Power Oscillations / Hunting at higher speeds

I recently bought a kit from ebikes .ca:
- TDCM_IGH_305
- 25A Controller
- CA3
- 12Pole PAS

I'm having issues with it hunting to find a wattage.

I've set the autoPAS to give 300 Watts when activated, no change at different RPMs.

I have had to pull Wgain down to 5.00 so that hunting didn't cause the bike to be too jerky. Anything lower and it really takes a long time to get up to the power level.

At any speed below 23kph, I don't have any issue with watt hunting. As I go faster the watts swing farther below and above 300 the faster I go.

I've tested to see if it happens at autoPAS set to other powers (350, 400) and it does. It also happens at any pedal cadence with any amount of force that I put into the bike.

Any idea what I should be looking for? The power oscillations/Hunting are really annoying.

Xoc said:

I recently bought a kit from ebikes .ca:
- TDCM_IGH_305
- 25A Controller
- CA3
- 12Pole PAS

...
I've set the autoPAS to give 300 Watts when activated, no change at different RPMs.

Click to expand...

What firmware are you running? There hasn't been an 'AutoPAS' setting for a long time.

Start by flashing to 3.12 which is the latest supported version.

It's surprising that this is happening with such low power. Examine the Limit Flags on the Diagnostics screen to ensure it is really Watts limiting.
Many settings interact - instead of playing 20 Questions to find out how it's configured, it's easiest thing to just review your config. Extract the configuration showing this behavior from the CA (don't use an existing file) and post it up here.

Thank you for the direction so far.
I'm sorry, it's not AutoPAS. It's labeled as "Basic (PWR)" under PAS -> PAS Mode
I'm running 3.12
I guess I was just reading on Grin's website AutoPAS everywhere.

I've checked the Diagnostic screen and it's only showing a capital W when it's happening.

I don't have a COMM cable, but it looks fairly simple so I'll try to make one. I'll post the config file once I've figured it out.

Thanks

Xoc said:

it's not AutoPAS. It's labeled as "Basic (PWR)" under PAS -> PAS Mode
I'm running 3.12
I guess I was just reading on Grin's website AutoPAS everywhere.

Click to expand...

Okay. 3.1 doc hasn't come along yet in favor of the Setupup Utility Help File and Grin CA3 videos.
Download the Setup Utility and use it to pull down the newest firmware. Then open a NEW 3.12 setup file. The Help file and screen ToolTips will then give you 3.12-specific info.

Xoc said:

I've checked the Diagnostic screen and it's only showing a capital W when it's happening.

Click to expand...

Okay - so a real power limit issue....

Xoc said:

I don't have a COMM cable, but it looks fairly simple so I'll try to make one. I'll post the config file once I've figured it out.

Click to expand...

Well, that can turn into a Science Project....
Meanwhile, just transcribe these settings:

ThrI->CntrlMode =

ThrO->DownRate =
ThrO->UpRate =
ThrO->PASRate =
ThrO->FastRate =
ThrO->FastThrsh =

PLim->MaxCurrent =
PLim->MaxPower =
PLim->AGain =
PLim->WGain =

PAS->PASMode =
PAS->StrtLevel =
PAS->ScaleFactr =
PAS->StrtThrsh =
PAS->StopThrsh =

Also - about what does the bike+rider weigh?
Are you using an Aux Input control?

ThrI->CntrlMode = Pass-thru

ThrO->DownRate = 4.00 Vlt/sec
ThrO->UpRate = 2.00 Vlt/sec
ThrO->PASRate = 1.00 Vlt/sec
ThrO->FastRate = 4.00
ThrO->FastThrsh = 3.00 Amps

*PLim->MaxCurrent = 25.0 Amps
PLim->MaxPower = 1000 Watts
*PLim->AGain = 60 Gain
*PLim->WGain = 5.0 Gain

*PAS->PASMode = Basic (Pwr)
*PAS->StrtLevel = + 300 Watts
PAS->ScaleFactr = 0.00 W/rpm
PAS->StrtThrsh = 0.30 sec/pole
PAS->StopThrsh = 0.19 sec/pole

I've *'d the ones that I've changed from default.

Thanks for the tip on finding the Settings Summary on the setup utility. It actually matches what I see on the CA.

