Cycle Analyst V3 preview and first beta release

mrbill · Jan 9, 2015

justin_le said:
This behavior should be somewhat better in the P7, but I know what to do to make it work perfectly.

I went out riding this afternoon to test a bunch of stuff on my bike: new front wheel, reversed usual spoke orientation on rear wheel non-drive side to better handle disk brake torque under hard braking, and the CAV3, P7 firmware. Up and down lots of hills.

http://www.trimbleoutdoors.com/ViewTrip/2874325

The surging or kick of motor power upon resumption of PAS-mediated throttle is less than it was with P6. I still get kicked occasionally, but most of the time the initial application of power is kick-free. I did change some of the parameters: upRamp -> 2.5v/sec (lower was too sluggish for when I press the throttle to get power NOW!), and fastRampCurrent I increased from 0 to 1.7 Amps. With P6 setting this to anything but zero caused more kick, but it seems to work as expected in P7.

mrbill said:
mrbill said:

It feels like the CAV3 is ignoring the prior assist factor (set by aux pot) or throttle ramping when power is re-applied after a 3-second break in pedaling or after a 3-second application of the e-brake.

Click to expand...

What's actually happening is that the power feedback loop is still attempting to output a higher and higher signal if you pedal during the ebrakes (which separately forces the output low) and then on releasing the ebrakes there is a built in "wind-up" in the feedback term. The up-and down ramp rates are always being followed, but you have them set so high that it's not having much effect. I think that's why I haven't noticed this since I've typically had a much more subdued up-ramp, and with hub motors I don't think an overshoot is felt with quite as much kick.

Your explanation makes sense. I still do get an occasional kick, and it's hard for me to remember all the conditions that led up to the misbehavior. I think it occurs more often (but not always!) after running the motor for a while in PAS mode, then taking a short pedaling break, then resuming pedaling with a kick.

(The other rough edge, a minor nuisance, is the wide hysteresis of the speed limit when approached from above the max speed. If decelerating through the max speed, resumption of power is delayed unless one is decelerating very slowly. E.g. if max speed=20mph and bicycle decelerates quickly from 25 mph through 20 mph as when a downhill is following by an uphill, power might not be applied until speed is down to 15 or 16 mph and momentum lost. PSGain=0.85v/mph, IntSGain=200, and DSGain=200. The parameters were tweaked for approaching the speed limit from below.)

Click to expand...

Ha yes, this is part of the asymmetric control problem (CA can apply power to speed up the bike, but has no ability to control the deceleration rates) and I'm glad you reminded me of it since there may be some room for algorithm changes to better cope with that. If you do increase the DSGain to a higher value though you should find that the situation is improved, as then it will sense your rapid deceleration and attempt to increase the throttle output accordingly.

OK, I will try some higher values for DSGain and see if that improves the behavior of approaching speed limit from above without causing too much oscillation when approaching speed limit from below.

It would be nice if PAS-mediated throttle signal observed the throttle ramping parameters. I'm using up-ramp of 3.5v/sec, fast-ramp of 5v/sec and down-ramp of 40v/sec.

Click to expand...

Try in this case having your standard up ramp at 1V/sec, and if you find that it is switching from the fast to the standard up ramp too soon then either increase the fast ramp current threshold from 3A to more like 5 or 6A, OR reduce the fast ramp rate a bit so that the current draw of the motor as it is ramping up to an RPM that engages the drive chain is less than the current threshold. You want the throttle to quickly reach the voltage where the motor RPM is now matched to the drivechain RPM, and then take a slower path from that point on. 3.5 V/sec may not seem too fast, but given that at a given RPM the controller goes from no current to max current over about a 0.3-0.5V span (depending mostly on the motor's winding resistance), it means that the effective time to go from zero to full power is on the order of 0.1 seconds. That's not a whole lot of time for the CA to respond in.

On today's ride I realized my ramping was a bit too fast. I tried 1v/sec, but this resulted in too sluggish throttle behavior (and PAS behavior). I gradually increased the upRamp, 1.7v/sec, 2.2v/sec, and settled on 2.5v/sec. I also turned on the fastRamp by setting current to 1.7A, just under the full-throttle free-spin current of my motor. Since I updated to P7 I haven't tried higher fastRampCurrent values that would only be attained when the motor is loaded, although with P6, higher fastRampCurrents gave harder kicks. My goal is for the throttle (manual or PAS) to feel instantaneous or nearly so, but gradual enough not to overshoot the other limits or to give me a kick.

Let me know if subjectively things improve with the suggestions posted above and I'll work on addressing some of the key algorithm issues to optimize things further.

Thanks for looking into these two issues. I'll experiment with DSGain this weekend.

