Cycle Analyst V3 preview and first beta release

Get real world experience and user feedback on the electric bicycle products.
User avatar
justin_le   10 MW

10 MW
Posts: 2152
Joined: May 28 2007 3:27am
Location: Vancouver
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by justin_le » Nov 15 2018 1:46am

madnut wrote:
Nov 14 2018 2:29pm
I am assuming what happens is that CA notices that power is dropping and because my PAS is set to power mode it ramps up the out throttle to full trying to get more power. When I release the break throttle the CA is outputting fully open throttle 4V so the bike jumps.
Exactly. What you need to do is tell the CA that you are braking via the CA's ebrake input line. This is easy if you have a digital ebrake/regen input to the controller, but if it's an analog input to the controller then you'd have to set up a simple comparator circuit so that when you have the regen throttle activated, it pulls the CA's ebrake input low. If you're not EE inclined, this can be done mechanically too, with a simple limit switch that's activated by the rotation of your regen throttle.
2. Software wize (I dont know how well this would work but...) I think when the CA sees the power drop from lets say 500W to negative wattage it should put the throttle to low and not allow it to climb until the wattage is zero or positive.
You're spot on that this approach would work and it's actually been considered, but there are unexpected consequences to doing something like that which can solve one persons situation but cause grief in others. For instance you can have negative watts if you have a solar panel hooked up to the battery, and then the CA would think that there is regen and not apply and PAS power until you bike through a dark zone and the solar input goes up to zero. And then you'd have PAS again, but only so long as the PAS watts stayed higher than the negative solar watts or else you'll get locked out.

So just saying it would take some careful consideration to implement a feature that's making presumptions on what the rider is doing (ie applying regen brakes) based on what the CA sees for the consumption numbers.
Currently recovering from the Suntrip race on a back to back tandem solar powered row/cycle trike. 550 watt solar roof, dual Grin All Axle hub motors, dual Phaserunner controllers, 12 LiGo batteries, and a whole wack of gear.

Now back in Vancouver with my Big Dummy Frame (yes This One, thanks ES!) with Grin all-axle front hub, Phaserunner controller, and 52V 19Ah Cellman triangle pack
My website: http://www.ebikes.ca
Please contact via email, info@ebikes.ca, rather than PMs, which are disabled

madnut   100 mW

100 mW
Posts: 38
Joined: Nov 16 2016 12:40am

Re: Cycle Analyst V3 preview and first beta release

Post by madnut » Nov 15 2018 4:35am

justin_le wrote:
Nov 15 2018 1:46am
madnut wrote:
Nov 14 2018 2:29pm
I am assuming what happens is that CA notices that power is dropping and because my PAS is set to power mode it ramps up the out throttle to full trying to get more power. When I release the break throttle the CA is outputting fully open throttle 4V so the bike jumps.
Exactly. What you need to do is tell the CA that you are braking via the CA's ebrake input line. This is easy if you have a digital ebrake/regen input to the controller, but if it's an analog input to the controller then you'd have to set up a simple comparator circuit so that when you have the regen throttle activated, it pulls the CA's ebrake input low. If you're not EE inclined, this can be done mechanically too, with a simple limit switch that's activated by the rotation of your regen throttle.
2. Software wize (I dont know how well this would work but...) I think when the CA sees the power drop from lets say 500W to negative wattage it should put the throttle to low and not allow it to climb until the wattage is zero or positive.
You're spot on that this approach would work and it's actually been considered, but there are unexpected consequences to doing something like that which can solve one persons situation but cause grief in others. For instance you can have negative watts if you have a solar panel hooked up to the battery, and then the CA would think that there is regen and not apply and PAS power until you bike through a dark zone and the solar input goes up to zero. And then you'd have PAS again, but only so long as the PAS watts stayed higher than the negative solar watts or else you'll get locked out.

So just saying it would take some careful consideration to implement a feature that's making presumptions on what the rider is doing (ie applying regen brakes) based on what the CA sees for the consumption numbers.
Thanks a bunch, I'm glad to hear its a known issue and the solution being relatively simple. This is a perfect excuse for me to try out a ATTINY85. I experimented with a full size Arduino and can get it to signal when it sees 1V from the throttle and its adjustable.