I don't have an AUX switch.

I've had both 250LBS and about 350LBS total weight. The oscillations happen at both loads, and basically the same speeds, though I couldn't tell you if it was worse or better on either. I've got a LvH Bullitt.

The settings seem okay. 25A since it's the controller limit and I guess 1000W because you have a 36V battery (39V * 25A ~= 1000W)

Xoc said:

At any speed below 23kph, I don't have any issue with watt hunting. As I go faster the watts swing farther below and above 300 the faster I go.

Click to expand...

This seems to indicate you are running into a upper current/power limit - the PAS performance is okay but you are getting into trouble at high power levels. Since you are not getting a CA amps limit flag, you are below the 25A limit. The Wattage W flag is going to be in play pretty much all the time because the CA PAS system borrows the power PID and limits the MaxPower according to PAS assist and so always looks like watts limiting.

So - this is kind of odd. This is a very modest power DD motor. That said, according to the simulator you are running just about at the controller limit when the surging starts. So, a couple more questions to see if the power fluctuations are really coming from the CA:

1. What is your battery voltage?
2. Which model motor do you have (305 or 395rpm)?
3. Does the surging also happen when you run with throttle or only with PAS?
4. If it happens with throttle, does it still happen if you configure ThrI->CntrlMode = Bypass?
5. What happens (Limit Flags, surging) with either PAS or throttle if you set PLim->MaxPower = 2000 Watts?

Thank you for your help so far. I really appreciate it.

1. What is your battery voltage?
52V
2. Which model motor do you have (305 or 395rpm)?
305rpm
3. Does the surging also happen when you run with throttle or only with PAS?
At higher power levels (1000watts), no. When I set the max power of the system to 300 watts (or other similar levels) and use the throttle, the oscillations come back.
4. If it happens with throttle, does it still happen if you configure ThrI->CntrlMode = Bypass?
Haven't tried thist yet. I will this evening.
5. What happens (Limit Flags, surging) with either PAS or throttle if you set PLim->MaxPower = 2000 Watts?
Same state as before. W is the flag, surging happens at the same speeds.

I've also been in contact with Justin at Grin.

1. He asked me to check the Rshunt value. The sticker had 3.82 and the CA had 3.8 inputted. I've changed it and it's the same issue as before.
I noticed in the UUG that the typical range for the 6 mosfet Infineon controllers is 5.0-6.0. Mine seems quite outside that range.

2. He also had be try a wgain of 15 and 12. The higher the wgain causes the frequency of the oscillations to increase. A wgain of 5 caused the an oscillation to occur over about 2 seconds, whereas a wgain of 15 caused multiple oscillation over a second. This made the apparent magnitude of the oscillation much greater (my upper body was moving back and forth).

3. I've tried bumping the power up to 550 watts. The oscillations started to become apparent at 29kph instead of 23kph.

I've tried the bypass mode. I didn't get any surging. But it was using full power.

I also noticed on the diag screen that it was pulling 24amps at full throttle. That would seem to indicate that the rshunt value is correct.

Is it normal for this surging to occur? The lower the watts, the lower the speeds it happens at? I've ridden other ebikes from different manufactures (Ohm-Bionx, Specialized, Rad-Power) and never felt this surging on any power level at any speed.

Xoc said:

I've tried the bypass mode. I didn't get any surging. But it was using full power.

Click to expand...

Bypass mode prevents the CA from affecting the controller throttle in anyway - it's essentially like connecting the throttle directly to the controller. This was a simple test to verify that the issues were not arising in the controller or battery.

Xoc said:

I also noticed on the diag screen that it was pulling 24amps at full throttle. That would seem to indicate that the rshunt value is correct.

Click to expand...

Yep.

Xoc said:

Is it normal for this surging to occur?

Click to expand...

No. CAs normally run smooth as silk using the WGain = 12 setting that Justin related. This is typically a mindlessly simple setup. There is something very strange going on here.

Xoc said:

The lower the watts, the lower the speeds it happens at?

Click to expand...

Non-specific. Here I'm assuming you are talking about the PAS->StartLevel not the PLim->MaxPower, correct?