Oh, and I really like the new Aux Adjust display that appears when I adjust the Aux Pot. Nice to see the actual power level I'm dialing in. One improvement would be to have that display persist for a couple seconds after the Aux Pot value stops changing. It disappears too fast!

justin_le · Jan 11, 2015

mrbill said:
On today's ride I realized my ramping was a bit too fast. I tried 1v/sec, but this resulted in too sluggish throttle behavior (and PAS behavior). I gradually increased the upRamp, 1.7v/sec, 2.2v/sec, and settled on 2.5v/sec. I also turned on the fastRamp by setting current to 1.7A, just under the full-throttle free-spin current of my motor.

Ah this is a misundertanding. You want the setting to be higher than the unloaded motor speed, and not just while it is free spinning but more specifically while it is in the process of accelerating up to the free spin RPM at the fast ramp rate. If the unloaded draw is 1.7A, I suspect while accelerating up to that speed the load from the motor's own inertia probably results in more like a 3-4A current draw, and whatever it is you want the fast ramp threshold to be higher than that. With a direct drive hub motor, you can have a very low threshold because there is no current draw from the motor until the throttle is at the unload RPM point (I set mine at 0.5A for instance), but with a geared or mid-drive motor you'll need to be higher than this for or the CA will go prematurely to the slow ramp rate.

I think this is why you've experienced a 'sluggish' throttle response with the recommended settings. To help dial it in, set you standard ramp rate really low (like <0.1 V/sec) and then watch the diagnostics screen and you'll be able to see quite clearly when it has switched from the fast to the low rate, and set your threshold current so this only happen when the motor has caught up to the pedals and is being loaded by the bike, not before.

Let me know if subjectively things improve with the suggestions posted above and I'll work on addressing some of the key algorithm issues to optimize things further.

Click to expand...

Thanks for looking into these two issues. I'll experiment with DSGain this weekend.

OK, another point I meant to mention is that the default value of 50 for the Watts gain is really too high, and we've found on most PAS assist systems that the ideal gain setting is more in the 12-20 range. Our default now with the CycleStoker drive is a gain of 15 here, so reducing that should help with the overshoot too.

I've attached a Prelim8 version of the code that should address instances of power surge after releasing ebrakes while pedaling in a PAS mode, it also increases the debounce time from the magnet sensor which should help reduce spurious speedometer readings from misaligned speedo sensors, and I've increased how long the Aux Adjust display lingers a bit as per your suggestion below.

Oh, and I really like the new Aux Adjust display that appears when I adjust the Aux Pot. Nice to see the actual power level I'm dialing in. One improvement would be to have that display persist for a couple seconds after the Aux Pot value stops changing. It disappears too fast!

mrbill · Jan 11, 2015

Thanks for the quick turnaround on P8 firmware.

justin_le said:
Ha yes, this is part of the asymmetric control problem (CA can apply power to speed up the bike, but has no ability to control the deceleration rates) and I'm glad you reminded me of it since there may be some room for algorithm changes to better cope with that. If you do increase the DSGain to a higher value though you should find that the situation is improved, as then it will sense your rapid deceleration and attempt to increase the throttle output accordingly.

I rode again today using P7 and tried setting DSGain to 500, but the problem of undershooting the speed limit after decelerating from a higher speed persists about the same as before. Moreover, when approaching the speed limit from below, there is now mildly annoying "pumping the gas pedal" oscillation that damps out after about 30 seconds. With DSGain at 200 there is only one overshoot to speed+1mph before settling back to the max speed. I suspect the speed limit undershoot problem requires some developer intervention.

justin_le said:
mrbill said:

On today's ride I realized my ramping was a bit too fast. I tried 1v/sec, but this resulted in too sluggish throttle behavior (and PAS behavior). I gradually increased the upRamp, 1.7v/sec, 2.2v/sec, and settled on 2.5v/sec. I also turned on the fastRamp by setting current to 1.7A, just under the full-throttle free-spin current of my motor.

Click to expand...

Ah this is a misundertanding. You want the setting to be higher than the unloaded motor speed, and not just while it is free spinning but more specifically while it is in the process of accelerating up to the free spin RPM at the fast ramp rate. If the unloaded draw is 1.7A, I suspect while accelerating up to that speed the load from the motor's own inertia probably results in more like a 3-4A current draw, and whatever it is you want the fast ramp threshold to be higher than that. With a direct drive hub motor, you can have a very low threshold because there is no current draw from the motor until the throttle is at the unload RPM point (I set mine at 0.5A for instance), but with a geared or mid-drive motor you'll need to be higher than this for or the CA will go prematurely to the slow ramp rate.

I think this is why you've experienced a 'sluggish' throttle response with the recommended settings. To help dial it in, set you standard ramp rate really low (like <0.1 V/sec) and then watch the diagnostics screen and you'll be able to see quite clearly when it has switched from the fast to the low rate, and set your threshold current so this only happen when the motor has caught up to the pedals and is being loaded by the bike, not before.

Ah, I see. Never thought to measure the peak current while the motor is spooling up unloaded. The full-throttle free-spin current is about 2 to 2.2A. I initially set the fastRampCurrent to 2 Amps, but 1.7A gave less kick. I'll attempt to measure the peak current draw using a peak-reading clamp meter on the battery cable. Shall I initially set fast ramp current to something just above this peak? E.g. if peak current is 4.5A, set fast ramp current to 5.0A? Or would it be better to go a bit higher to, say, 6.0A? How much margin would you recommend?