User avatar
Tom on 101   10 mW

10 mW
Posts: 26
Joined: Jun 17 2016 10:12am
Location: Aptos, California

Re: Cycle Analyst V3 preview and first beta release

Post by Tom on 101 » Nov 19 2018 3:16pm

justin_le wrote:
Nov 14 2018 7:04am
Tom on 101 wrote:
Nov 09 2018 8:51pm
We are getting a W flag on the diagnostic screen which to me makes no sense.
In your setup, you are running in a torque assist PAS mode, which means while you are pedaling whether hard or lightly, the CA is ALWAYS trying to maintain a regulated constant wattage output, hence the W flag should always be capitalized since you are always limiting the output based on a target wattage.

Where did your 160 Nm/V torque scaling and 0.92V offset come from? Right now you have your maximum assist set to 2x, so to get 600 watts of output power from the controller the CA3 will need to see 300 watts of human input power.
OK I see, I was under the false impression that if I saw the capital W that means the limit was hit.

As far as the 0.92V, that was the offset measured at set up. The 160Nm/V was reach by experimenting with a few different numbers, starting at 20Nm/V and doubling the number until we had some response from the PAS.

I guess Im still confused about the maximum assist at 2x (is that the same as Assist Scale Factor?) which I have set 2W/HW? So really I should have something like 5W/HW so that when someone hits 100 Watts the motor would be around 500 watts output?

User avatar
justin_le   10 MW

10 MW
Posts: 2152
Joined: May 28 2007 3:27am
Location: Vancouver
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by justin_le » Nov 21 2018 5:27am

Tom on 101 wrote:
Nov 19 2018 3:16pm
OK I see, I was under the false impression that if I saw the capital W that means the limit was hit.
Well it is, in that the watts limit is the lower of your maximum watt setting, or the watts that is commanded by the PAS control system. The flags are there mostly to give insight into what is controller the throttle output voltage that you see on that same screen.
As far as the 0.92V, that was the offset measured at set up. The 160Nm/V was reach by experimenting with a few different numbers, starting at 20Nm/V and doubling the number until we had some response from the PAS.
OK more ideally you would look at the human watts (3rd display screen) and adjust it until it showed a value that correlates with what you know for your human power output under different amounts of exertion. Normal riding is like 100-150 watts, pedaling hard is like 250 watts, really giving it for a sprint would be like 350-400 watts.

If you see 400 watts when you're just pedaling along, you've set this scaling way too high.
I guess Im still confused about the maximum assist at 2x (is that the same as Assist Scale Factor?) which I have set 2W/HW? So really I should have something like 5W/HW so that when someone hits 100 Watts the motor would be around 500 watts output?
Correct, but if you've set it up so that the CA thinks your doing 250 watts when you're really only putting in 100 watts, then you'd wind up with a lower desired W/HW. At the end of the day it doesn't matter so much just for PAS assist control, but it's better if the human watts numbers bear a meaningful relation to your actual pedal power.
Currently recovering from the Suntrip race on a back to back tandem solar powered row/cycle trike. 550 watt solar roof, dual Grin All Axle hub motors, dual Phaserunner controllers, 12 LiGo batteries, and a whole wack of gear.

Now back in Vancouver with my Big Dummy Frame (yes This One, thanks ES!) with Grin all-axle front hub, Phaserunner controller, and 52V 19Ah Cellman triangle pack
My website: http://www.ebikes.ca
Please contact via email, info@ebikes.ca, rather than PMs, which are disabled

User avatar
justin_le   10 MW

10 MW
Posts: 2152
Joined: May 28 2007 3:27am
Location: Vancouver
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by justin_le » Nov 21 2018 5:36am

madnut wrote:
Nov 09 2018 1:45am
Ok I have updated and things are much better.
Couple of bugs to report (this is under high power calibration):
1. If I set 2A in the Fast Ramp Threshold, on the device it looks like its actually sets 20 Amps. This results in a very jumpy DD hub bike in PAS mode, not very safe :) (I fix it by setting it on the device itself).
2. If I apply the ebrake and keep peddling, and then while pedaling release the ebrake, the throttle out will climb very slowly around 0.1V/s.
3. I have two presets with #2 as starting one, if I try to switch I get watchdog and a crash. Here is my setup https://1drv.ms/u/s!AmHyrxxOO3jpgYYrteGTTlkYyNXtSw
Thanks again for the quick feedback, CA3.13 Beta2 is now available if you click the "get new firmware" link in the CA setup utility, and that should fix all of these thing. We don't have too many people using PAS in high range mode so you helped us track down some oversights that go back a while.
Feature request:
Not sure if this is doable but in high power mode the smallest power limit I can set is 1kw. I would love to have at least one decimal point so i can create a legal 0.7kw preset. Ideally if we could have 2 decimal points I would set it to 0.75kw.
So this was a small bug in the displayed resolution. You'll notice that it gave two decimals of precision in the normal range watts mode, but no decimals in kW, when it was supposed to be the other way around. If you go File -> Remove All Firmware, and then hit the button "get new firmware" to re-download and unpack them all, then it will properly give you two decimals of resolution for the kW in high range mode on this and the previous firmwares too.
RemAllFW.jpg
RemAllFW.jpg (70.57 KiB) Viewed 585 times
Let us know if that does the trick on all counts for ya.
Currently recovering from the Suntrip race on a back to back tandem solar powered row/cycle trike. 550 watt solar roof, dual Grin All Axle hub motors, dual Phaserunner controllers, 12 LiGo batteries, and a whole wack of gear.

