Page 183 of 183

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 01 2020 3:08am
by ebike11
Hi..I have another odd problem, just for fun, I connected my 80A 3000W controller to my qs205 and cycle analyst.
I set me Power limits in the CA setup menu to 80A and left the Watts at factory 9900W
However while riding, I noticed my Amps went well over the 80A, even reaching 120 or so.
Why would it do this if its supposed to cut off current? thx

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 3:25am
by justin_le
madin88 wrote:
May 23 2020 3:38am
I have installed a new torque PAS with Sempu T4.3 sensor which came as 1-wire type with 48 poles, and noticed the CA does not allow to set more than 32 poles. Any numbers higher than that are not applied.
Why it cannot be set higher than that? Is there a reason for?
There is a reason which is related to limited memory in the device. The way that the CA3 calculates the pedal RPM and the human power is that it needs to keep a running history of the timing of each pedal sense event and the averge torque that it measured for that portion of the rotation, and so when you have a 48 pole sensor that means 48 separate time measurements (3 bytes each) and 48 separate torque values (2 bytes each) = lots of bytes for a small 8bit micro. But certainly there could be workarounds in a next release. At the time that this was initially coded the max PAS sensor available on the market was 24 poles, so 32 seemed like a decent margin.
With about half the Nm/V setting the power it reads seems to be quite accurate but still no overall nice solution.
Agreed. We will aim to support inputting higher PAS pole capabilties in future releases.
One issue i noticed is when i stop to apply fore on the pedals (only spinning them freely), with human power being lower as the "start value" which is set up, the motor does not stop.
Lets say you have set 25W human power, the motor will begin to assist if it goes past this value, but it will not stop if it get lower than this when reducing force and spinning the cranks freely it reads around 5-10W and still applying power to the motor for lets say 3-5 sec before it goes out.
Shouldn't it stop instant?
It's not instant in this case because it's relying on the constant wattage feedback loop to bring the power back down. If it went down instantly, then if you were pedaling just around the start value watts, every time it dipped below that the throttle output would drop to min and then take time to ramp back up to a value that does anything causing a very annoying lag. That's actually how it was originally implemented in the CA3 Prelim firmwares and we changed it so that the feedback loop was always active even if commanded watts was zero. If you actually stop pedaling the power will cutout immediately as it should, but if your human power goes down below the threshold while you are still pedaling, it's simply commanding 0 watts from the feedback controller and it will ramp down based on WGain.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 3:28am
by justin_le
ebike11 wrote:
Jun 01 2020 3:08am
Why would it do this if its supposed to cut off current? thx
It will only have the means to scale back current draw if you have the throttle output signal of the CA wired up to the throttle input line of your motor controller. Can you go to the diagnostics screen of the CA3 (press the left button once) and tell us what you see the throttle output voltage signal do when your current is exceeding the 80 amps threshold?

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 3:35am
by justin_le
mrbill wrote:
Apr 23 2020 1:16pm
To disable my throttle while not pedaling I set the "Max No-Pedal Speed" under "Speed Limit Settings" to 0 (zero). This appears to do exactly what I want except when I'm commanding varying levels of regenerative "motor" braking.
When using the motor to brake, I get full motor braking only. I checked the Out stat on the CA3 Diagnostics screen and discovered that it is pegged to 0.0 volts out no matter the voltage of In. I don't believe it is reasonable to require pedaling to make use of the throttle as a braking force control.
Thanks Bill for bringing this to our attention. It wasn't a deliberate descision to do this, rather the "max no-pedal speed" throttle function was implemented well before we did the proportional throttle regen. We didn't think to make an exception for ignoring the throttle while ebrake was applied so that the proportional regen was still functional here in this particular usage case. This detail wasn't on our radar and we'll have it addressed in an upcoming CA release!

You are correct that the spirit of this setting was to disable throttle from powering the bike if you weren't pedaling, but it doesn't make sense that it would restrict the regen functionality when you aren't pedaling too.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 3:39am
by ebike11
justin_le wrote:
Jun 03 2020 3:28am
ebike11 wrote:
Jun 01 2020 3:08am
Why would it do this if its supposed to cut off current? thx
It will only have the means to scale back current draw if you have the throttle output signal of the CA wired up to the throttle input line of your motor controller. Can you go to the diagnostics screen of the CA3 (press the left button once) and tell us what you see the throttle output voltage signal do when your current is exceeding the 80 amps threshold?
Ah my mistake....my throttle is directly plugged into the controller. So if the CA is all connected but the throttle is going directly to the controller..does this mean that the motor will see all the full power of the battery/controller with no power limits?