Next time I go out I'll test P8 after setting the throttleOut parameters per your advice.

justin_le · Jan 11, 2015

mrbill said:
I rode again today using P7 and tried setting DSGain to 500, but the problem of undershooting the speed limit after decelerating from a higher speed persists about the same as before. Moreover, when approaching the speed limit from below, there is now mildly annoying "pumping the gas pedal" oscillation that damps out after about 30 seconds. With DSGain at 200 there is only one overshoot to speed+1mph before settling back to the max speed. I suspect the speed limit undershoot problem requires some developer intervention.

OK great, thanks for trying that out and reporting here and yes I'll consider approaches to address this asymmetry down the road. Easiest would be to have a separate set of PID feedback control parameters for when you are above the speed limit coming down vs. below it coming up, but presenting people with a total of 6 variables to adjust to dial in their speed control would be a bit much; 3 variables is already a little daunting to tweak.

mrbill said:
The full-throttle free-spin current is about 2 to 2.2A. I initially set the fastRampCurrent to 2 Amps, but 1.7A gave less kick.

The reason for that is that it had already switched to using your standard ramp rate rather than the fast ramp rate. For a normal controller that has a voltage throttle, your standard up-ramp rate can be like 0.5 V/sec without any apparent loss of "responsiveness", so long as your fast ramp gets you up to the point fairly quick.

I'll attempt to measure the peak current draw using a peak-reading clamp meter on the battery cable. Shall I initially set fast ramp current to something just above this peak? E.g. if peak current is 4.5A, set fast ramp current to 5.0A?

Yes, exactly, but just be mindful that current peak will be a function of the fast ramp rate. So if you ramp the motor up at 2V/sec it will draw less current accelerating than if you ramp it up at 4V/sec. If you take a couple peak readings of your startup amperage draw at different fast ramp rates that would be quite interesting to see here.

Given that the useful input throttle range of most controllers is a 2V span (from ~1.5V to ~3.5V), and under most circumstances you will hit the motor power about half way up, a 4V/sec fast ramp will typically get you to the active throttle region in about 1/4 second, and in a very worst case about 1/2 second. If you go really high on the fast up ramp, like 10 V/sec or more, then there is quite a bit of overshoot with many controllers as the CA's output is rising faster than the controller is responding to the signal. Most controllers have their own input filters on how fast the throttle can change, so if the CA goes up faster than this you'll certainly get a 'kick' once the controller's output catches up to the fast rising input signal.

Or would it be better to go a bit higher to, say, 6.0A? How much margin would you recommend?

Unfortunately I don't have enough empirical stats to make an educated suggestion on the needed margin, but in principle you shouldn't need very much. Even if momentarily the current spiked above the threshold, the throttle would only briefly be rising at the standard rate, once the measured current was down again it would resume at the fast rate.

Next time I go out I'll test P8 after setting the throttleOut parameters per your advice.

Great, and don't forget to try setting the WGain lower too. I just finished a 10km night ride using both Torque and PAS assist modes with these settings and felt it really hit the right mark.

-Justin

mrbill · Jan 12, 2015

justin_le said:
mrbill said:

I rode again today using P7 and tried setting DSGain to 500, but the problem of undershooting the speed limit after decelerating from a higher speed persists about the same as before. Moreover, when approaching the speed limit from below, there is now mildly annoying "pumping the gas pedal" oscillation that damps out after about 30 seconds. With DSGain at 200 there is only one overshoot to speed+1mph before settling back to the max speed. I suspect the speed limit undershoot problem requires some developer intervention.

Click to expand...

OK great, thanks for trying that out and reporting here and yes I'll consider approaches to address this asymmetry down the road. Easiest would be to have a separate set of PID feedback control parameters for when you are above the speed limit coming down vs. below it coming up, but presenting people with a total of 6 variables to adjust to dial in their speed control would be a bit much; 3 variables is already a little daunting to tweak.

I wouldn't mind having ability to tweak speed deceleration parameters. And, if defaults can be established that work across a wide range of cases, then most users oughtn't complain.

I'll attempt to measure the peak current draw using a peak-reading clamp meter on the battery cable. Shall I initially set fast ramp current to something just above this peak? E.g. if peak current is 4.5A, set fast ramp current to 5.0A?

Click to expand...

Yes, exactly, but just be mindful that current peak will be a function of the fast ramp rate. So if you ramp the motor up at 2V/sec it will draw less current accelerating than if you ramp it up at 4V/sec. If you take a couple peak readings of your startup amperage draw at different fast ramp rates that would be quite interesting to see here.