Now back in Vancouver with my Big Dummy Frame (yes This One, thanks ES!) with Grin all-axle front hub, Phaserunner controller, and 52V 19Ah Cellman triangle pack
My website: http://www.ebikes.ca
Please contact via email, info@ebikes.ca, rather than PMs, which are disabled

madnut   100 mW

100 mW
Posts: 38
Joined: Nov 16 2016 12:40am

Re: Cycle Analyst V3 preview and first beta release

Post by madnut » Nov 21 2018 12:13pm

Oh o! I tried to flash 1.13b but something went wrong with my cable mid flash, and now instead of a boot screen i get just blank screen when I turn the CA on. How do I go about recovering it?

User avatar
Adamlivi   10 mW

10 mW
Posts: 25
Joined: Nov 26 2016 2:52pm
Location: Norcross, Georgia

Re: Cycle Analyst V3 preview and first beta release

Post by Adamlivi » Nov 21 2018 1:26pm

teklektik wrote:
Jul 28 2018 12:36pm
Adamlivi wrote:
Jul 28 2018 9:01am
...write failure at 2% with connection issue
...
Any help or is this a "send in" repair?
It's not bricked. No 'send in ' required.
This is the comms problem referenced recently in this thread. The difficulty appears to be some very close timing issue in the Setup Utility 1.54 but we are sort of working around it for now with CA firmware release 3.12. That said - when you get a timeout like this when the boot install process is underway, it's out of the scope of a CA firmware patch to resolve. My recommendation is to try again, checking the 'reset defaults' box. Power up the CA **AFTER** you begin the download. This will send control directly to the bootloader with no interference by the CA firmware.

As much as a PITA as this seems, you may be able to get this to load with a few tries. Since it's timing related, you can also try another PC if you have one handy.

If this remains unresolved, email Grin and reference your post in this thread.

madnut   100 mW

100 mW
Posts: 38
Joined: Nov 16 2016 12:40am

Re: Cycle Analyst V3 preview and first beta release

Post by madnut » Nov 21 2018 1:33pm

Adamlivi wrote:
Nov 21 2018 1:26pm
teklektik wrote:
Jul 28 2018 12:36pm
Adamlivi wrote:
Jul 28 2018 9:01am
...write failure at 2% with connection issue
...
Any help or is this a "send in" repair?
It's not bricked. No 'send in ' required.
This is the comms problem referenced recently in this thread. The difficulty appears to be some very close timing issue in the Setup Utility 1.54 but we are sort of working around it for now with CA firmware release 3.12. That said - when you get a timeout like this when the boot install process is underway, it's out of the scope of a CA firmware patch to resolve. My recommendation is to try again, checking the 'reset defaults' box. Power up the CA **AFTER** you begin the download. This will send control directly to the bootloader with no interference by the CA firmware.

As much as a PITA as this seems, you may be able to get this to load with a few tries. Since it's timing related, you can also try another PC if you have one handy.

If this remains unresolved, email Grin and reference your post in this thread.
Thanks! I will try this out.

