, otherwise it is probably not related, unless it is a bad connection between the battery input to the SA model's shunt, and the output to the controller.

Or is it only connected to monitor battery current flow, and speed sensing?

If so, it is unlikely the CA causes the cutout issue, and is probably separate. Probably still a bad connection somewhere, though.



The regen thing may just be an issue where you just need to zero the current in the menu, if it has always done this. If it just started, soemthing may have become corrupt in a CA setting, and zeroing should still fix it. If not, and it begins drifting again, I'd ask ebikes.ca about it.


The battery not reading the correct Ah may be a shunt issue. If you have the SA model, make sure your shunt is set to around 1.3mOhm (it might be a bit above or below that). If you have the DP, you'll need to know what the shunt value is supposed to be for your controller, or else find out via one of the methods posted around ES.

 

[chvidgov.bc.ca]

Check the CA setting "number of poles" - get the right number from Justin, for your motor. I recall once my motor was acting flakey and I had the wrong number of poles set for the motor.

 

[Mark Stosberg]

Thanks to everyone for the replies! It's always so helpful here. I hadn't checked back sooner as I was expecting to be notified by emails of replies. I'll try review the advice again in detail, try some things out and report back on my progress.

 

[NeilP]

Check the CA setting "number of poles" - get the right number from Justin, for your motor. I recall once my motor was acting flakey and I had the wrong number of poles set for the motor.

Only if you DO NOT have a speed sensor. I have a CA-DP with speed sensor.

I first setup my unit with number of poles for my motor and got totally wrong speeds. If you have a speed sensor ( And one magnet on the wheel) set just one pole. But I do not think that is the problem here.

I would unplug the CA to start with and see if the problem persists...at least then the issue can be narrowed down to controller or CA settings or fault.

Oh, and go to your User Control Panel/Board Preferences/Edit Posting Defaults to Notify me upon replies by default


to save having to remember to check the box each time

 

[Mark Stosberg]

To answer some questions:

• I've the bike and electric system about a year, which makes it seem fairly "new" to me, but during that year, I've already replaced the battery, controller and charger due to failures, so there is a history of problems. At once point, the charger and battery were mismatched (35A charger and 10Ah battery), which seemed to kill the battery, and may not have been good for the CA either.
• I *do* have a 6-wire connector from the CA to the controller, *and* a wire running down to a wheel sensor. I also manually soldered in a third wire for a front LED light. That part keeps to go smoothly. ( There's now a related splitter elsewhere in the wiring to power the rear light.
• The "random pulsing" is new. I've experienced low-voltage pulsing before. And this is different in a couple ways. The low voltage only starts after after 20 miles / 6 Amps, and has a rhythmic nature, getting steadily worse. The new pulsing happens much earlier in the ride at higher voltage, and feels more random. There is no "steady decline" effect with it. After some pulsing, the bike will recover back to full speed for some significant period.
• The "random speed" reading is also new. Perhaps it started at the same time as the random pulsing issue. I can't be sure.
• The "ghost regen" issue the fact that my battery capacity only appears to 7Ah are old issue. They don't seem to have any ill effect besides reporting, so I haven't worried about them yet.

The advice here centered around the CA. I've unplugged it for now, and will ride it long enough to see if that seems to fix the random pulsing issue. Although electrical issues are not my strength (I'm a software engineer), I did get a multimeter for Christmas and enjoy directed experiments with it if that would help helpful later in the diagnosis.