Given that the useful input throttle range of most controllers is a 2V span (from ~1.5V to ~3.5V), and under most circumstances you will hit the motor power about half way up, a 4V/sec fast ramp will typically get you to the active throttle region in about 1/4 second, and in a very worst case about 1/2 second. If you go really high on the fast up ramp, like 10 V/sec or more, then there is quite a bit of overshoot with many controllers as the CA's output is rising faster than the controller is responding to the signal. Most controllers have their own input filters on how fast the throttle can change, so if the CA goes up faster than this you'll certainly get a 'kick' once the controller's output catches up to the fast rising input signal.

I'll try to gather some data this afternoon and report back.

Next time I go out I'll test P8 after setting the throttleOut parameters per your advice.

Click to expand...

Great, and don't forget to try setting the WGain lower too. I just finished a 10km night ride using both Torque and PAS assist modes with these settings and felt it really hit the right mark.

So many things to try! OK. Maybe I'll start with WGain of 5 or 10, a change that I'll notice. Then work my way back up to a pleasing value.

Shall I change the AGain value similarly, or does only WGain apply to the PAS level (since PAS level adjusts the power limit)?

While I have your attention...

Another quibble to address in P9:
I have my MaxPower limit set to 750 watts on my "Legal" preset and 1000 watts on my other presets. AutoPAS power level is set to 1000 watts. When I am using the "Legal" preset, the Aux Adjust for PAS Level varies from 0 to 1000 watts. It would be nice if that upper bound were the lesser of AutoPASPower and MaxPower instead of being fixed to AutoPASPower regardless of MaxPower. That would give the dial a bit more resolution when the maximum power is lower. So the maximum adjust value would be 750 watts when using the "Legal" preset and 1000 watts when using the other presets.

Another option that simplifies things would be to get rid of AutoPAS Power Level (global value), using MaxPower that is preset-specific. But, perhaps you need both due to Byzantine PAS vs throttle power regulations in certain jurisdictions.

And, a quibble in the Setup Utility:
When downloading parameters to my laptop, the downloaded Odometer Distance is (Overall Distance) - (Trip Distance). The Trip Distance is not reflected in the downloaded value, even though Overall Distance includes Trip Distance on the CAV2 itself. If I do a trip reset before downloading (zeroing out Trip Distance), then the Overall Distance is the same as the downloaded Odometer Distance.

Shall I submit these through your online form?

mrbill · Jan 12, 2015

justin_le said:
If you take a couple peak readings of your startup amperage draw at different fast ramp rates that would be quite interesting to see here.

Here's some data to chew on.

I have two bikes. One uses a true mid-drive where the motor is connected to a jack-shaft between the crank and the rear wheel:
http://mrbill.homeip.net/albums/powerPursuitBuild/single-image/20100902-powerPursuitBuild66-photo.html#anchor

The other bike has the motor driving a planetary gearbox that drives the bottom bracket:
http://mrbill.homeip.net/albums/powerGoldRushBuild/single-image/20110602-powerGoldRushBuild85-photo.html#anchor

Both bikes use the same motor:
http://mrbill.homeip.net/albums/mac_bmc_600w_motor/index.html

I measured peak currents for both bikes using firmware P7. When flipping the throttle from zero to full, I first spun the cranks by hand to get the rear wheel spinning to reflect reality and so that I was measuring the peak current to spin up only the drivetrain.

I used the following relevant parameters:

WGain = 20
AGain = 50
fastRampCurrent = 0
fastRamp = max(upRamp, 5.0) V/sec

Bike w/planetary gearbox:

Code:

upRamp (V/sec)   peak currents observed (A)
---------------  ---------------------------------
20.0             4.43, 4.38, 8.93, 5.08, 4.51
10.0             5.26, 4.60, 5.13, 5.20, 4.99
5.0              4.93, 7.11, 2.94, 6.65, 5.02, 5.16, 8.16, 3.78
2.5              4.60, 5.72, 5.76, 4.97, 5.55, 4.48, 4.09
1.0              3.54, 3.43, 3.41, 3.74, 3.08

Bike w/o planetary gearbox:

Code:

upRamp (V/sec)   peak currents observed (A)
---------------  ---------------------------------
20.0             6.42, 4.67, 6.61, 5.43, 4.46
10.0             4.55, 4.49, 4.49, 4.40, 4.84
5.0              4.43, 4.24, 4.35, 5.08, 5.26, 4.59
2.5              5.41, 5.22, 4.51, 5.17, 6.00, 5.85
1.0              3.91, 3.50, 3.28, 3.24, 3.33, 3.34

I suspect that at the slow 1.0 V/sec ramp rate the peak current is being limited by the throttle response. At 2.5 V/Sec and higher, we're probably seeing the true peaks.

Given these figures I settled on fastRampCurrent = 6.0 Amps as a starting point.

justin_le · Jan 13, 2015

mrbill said:
Here's some data to chew on.

Thanks Bill, this is great. Appreciate you taking the time to dig deep here.

Given these figures I settled on fastRampCurrent = 6.0 Amps as a starting point.