User avatar
justin_le   10 MW

10 MW
Posts: 2152
Joined: May 28 2007 3:27am
Location: Vancouver
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by justin_le » Nov 21 2018 7:12pm

madnut wrote:
Nov 21 2018 12:13pm
Oh o! I tried to flash 1.13b but something went wrong with my cable mid flash, and now instead of a boot screen i get just blank screen when I turn the CA on. How do I go about recovering it?
You can always always always just apply power to the CA after you have clicked the "update firmware" button and then the software will catch the CA while it is still in the bootloader and start updating. As Adam quoted, you'll just need to check the "reset settings to default" checkbox with this approach. That's because when you get to the bootloader this way the software has no way to know what firmware is on the CA and if the internal settings are compatible, so it will want you to load on default settings known to be appropriate for that firmware.
Currently recovering from the Suntrip race on a back to back tandem solar powered row/cycle trike. 550 watt solar roof, dual Grin All Axle hub motors, dual Phaserunner controllers, 12 LiGo batteries, and a whole wack of gear.

Now back in Vancouver with my Big Dummy Frame (yes This One, thanks ES!) with Grin all-axle front hub, Phaserunner controller, and 52V 19Ah Cellman triangle pack
My website: http://www.ebikes.ca
Please contact via email, info@ebikes.ca, rather than PMs, which are disabled

madnut   100 mW

100 mW
Posts: 38
Joined: Nov 16 2016 12:40am

Re: Cycle Analyst V3 preview and first beta release

Post by madnut » Nov 22 2018 12:42am

justin_le wrote:
Nov 21 2018 7:12pm
madnut wrote:
Nov 21 2018 12:13pm
Oh o! I tried to flash 1.13b but something went wrong with my cable mid flash, and now instead of a boot screen i get just blank screen when I turn the CA on. How do I go about recovering it?
You can always always always just apply power to the CA after you have clicked the "update firmware" button and then the software will catch the CA while it is still in the bootloader and start updating. As Adam quoted, you'll just need to check the "reset settings to default" checkbox with this approach. That's because when you get to the bootloader this way the software has no way to know what firmware is on the CA and if the internal settings are compatible, so it will want you to load on default settings known to be appropriate for that firmware.
Thanks! its back up and running.

madnut   100 mW

100 mW
Posts: 38
Joined: Nov 16 2016 12:40am

Re: Cycle Analyst V3 preview and first beta release

Post by madnut » Nov 22 2018 12:49am

justin_le wrote:
Nov 21 2018 5:36am
madnut wrote:
Nov 09 2018 1:45am
Ok I have updated and things are much better.
Couple of bugs to report (this is under high power calibration):
1. If I set 2A in the Fast Ramp Threshold, on the device it looks like its actually sets 20 Amps. This results in a very jumpy DD hub bike in PAS mode, not very safe :) (I fix it by setting it on the device itself).
2. If I apply the ebrake and keep peddling, and then while pedaling release the ebrake, the throttle out will climb very slowly around 0.1V/s.
3. I have two presets with #2 as starting one, if I try to switch I get watchdog and a crash. Here is my setup https://1drv.ms/u/s!AmHyrxxOO3jpgYYrteGTTlkYyNXtSw
Thanks again for the quick feedback, CA3.13 Beta2 is now available if you click the "get new firmware" link in the CA setup utility, and that should fix all of these thing. We don't have too many people using PAS in high range mode so you helped us track down some oversights that go back a while.
Feature request:
Not sure if this is doable but in high power mode the smallest power limit I can set is 1kw. I would love to have at least one decimal point so i can create a legal 0.7kw preset. Ideally if we could have 2 decimal points I would set it to 0.75kw.
So this was a small bug in the displayed resolution. You'll notice that it gave two decimals of precision in the normal range watts mode, but no decimals in kW, when it was supposed to be the other way around. If you go File -> Remove All Firmware, and then hit the button "get new firmware" to re-download and unpack them all, then it will properly give you two decimals of resolution for the kW in high range mode on this and the previous firmwares too.

RemAllFW.jpg

Let us know if that does the trick on all counts for ya.
Thanks! most of the issues are gone.
But #2 still seems to be there, if i press ebrake and keep peddling, then when ebrake is released pas takes forever to increase throttle out.

I also have noticed a similar issue but i don't have a consistent repro for. I noticed that if I am going down hill and stop peddling sometimes when I resume peddling the power ends up stuck at like 100W and wont climb (My pass is set to 600W so it should climb). If I stop peddling and resume again its back to normal. The throttle might be doing the same thing here as issue #2 but i haven't had a chance to be in that menu when it happens to see what its doing.
I am hoping that fixing #2 will fix this as well. Or in the least it will help me isolate a repro because sometimes I'm not sure if I hit the break or brushed it accidentally.