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 8:19am
by DanGT86
Yes. The CA3 has no means to limit anything unless it is between the throttle and controller.

With your throttle wired directly to the controller your CA is just a fancy information gauge.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 9:57am
by mrbill
justin_le wrote:
Jun 03 2020 3:35am
mrbill wrote:
Apr 23 2020 1:16pm
To disable my throttle while not pedaling I set the "Max No-Pedal Speed" under "Speed Limit Settings" to 0 (zero). This appears to do exactly what I want except when I'm commanding varying levels of regenerative "motor" braking.
When using the motor to brake, I get full motor braking only. I checked the Out stat on the CA3 Diagnostics screen and discovered that it is pegged to 0.0 volts out no matter the voltage of In. I don't believe it is reasonable to require pedaling to make use of the throttle as a braking force control.
Thanks Bill for bringing this to our attention. It wasn't a deliberate descision to do this, rather the "max no-pedal speed" throttle function was implemented well before we did the proportional throttle regen. We didn't think to make an exception for ignoring the throttle while ebrake was applied so that the proportional regen was still functional here in this particular usage case. This detail wasn't on our radar and we'll have it addressed in an upcoming CA release!

You are correct that the spirit of this setting was to disable throttle from powering the bike if you weren't pedaling, but it doesn't make sense that it would restrict the regen functionality when you aren't pedaling too.
Thanks, Justin, for acknowledging this particular and peculiar usage case.

The local public lands managers in my area are investigating the possibility of allowing on some of their trails, perhaps on a trial basis, "Class 1" e-bikes. In preparation of this I'm dusting off my Class 1 or "Legal" e-bike settings. In the state of California Class 1 e-bikes are required to disable any throttle unless the user is pedaling. But, while riding with those settings I noticed that I lost my variable regeneration: it was full "on" or nothing.

I'm already running an older firmware version (3.12) due to MaxSpeed glitches discovered in the latest release (3.14). Would it be expecting too much for the behavior of the throttle used as motor brake (as well as the MaxSpeed glitches) to be addressed in the next release?

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 10:23am
by john61ct
Justin,

The fact you participate directly in the forums like this

adds so much value, I just wanted to express gratitude on behalf of the community

and to wish you maximum prosperity as we work our way toward whatever "new normal" is coming

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 5:23pm
by ebike11
john61ct wrote:
Jun 03 2020 10:23am
Justin,

The fact you participate directly in the forums like this

adds so much value, I just wanted to express gratitude on behalf of the community

and to wish you maximum prosperity as we work our way toward whatever "new normal" is coming
Yes..its great!

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 5:24pm
by ebike11
DanGT86 wrote:
Jun 03 2020 8:19am
Yes. The CA3 has no means to limit anything unless it is between the throttle and controller.

With your throttle wired directly to the controller your CA is just a fancy information gauge.
Would that mean then that the motor/contoller/battery is seeing full power with no ca power limits when the throttle is directly to the controller?

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 5:39pm
by NCC1941
ebike11 wrote:
Jun 03 2020 5:24pm
Would that mean then that the motor/contoller/battery is seeing full power with no ca power limits when the throttle is directly to the controller?
Yes. The CA3 imposes its limits by altering the throttle signal from CA to controller. It can't do that if it's not sitting between the throttle and the controller.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 03 2020 6:50pm
by ebike11
NCC1941 wrote:
Jun 03 2020 5:39pm
ebike11 wrote:
Jun 03 2020 5:24pm
Would that mean then that the motor/contoller/battery is seeing full power with no ca power limits when the throttle is directly to the controller?
Yes. The CA3 imposes its limits by altering the throttle signal from CA to controller. It can't do that if it's not sitting between the throttle and the controller.
Ok thanks

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 15 2020 8:21am
by chilau2000
Hi, is there a way for the CA to only enable throttle when the vehicle is moving.
I'm thinking of upgrading my Dualtron with a Phaserunner and CA and I would like to keep the enable throttle only when moving feature for safety.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 15 2020 8:41am
by amberwolf
Should be able to do that here:
https://www.ebikes.ca/product-info/cycl ... pSpeedLims


Speed Limit Setup Menu

Speed Limit Settings, preview line shows max speed, but there are also a minimum start speed limit and max throttle only speed setpointsThere are both minimum and maximum speed limit settings available, and the 2nd line preview shows the speed limit set point present in the current preset. If you don't want to use the Cycle Analyst to limit the vehicle speed in any capacity, then this can be left at the default setting of 199 kph.