After giving more thought to the algorithm I realized that we don't need to be too concerned with instantaneous current peaks, what would matter more is to set it a little higher than the average current draw of motor acceleration, rather than the peak draw. So if you find 6A has a bit too much punch, then I'd go down to the 3-4A range.

http://mrbill.homeip.net/albums/powerPursuitBuild/single-image/20100902-powerPursuitBuild66-photo.html#anchor

Hey wow, that is the legendary shimano megarange 11-34t freewheel, there was a time when those were selling on ebay for over $150.

Regarding the additional quibble points,
#1 is understood, I'll need to give a little bit of consideration if there aren't people who would be bothered by this change (for instance, someone wants their AUX controlled assist level to be be consistent among presets, when other parameters like the max speed and max power are changed, and then this could cause a variation from one preset to the next).

#2 Wouldn't work well, since most people have their PAS assist set to a lowish maximum wattage for background assist power (like 500W), and then have the throttle whenever they want full power. So it's important that the PAS watts can be set independently from the Max watts. But we'll be making the PAS Watts be per-profile rather than global in the 3.1 FW.

#3, Totally valid and good observation, I honestly didn't think people would make much notice of this. The CA adds the trip distance to the odometer in eeprom only when you hold the button and do a reset, until then it is stored as two separate values. At the moment the software isn't reading or making use of the separate trip distance number, but we should be able to make it read that from the CA too and add the value to the displayed odometer, exactly like the CA itself does. I'll put this on our future enhancements list.

I'll be away for most of the next 2 weeks so won't have any additional builds or comments until the 26th of January, but in the meantime if as many people can flash this P8 firmware to their CA's and confirm that all runs ace as expected, that will be wonderful, and if any glitches be found then report here and they'll be tackled on my return.

matsh · Jan 17, 2015

On my ca v3 with my crystalyte tc60 motor, the bike was left for Couple of days in a room that stays at 15 degree celsius. When i today turned bike on the reading was around 31 degree celsius on my ca. Thats allot off difference vs the room temperatur. And when it says 120c what temp does my motor then have...

madnomad · Jan 23, 2015

Here are a few ideas I'd like to see implemented in the CA v4 (any idea when its coming btw?):

A built in DC converter which can output 5v, 8.4v and 12v. I have spent soooo much time trying to find a solution to power my lights, and it seems like most people would love to power their lights from their CA, so this would save a lot of people a lot of time. It's also a pain having to conceal a bulky DC converter somewhere between the light and the CA.

I would like to see waterproof sockets on the CA to replace all the cables coming out of it. It's hard to know whether I should cut them off or if I'll want to use them at some point in the future.

I would like to see the screen replaced with a something a bit more up to date. LCD screens are pretty advanced these days and this one seems like something from the 80's (really not trying to be rude). It seems like the best idea would be to design it so an android or iphone could be used as the screen, and if this could be done via bluetooth then that would be fantastic. The mind boggles as to the new opportunities this would bring- music, GPS, rear view mirror (on screen), etc etc etc.

I think a few settings seem to be unnecessary and complicate things. Maybe I'm wrong here, but couldn't the recommended "gain" settings be calculated automatically based on the power limits you enter?

NeilP · Jan 23, 2015

I'd certainly like to see a change to the cabling. I have had to unsolder all of mine and extend them to go in to the frame triangle, so either a socket or a single thick cable coming out.

With a socket various plug options could be offered, but I could see issues with strength if a socket / plug arrangement were used. it would need to be a big plug, meaning it sticking out a long way. I think a socket idea would only be feasible if complete box redesign was down too.
For me just longer cables or a single bigger multi core cable would be better to get the black non water resistant connectors off the handle bars.

izeman · Jan 23, 2015

regarding "old fashioned" screen. yes. it's not really fancy - but for sure the best regarding readability. it's backlit and has a high contrast. if you ever tried to read something from your phone in bright sunlight you know what i mean.
the rat's nest coming out of CA is not nice. that's true. i always desolder the wires and install new sets as soon as i buy a new unit.

NeilP · Jan 24, 2015

I do have a very urgent request for something that does need changing ..before I do physical damage to something due to my in ability to understand there CA Setup Utility!

Any chance of a change to the SEND Icon or a timer that pops up a warning on the Utility when sending new settings back to the CA. Does not need to be a confirmation box, maybe click the button to SEND TO the CA..which then brings up a BIG RED BOX with flashing lights and sirens with a message sosmeth gliek "Don't be a frock wit...do you really want to do this"

Yet again I have managed to overwrite my CA with blank settings from a newly installed version of the software on a new computer. The last computer died, ..the data hard DD did not like the damp shed. so all my old config settings were on the that machine

So the first thing I wanted to do was read the settings on the CA.. I have always had an issue with the Send or Receive Icons. I always look at them and have to think..OK, which way are the arrows pointing ...and foe some reason ...get it wrong so very often.

For weeks I had been tinkering with gain settings etc, tuning it all to 'perfection', then a few months back other stuff intervened, then the PC dies, then I stripped i, the bike down to fit new ignition switch, and new DC=DC block, do lights final wiring etc. controller died, repaired, died again... >>agggggghhhbh and now I have wiped my only set of settings of my 'final config' by writing To the CA and not Reading FROM it. What a f in g muppet.