User avatar
Merlin   100 kW

100 kW
Posts: 1474
Joined: Jan 29 2013 10:00am
Location: europe

Re: Cycle Analyst V3 preview and first beta release

Post by Merlin » Nov 26 2018 2:55pm

hi@all CA lovers.

3 years ago i build a bike with CA + Pot for regulating Power level from 0 to 1250w.

he had some PAS issues so as i saw its an reaaally old CA Firmware, i updated to the latest now.
PAS has no issue anymore but the POT doesnt work correct anymore.

analog aux limits are set to 0.01 to 4.95v
thats the range of the pot and you can see in the screen while twisting the knob it raises from 0.00v smoothly as you turn to 4.95v
but the %PWR raises from

0.00v - 00%PWR
to
0.05v - 99%PWR

of course the knob is only twisted a single millimeter. the rest of the range is dead.

any ideas?


User avatar
amberwolf   100 GW

100 GW
Posts: 25135
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Power limit max 1kW in high range, 100W in low range?

Post by amberwolf » Dec 06 2018 7:05pm

Teklektik wrote:The 40x multiplier was chosen to provide a reasonable maximum controllable power for PAS vehicles. Your infirmity that limits pedal torque and the exceptionally heavy vehicle fall somewhat outside those CA design parameters. That said:
  1. Since your primary issue appears to be with startup, you should first try the standard adjustment and increase the startup assist as described in the 3.1 ToolTips and Help, perhaps with a value of (-25HW) to (-50HW).


    CA3-1_StartLevelHelpInstructions.png
    .
  2. You can go after the 40x assist scaling limit by increasing PASD->TrqScale to yield a higher apparent crank torque. This will have an effect similar to increasing the Assist Scale Factor but will distort the displayed HW. However, choosing an integral multiplier will make it easy to mentally divide the displayed HW while essentially expanding the Assist Scale Factor. For instance, doubling TrqScale will result in a displayed HW 2x actual while expanding the maximum assist to 2x40 = 80x HW.
Some combination of these two strategies should resolve your issue.
It took me quite a while to get time enough to update firmware / reset to defaults, etc., and then go thru the settings and experiment, but thanks to your help, it's now working well enough as a basic starting point. :) So thank you, very much.

The false power limit thing must've been an invisibly corrupt CA setting, because it was still like taht after the FW flash to beta 3.13b2 (latest available via the utility ATM), and only checking the box to reset settings to defaults at flash fixed it (and subsequently reloading a new setup done from scratch on the utility).

I'm still experimenting with settings that will still let me easily control the power via pedalling, while giving full power from startup as early as possible. I figure that will take a while.

At present with the settings I had to use to get the full startup power via PAS, it's always accelerating if I am pedalling, at any cadence, until it hits the speed limit (20MPH) set in the CA, and then it cuts back to hold me there.

I still need to be able to pedal at low speeds (parking lots, etc.), so I'm still re-reading everything and experimenting with settings to find a happy medium.

But at least now I can, if necessary, just control the trike motor power via pedals for my commutes and such.

Marc S.   100 W

100 W
Posts: 216
Joined: Jul 04 2012 12:23pm
Location: Hamburg, Germany
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by Marc S. » Dec 16 2018 11:51am

Hi Tek or Justin,

I like to put the right front case button on a separate momentary switch at the handle bar*.

I got the temp sensor wire on a separate cable (ex speed sensor cable of the CAv3 DPS), hence I got the yellow wire free to connect to the B2 pad of the 3-pin front case connector.

-can I use a switch to close B2 with GND?
-I've measured the 3-pin plug of the front case, when I press the button, there is a 180 Ohm resistance between the middle connector and B2.

Now I'm confused... :)

Cheers
Marc


*the Cycle Analyst will sit at arms length at the bottom of the front boom of my Milan velomobile, between my thighs.
I can reach it (barely) wile I ride, but would prefer to toggle the CA screens form a handlebar switch.

User avatar
amberwolf   100 GW

100 GW
Posts: 25135
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by amberwolf » Dec 16 2018 7:55pm

Any momentary switch can be used, wired between the center pin on the CA board inside the case, and the outside pin that corresponds to the button you want to activate.

It can be wired in parallel with the existing button, so you can still use the front panel buttons as normal.

If you like, you can use a phono jack mounted to the CA case, with it's tip, ring, and sleeve wires to the CA pins, and then use a phono plug and wire from your switch module, so you can plug / unplug it as needed for servicing of different things.

Post Reply