The behavior seems to be indicating that power level limiting is misbehaving. Lowering the StartLevel simply changes the power level (and hence speed) at which the limiting cuts in and the misbehavior occurs.

Xoc said:

3. Does the surging also happen when you run with throttle or only with PAS?
At higher power levels (1000watts), no. When I set the max power of the system to 300 watts (or other similar levels) and use the throttle, the oscillations come back.

Click to expand...

Again, I don't understand what power levels you are talking about: CA power display, MaxPower, or StartLevel?

In any case, because this is happening with either throttle or PAS it seems that this is not the PAS code specifically but rather the common PAS/Throttle power-limiting code. There is a minor bug in that bit of code, but it shouldn't have this effect and I believe this is the only instance we've seen of this behavior.

The apparent symptoms might indicate an incorrect RShunt setting or a much too large WGain setting. Either of these would cause power PID overcorrection (cyclic overshoot/undershoot) and the surging you experience - but both settings seem okay. Odd.

In the absence of any new ideas, I can only suggest that there may be some unexplained corruption of you Settings in EEPROM. I would re-flash to defaults and then manually reconfigure without using any previously saved setup files. This is the standard Hail Mary maneuver when all else fails, but...

Post back with results. If this is unsuccessful I'll talk with Justin about other options to get this resolved.

Sorry for not being more specific:

teklektik said:

Xoc said:

The lower the watts, the lower the speeds it happens at?

Click to expand...

Non-specific. Here I'm assuming you are talking about the PAS->StartLevel not the PLim->MaxPower, correct?

Click to expand...

Yes, I'm talking about the PAS->StartLevel.

teklektik said:

Xoc said:

3. Does the surging also happen when you run with throttle or only with PAS?
At higher power levels (1000watts), no. When I set the max power of the system to 300 watts (or other similar levels) and use the throttle, the oscillations come back.

Click to expand...

Again, I don't understand what power levels you are talking about: CA power display, MaxPower, or StartLevel?

Click to expand...

When I had the PLim->MaxPower set to 1000 watts I didn't feel any surging. When I set the PLim->MaxPower to 300 watts and used the throttle I did feel the surging as normal.

teklektik said:

In the absence of any new ideas, I can only suggest that there may be some unexplained corruption of you Settings in EEPROM. I would re-flash to defaults and then manually reconfigure without using any previously saved setup files. This is the standard Hail Mary maneuver when all else fails, but...

Post back with results. If this is unsuccessful I'll talk with Justin about other options to get this resolved.

Click to expand...

I'll try that. Do I need to use a computer with a TTL cable or is doing the factory reset from within the CA3 menu enough?
I don't have a TTL cable, but I am in Vancouver and can stop by the Grin Office. I'll do the factory reset on the CA3 menu to start with anyway.

Thanks for your help!

Ok - thanks for the clarifications.

Xoc said:

Do I need to use a computer with a TTL cable or is doing the factory reset from within the CA3 menu enough?
I don't have a TTL cable, but I am in Vancouver and can stop by the Grin Office. I'll do the factory reset on the CA3 menu to start with anyway.

Click to expand...

Depends. Try the CA-only reset, but re-flash is absolute.

I recommend that you get a cable. New software is always coming along... (There are new beta releases on the horizon!)

Turns out that it wasn't the CA, but instead an incompatibility between the controller (25A Grinfineon) and motor (TDCM_305_IGH) I got from them. Justin was saying that everything worked fine in sensor less mode, but not in sensor mode. He'd be able to explain in more detail I'm sure. He also said it was the first time they'd come across it.

Anyway, they got me sorted out. Thanks Justin and Steve at Grin.

And thanks teklektik for all the help. I also picked up a TTL cable.

Xoc said:

Turns out that it wasn't the CA, but instead an incompatibility between the controller (25A Grinfineon) and motor (TDCM_305_IGH)....
Justin was saying that everything worked fine in sensor less mode, but not in sensor mode. ...

Click to expand...

Whoa - that's amazingly strange!
The sensorless vs sensored mode does explain why the throttle Bypass mode test seemed to indicate the CA - the motor was in sensorless mode at your 'full power' test condition.

I'll be interested to hear the details from JLE - certainly a weird one.

Thanks for posting back. Glad it got cleared up.