Click for Explanation of Speed Limit Settings

The CA can be configured with a speed limit to cut motor power when the bike exceeds that limit. This speed limit is implemented as a PID controller and so there are three gain terms that may need to be tweaked to achieve the desired responsiveness when the bike hits the limit point. Surging or oscillation at the speed limit setpoint or power cutouts on hard acceleration (powerful bikes) are indications that speed gain adjustments are needed.

[ SLim->Max Speed ]*
Upper speed limit. The CA rolls back throttle output voltage whenever speed exceeds this setting.
OEM->AbsMaxSpeed determines the maximum value for this setting.

[ SLim->Strt Speed ]*
The minimum starting speed that must be reached before the CA will allow power application. This is useful for RC drives or systems with sensorless controllers that do not work well from a dead start. The rider must pedal the bike up to StrtSpeed before power is applied.


[ SLim->MxNoPdlSpd ]*
The maximum speed at which the throttle will apply power without pedaling in any PAS mode. This feature supports European pedalec legislation which requires pedaling the bike for motor power, but which provides exemption below a certain speed (e.g. 6kph) in which case use of a throttle without pedaling is permitted. Applies only if a PAS device is installed.

[ SLim->Regen Lmtg ]
Chooser to select if Speed Limiting should use regenerative braking to help control the maximum speed.

Disabled: Speed is limited exclusively by reducing power. Above the speed limit the motor power is reduced to zero, but regenerative braking is not applied allowing the rider to continue to coast or pedal above this speed limit. [Default]

Enabled: Regenerative braking is used to enforce the speed limit setpoint. This means that regen may engage automatically when descending hills even without the rider applying the ebrakes. Proportional throttle regen (EBk->PropRegen) need not be enabled. This feature requires a controller with 0.0 - 0.8V throttle regen function (e.g. Grinfineon or Phaserunner).

[ SLim->IntSGain ]
Integral (I) feedback gain for the speed PID control loop. Lower values give smoother control and less likelihood of hunting, but can increase the time it takes for the speed limit to stabilize. This adjusts the correction for accumulated 'past' speed error.

[ SLim->PSGain ]
Proportional (P) feedback term for the speed PID control loop and displayed in terms of Volts / (mph or kph). If set to 0.5V/kph, the throttle output will immediately drop by 0.5V for each km/hr the bike exceeds the set speed limit. This adjusts the correction for the 'present' speed error.

[ SLim->DSGain ]
Differential (D) feedback term for the speed PID control loop. This is used to dampen speed oscillations by determining the ability of the CA to scale back power in anticipation of breaching the speed limit due to acceleration towards that limit. Values in the 100-300 range seem to work well, but lower values are often required for vehicles capable of hard acceleration. This adjusts the correction for 'future' anticipated speed errors.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 19 2020 10:30am
by ebike11
Hi guys..
Im planning on wiring an external shunt to my CA.
On the JST plug it says red wire goes to controller positive. I have a Sabvoton controller. Do I connect it to the main controller terminal positive screw or the controller turn on wire?
Thanks

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 19 2020 11:35pm
by DanGT86
The red V+ needs to be connected to your battery positive wire. The main controller positive terminal Is a safe bet. If the controller turn on wire comes straight from the battery then I suppose that would work as well.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 20 2020 11:23am
by ebike11
DanGT86 wrote:
Jun 19 2020 11:35pm
The red V+ needs to be connected to your battery positive wire. The main controller positive terminal Is a safe bet. If the controller turn on wire comes straight from the battery then I suppose that would work as well.
ok thanks
I plan to wire up the jst plug for high current external shunt, not the one that Grin sells. I have a Grin one but its rated lower

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 23 2020 9:24pm
by ebike11
I read the CA manual for wiring. Id like to do pass thru wiring with the grin shunt and have the throttle wire ONLY going from the shunt to the controller green wire. But at the handlebars, do I plug the throttle directly into the CA unit controller connector with ALL 3 wires as normal or only red and black WITHOUT connecting the green?
Thx

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 29 2020 1:21am
by Scotty T
I've moved my CA to a new bike and played around with settings, turns out I was limiting amps and watts which was causing the throttle issue of surging so removing those limits fixed that. Also throttle out voltage was too high for the controller so I dropped it a little and all good now. Have got three presets for PAS level, limited to 1200w 25km/h ("legal") 2000w 35km/h (it only does ~33 tops) and 750w/20km for slow cruising, using the digital aux to control.