But seriously for some reason those icons are a bit ...well not confusing...well they should not be..too small maybe? the arrows too small ? Maybe a Prompt , which can be disabled, when the send button is hit..warngin theta the CA is going to be overwritten....please before I frock it up again.

mrbill · Jan 26, 2015

justin_le said:
After giving more thought to the algorithm I realized that we don't need to be too concerned with instantaneous current peaks, what would matter more is to set it a little higher than the average current draw of motor acceleration, rather than the peak draw. So if you find 6A has a bit too much punch, then I'd go down to the 3-4A range.

I settled on 4A, although it did not behave noticeably differently than 6A. I did start to notice some sluggishness off the line at 3A and esp. at 2A.

The speed limit undershoot problem still exists to the same degree as in P7 and P6 firmware.

I appreciate that the AuxAdjust display persists for maybe one second (although sometimes less following very small adjustments) rather than disappearing the moment the user stops making an adjustment. But, I think the persistence of this display should be longer still.

I often find myself adjusting the PAS watts while focused on distant objects. When conditions allow I will take my eyes off the road and cast my gaze on the CAV3 on my handlebars. Having the display persist after an adjustment for as long as the alternating "mi" and "Ah" display on the main screen, or about 4 seconds would not be too long and would be consistent with the user interface for those other alternating displays, especially since there is no way to recall this display without changing the Adjust level.

Overall P8 improves the manners of the PAS when controlling a mid-drive system. I may tweak the settings further, but for now I'm reasonably happy with the behavior. Response is much smoother than even P7, and I like how the power dropout when switching from throttle to PAS mode is minimized, as would occur when starting from a stop using throttle and pedaling, then releasing the throttle to switch over to PAS mode while continuing to pedal. The dropout used to be as much as one second before the PAS kicked in.

Issues:

I experience unbidden motor power under the following conditions: My Aux Pot is adjusted for 0 (zero) PAS watts. I start from a stop by both pedaling and throttling at moderately-high power. When I reach a comfortable cruising speed I release the throttle. Motor power drops to zero, then surges before decaying slowly to zero. Motor power should never be more than zero when PAS watts is zero.

I observed a few instances of misbehavior that I haven't yet been able to reproduce or analyze. I'm reporting them as possible but not confirmed problems.

1) Upon "turning on" my controller with the power switch, the CA display remained blank. Turning power off and on again restored normal functionality.

2) After coasting down a hill with the PAS watts set to a moderate power level, I resumed pedaling and got no power. Then I attempted throttle and got no power. Cycling power to the CAV3 restored normal functionality.

3) Shortly after the observation in (2) but after I had restored normal functionality, I coasted to a stop at a red light and upon releasing the e-brake observed the motor creeping at low power (30-40 watts). When I no longer had traffic behind me, I checked and confirmed that min. ThO voltage was 0.62 while the controller's apparent minimum throttle voltage (before motor began to creep) was 0.74 volts.

(1) occurred on "Bike A", while (2) and (3) occurred on "Bike B".

vanbhills · Feb 1, 2015

I think the draft manual should start with resetting the configuration in the cycle analyst.
I started with section: 4.2.1.1 (The throttle test to confirm controller compatibility)

I made only those changes to do the throttle test. When the motor didn't enable I was confused and thought I had been provided an old controller with a very recent purchase. After puzzling over it for awhile I found a topic on using the setup utility and noticed lots of differences between the "default" and the settings loaded from my CA including a THUN enabled and many other differences. I glossed over these settings earlier without making any changes as this is my very first Cycle Analyst (and ebike) I didn't have much experience to draw on.

I sent the default config to the CA and the throttle works fine (along with all the other components excluding the TDCM).

On the downside, my TDCM sensor seems to report 4.77 volts regardless of any weight on the pedals and I can't get the PAS/Torque config to work. Annoying if that little tip wasn't in the slot properly. It is a fiddly thing. Is there anything I can check before I pull the cranks and try to re-install the sensor again?

teklektik · Feb 1, 2015

It's always a little disappointing when the Guide don't fulfill someone's needs.
Here's a few thoughts:

Requiring a reset to defaults as an initial step problematic. The Guide is designed to allow setup without use of the Setup Utility - not everyone has a PC handy for the job. The other strategy to restore defaults by flashing the unit may also introduce difficulties since not everyone has a programming cable. The Guide instead expects a factory-fresh unit, mentions defaults, and provides settings that should disable ill effects of advanced functions.