Finally got it all worked out using the basic PAS with watt limits which works very similar to the off the shelf Bafang PAS levels pre the CA being fitted. I was quite surprised to see peak power got just over 4000w at one stage doing a full throttle test.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jun 29 2020 9:05am
by DanGT86
ebike11 wrote:
Jun 23 2020 9:24pm
I read the CA manual for wiring. Id like to do pass thru wiring with the grin shunt and have the throttle wire ONLY going from the shunt to the controller green wire. But at the handlebars, do I plug the throttle directly into the CA unit controller connector with ALL 3 wires as normal or only red and black WITHOUT connecting the green?
Thx
The throttle connection from the throttle to the CA will be all three wires. The CA3 6 pin connector will have a throttle out wire that goes to the controller throttle input plug. This is usually the green wire on most controllers. The CA shunt from grin that has the 6 pin in and yellow and green wires out will use the green wire out to go to the controller throttle signal in wire. Again usually this wire is green on most controllers I have seen.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jul 01 2020 1:39am
by Scotty T
Some obeservations and questions...

If I limit amps or watts with power limits I get surging on the throttle, it seems to not like power limts in my setup.

PAS is working really well, just having trouble on a different crankset getting the magnet ring to line up. Running it start stop for half an hour didn't even get the controller warmer than 30 degrees.

I have a 25 AMP Grinfineon with 48 volt 11 Ah battery running a Bafaing CST 350w, and with no limits full throttle as mentioned it reads up to 4125 watts, and up to 97 AMPS. How is that possible or am I misunderstaning that information?

Re: Cycle Analyst V3 preview and first beta release

Posted: Jul 01 2020 9:29am
by DanGT86
Sounds like you might not have your shunt value set properly in the CA setup menu. If you are running a grinfineon controller then the shunt value should be written on the bottom of the controller. It will be some value expressed in milliohms. In the CA calibration menu there will be a shunt value setting. Make sure these numbers match. That is likely why you are seeing 4125 watts out of a 1200 watt setup.

As for the surging, when the Amp gain and Watt gain values are too high there will be surging when using power or amp throttle mode. Having the wrong shunt value will drastically amplify the problem. Once you have the above mentioned shunt issue sorted such that your CA is reading correct values you can tune these gain values. They are in the power limit menu. For instance if you are using amp throttle limiting then set your amps limit to the max you expect to use plus maybe a tiny bit of overhead. For your 25 amp controller 25-27 amps is a good guess. Set the Amps limit GAIN until there is some surging/oscillation on power when accelerating. Once you get the oscillation reduce that gain value by about 20-30% and try again. Keep going until the surging is gone. Ideally you want the highest gain value that doesn't cause surging. The cycle analyst should be super smooth running watt or amp throttle with only a 25amp controller especially one from grin.

Its well worth tuning in because the amp or watt throttle is such a nice mode to ride compared to speed based throttle.

Based on past experience you should make sure your fast throttle up setting is low or disable fast throttle by zeroing the fast throttle amp threshold while you are tuning the amp/watt gain settings. You can add it back when you get it dialed in but its just one more thing to confuse the process when trying to fine tune the jerky acceleration out.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jul 01 2020 6:47pm
by Scotty T
Thanks for the tips @DanGT86, I'll check that shunt value, I think it may have changed when I upgraded firmware and overwrote a setup.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jul 01 2020 7:57pm
by DanGT86
I think they default to 1mohm when you reset them. My 25 amp controller shunt is 4.66 mohms. That is very likely what happened.

Re: Cycle Analyst V3 preview and first beta release

Posted: Jul 01 2020 9:47pm
by Scotty T
DanGT86 wrote:
Jul 01 2020 7:57pm
I think they default to 1mohm when you reset them. My 25 amp controller shunt is 4.66 mohms. That is very likely what happened.
Thanks so much for this info, I have now set things up and it's a whole different bike, I had to drop all my limits as it took off like a rocket, I'd forgotten how punchy it was because of these issues after my firmware change and subsequent config overwrite.

I've found a very faded 4.19 written on the controller label that I'm noting here and is also now in my saved config, which I will also back up to the cloud :)

You are right, the limiting for watts and amps is great when working properly. This has made take off smoother, and now my PAS sensor which appeared to be playing up if it was not very close to the magnet is operating like it used to again, the bad shunt value was messing with everything. I still need to speed limit in my cruising mode as I have only got a single speed on this cruiser and I quickly run out of RPM in the legs.