That said, a section might be added to specifically address the matter of defaults if difficulties arise.
These settings should have disabled interactions from your PAS and all other advanced features. Since your PAS was enabled, I suspect you missed this step. Perhaps this should be reworded to verify these settings....
In general, the defaults will allow the CA to operate out-of-box for modestly powered bikes of 'usual' weight. The Guide procedures allow you to get the most out your CA by adjusting the throttle and enabling/tuning advanced features. You report that the defaults 'work' - which is no surprise, but I encourage you to pursue the setup steps to get the most out of your purchase.
The fixed voltage on your TDCM is a problem. There are no Setup errors that can account for that behavior. The applicable plugs are simple and unique so that is off the table as well. I suggest you contact ebikes.ca Support for assistance.

vanbhills · Feb 1, 2015

[*]These settings should have disabled interactions from your PAS and all other advanced features. Since your PAS was enabled, I suspect you missed this step. Perhaps this should be reworded to verify these settings....

The settings you referenced 4.6.1 are after the section I was attempting 4.2.1.1. I didn't "miss" them because I had yet to get to this point and was following "step by step". My settings from Grin fresh from the box had the PAS enabled for a THUN along with many other differences from the expected out of box defaults. Thus, I was trying to suggest moving some portion of these initial default settings up.

[*]In general, the defaults will allow the CA to operate out-of-box for modestly powered bikes of 'usual' weight. The Guide procedures allow you to get the most out your CA by adjusting the throttle and enabling/tuning advanced features. You report that the defaults 'work' - which is no surprise, but I encourage you to pursue the setup steps to get the most out of your purchase.

Of course, I'll follow the tuning.... after I resolve the sensor issue.

[*]The fixed voltage on your TDCM is a problem. There are no Setup errors that can account for that behavior. The applicable plugs are simple and unique so that is off the table as well. I suggest you contact ebikes.ca Support for assistance. [/list][/quote]

I expected as much. I'll try and re-install the bracket first. I was hesitating because its non-trivial and I had already had made a simple error.

Thanks for the response.

teklektik · Feb 1, 2015

vanbhills said:
The settings you referenced 4.6.1 are after the section I was attempting 4.2.1.1. I didn't "miss" them because I had yet to get to this point and was following "step by step". My settings from Grin fresh from the box had the PAS enabled for a THUN along with many other differences from the expected out of box defaults. Thus, I was trying to suggest moving some portion of these initial default settings up.

Ah - I see your point. :idea:

The position of the settings in 4.6.1 was carefully arranged - but section 4.2.1.1 was added some time afterward - and it seems the Big Picture was not well re-evaluated. If the CA is so fouled up it won't spin the wheel, then clearly 4.2.1.1 is in jeopardy. ... This needs some thought. Thanks!

vanbhills said:
Of course, I'll follow the tuning.... after I resolve the sensor issue.

There's no particular reason to delay unless it's convenient for some other reason - PAS should be disabled to do the initial throttle tuning anyway.

Motobikewheels · Feb 2, 2015

Anybody know of a good way to hook up 10k thermistors for both motor and controller at the same time?

NeilP · Feb 2, 2015

It has been discussed before somewhere in this monster thread. Good luck!

best bet is a switch so you toggle between one of the other.

justin_le · Feb 2, 2015

mrbill said:
I appreciate that the AuxAdjust display persists for maybe one second (although sometimes less following very small adjustments) rather than disappearing the moment the user stops making an adjustment. But, I think the persistence of this display should be longer still.

OK, I'll need to get a few more people's opinions on this though, since this kind of thing can have a wide range of preferences and it'd be best to be somewhere in the middle, I'm definitely on the impatient side and like the screen to return to normal as soon as I'm happy with where the pot is at. Of others who've updated to the P8 firmware and who use the Aux Pot adjustment, what do you think of the duration of display persistance?

I experience unbidden motor power under the following conditions: My Aux Pot is adjusted for 0 (zero) PAS watts. I start from a stop by both pedaling and throttling at moderately-high power. When I reach a comfortable cruising speed I release the throttle. Motor power drops to zero, then surges before decaying slowly to zero. Motor power should never be more than zero when PAS watts is zero.

This is curious but it sounds like there was some kind of 'pent-up' PAS watts assist during the time that you were throttling, and only on releasing the throttle and re-enabling PAS mode did this get a chance to power and then settle to zero. You are saying that this particular phenomenon is easy to replicate?

3) Shortly after the observation in (2) but after I had restored normal functionality, I coasted to a stop at a red light and upon releasing the e-brake observed the motor creeping at low power (30-40 watts). When I no longer had traffic behind me, I checked and confirmed that min. ThO voltage was 0.62 while the controller's apparent minimum throttle voltage (before motor began to creep) was 0.74 volts.

This kind of slow creep is what can happen if your command watts isn't 0 but something very close, like just a couple watts. Is it possible that the Aux Input was just a hair trigger above the 0W point?

justin_le · Feb 2, 2015

matsh said:
On my ca v3 with my crystalyte tc60 motor, the bike was left for Couple of days in a room that stays at 15 degree celsius. When i today turned bike on the reading was around 31 degree celsius on my ca. Thats allot off difference vs the room temperatur. And when it says 120c what temp does my motor then have...

Are you sure that you have a 10K NTC thermistor with appropriate beta constant as your temperature sensor?? You can easily check it with a multimeter. If the motor is at 15 degrees you should measure about 15.5 kOhm of resistance between the thermistor pins. 30oC on the CA display would normally mean about 8K of resistance on the thermistor.

justin_le · Feb 2, 2015

madnomad said:
Here are a few ideas I'd like to see implemented in the CA v4 (any idea when its coming btw?):

Ha ha, when you see a thread here that says "CA V4 preview and first beta release" would be the time for suggestions along these lines. For a number of reasons, for now I gotta squeeze everything we can from the current hardware, which means no realtime clocks, no onboard bluetooth wireless, no graphics display screens, no enclosure level connectorization etc.

I think a few settings seem to be unnecessary and complicate things. Maybe I'm wrong here, but couldn't the recommended "gain" settings be calculated automatically based on the power limits you enter?

Not exactly unfortunately. The mass of the vehicle and the torque vs. throttle relationship of the particular motor/controller pair matter in a huge way. So for a given incremental change in the throttle voltage, how quickly does the vehicle start to accelerate? It would be possible though to have an application outside of the CA itself (like a web app or something built into the CA's setup utility) which could provide a set of recommended gain values based on a few parameters like the vehicle mass, motor kV value, wheel diameter, battery voltage, and a few other things. The problem is that I don't completely know myself how best to predict that computationally.

justin_le · Feb 2, 2015

NeilP said:
But seriously for some reason those icons are a bit ...well not confusing...well they should not be..too small maybe? the arrows too small ? Maybe a Prompt , which can be disabled, when the send button is hit..warngin theta the CA is going to be overwritten....please before I frock it up again.

Hi Neil, that's a totally valid point. I thought when sketching them that the "into CA" and "out of CA" arrows would be nice and visually clear, but in use I also need to do a double take and hover over to see the dialogue "write setup to CA" or "read setup from CA" to be sure. A more clear warning sign prior to writing and and a larger and more differentiated read/write icon would help eliminate accidental over-writes like this. We'll do it. Thanks for the suggestion.

teklektik · Feb 2, 2015

roadrash said:
Anybody know of a good way to hook up 10k thermistors for both motor and controller at the same time?

NeilP said:
It has been discussed before somewhere in this monster thread. Good luck!

best bet is a switch so you toggle between one of the other.

Justin called out two techniques for dual sensors way back on page 1 (!), but to my knowledge they have not been implemented. However, the reasoning behind using a single probe on the motor seems to lead to a simple operational solution - works for me...

justin_le said:
...But your other suggestion actually works OK with NTC thermistors. If you get two 20K thermistors and wire them up in parallel, then it behaves like a single 10K thermistor that (due to the high nonlinearity) is mostly weighted to show the higher of the two temps.

You could also have two semiconductor resistors that have a voltage output and link them up to the temp input via a diode, and then just the highest one would show.

I think in most setups:

motor winding temp > controller temp >> battery temp,

so it's usually the motor that hits thermal issues first. Interestingly though, both the controller and the motor have the same primary heating source (I^2R on the phase current), so limiting and regulating one will also limit the other in tandem.

-Justin

mrbill · Feb 4, 2015

justin_le said:
I experience unbidden motor power under the following conditions: My Aux Pot is adjusted for 0 (zero) PAS watts. I start from a stop by both pedaling and throttling at moderately-high power. When I reach a comfortable cruising speed I release the throttle. Motor power drops to zero, then surges before decaying slowly to zero. Motor power should never be more than zero when PAS watts is zero.

Click to expand...

This is curious but it sounds like there was some kind of 'pent-up' PAS watts assist during the time that you were throttling, and only on releasing the throttle and re-enabling PAS mode did this get a chance to power and then settle to zero. You are saying that this particular phenomenon is easy to replicate?

Yes. This occurs every time for me. Your description of "'pent-up' PAS watts" is exactly what this feels like. An alternate PAS-mediated throttle is running in the background and is only unmasked after I release the physical throttle that overrides the PAS throttle. The problem is that the PAS throttle should be zero at all times when PAS watts is adjusted to zero.

3) Shortly after the observation in (2) but after I had restored normal functionality, I coasted to a stop at a red light and upon releasing the e-brake observed the motor creeping at low power (30-40 watts). When I no longer had traffic behind me, I checked and confirmed that min. ThO voltage was 0.62 while the controller's apparent minimum throttle voltage (before motor began to creep) was 0.74 volts.

Click to expand...

This kind of slow creep is what can happen if your command watts isn't 0 but something very close, like just a couple watts. Is it possible that the Aux Input was just a hair trigger above the 0W point?

It's possible. I haven't experienced this behavior since this one occasion.

Cycle Analyst V3 preview and first beta release

10 kW

Administrator

Attachments

10 kW

Administrator

10 kW

10 kW

Administrator

1 W

1 W

1 GW

1 GW

1 GW

10 kW

10 µW

100 MW

10 µW

100 MW

100 W

1 GW

Administrator

Administrator

Administrator

Administrator

100 MW

10 kW

Similar threads