Cycle Analyst V3 preview and first beta release

snellemin said:
I had retry the boot fix a few times before it went through(where you can enter the numerical value).
but after that it says something like "success" and when you cycle power it starts from the beginning, correct? and you have to flash the regular firmware again to get it going.
 
i'm just about to give up. i just can't get a satisfying riding experience with a smooth throttle response and steady power output.
as you may know, it's a 2.2kw middrive driving the wheel through the rear gear cassette. so power (torque) demand varies a lot depending on the gear set. i have a 3 way switch installed (like in the CA manual, 3 resistors that set POWER limit to three defined levels) so i can switch down power easily.
example: i ride up a hill. throttle is somewhere in the middle position, and throttle in/out voltage doesn't change - checked at the '#11 diagnostic screen'. still the amps jump around like crazy. from 12a up to 20a in a fraction of a second. if i apply WOT then it accelerates quite ok (still not smooth) but starts cutting/applying power when full speed is reached.
i tried adjusting only ONE value which is Pgain. i trying from 100 (undriveable) to 3 (quite ok, but still not smooth, and for the cost of super slow power application).

if i remove ALL limiting it rides quite nice. but this totally make all PRESETS and AUX SWITCH useless.

i made a screenshot of my settings. maybe i follow a totally wrong path and try adjusting the wrong parameters. i just can think that i'm the only one with a mid drive that has is not able to adjust the settings to be satisfying.

these are my settings:


any help is really appreciated - just to keep me from banging my head to the wall :)
 
please somebody watch this video and tell me what to do ?!? i start to believe that this must be something different but the CA. maybe it's the controller that is not able to hold a steady current? it seems to overshoot permanently.
and it got worse. even with all limiting disabled, only WOT was nice to ride. so i shifted to 1st gear and rode WOT (25km/h) to ride next to my daughter on another ebike.
there is some stuttering present from the beginning, but it's worst at 0:13 to 0:20. you can see that wattage is quite steady at 1000w (of course this includes the CA's averaging)

[youtube]lgjSuiz-32Y[/youtube]
 
I saw that you have the power limited to 600W, but your Amp limit at 99A. Increase your watts limit to 4000W and see what's up.
 
snellemin said:
I saw that you have the power limited to 600W, but your Amp limit at 99A. Increase your watts limit to 4000W and see what's up.
there are 3 presets. 600,1000 and 2200w. even if is set amp to 99 and watts to 9900 it doesn't work if throttle is in the middle range. and if i lower power, be it with a preset or via 3-way switch, it's getting worse.
i did some research for the last few hours, and it seems it's not the CA, it's the controller's programming itself. it's an EB309 with additional shunt added and battery amps set to 40a and phase to 80a. that may not be ok. so i will play with phase amp a bit more, or remove the second shunt. this may cause issues.
i'm sorry if this is not really a CA issue and i posted wrong in this thread. it was not in purpose, and i'm still not sure what the problem is.
 
did you tried the wgain setting to a lower value?

my wgain setting is 20. and a buddy of me with a 306 had a similar problem from a fresh build.
you are limiting your system like me only on wattage.
so the again setting will not come in play.

just a try.
 
izeman-
Sorry to see your difficulties. Here are some thoughts:
  • You need to verify that all is well without the CA connected. I thought this was out of the way in an earlier post, but the question seems to have come up again.
  • Assuming this is a CA issue, the real cause of the limiting problem is unclear. You need to watch the Diagnostic Screen to see which Limit Flags are in play. This will tell you immediately which parameters to adjust.
  • I do not understand your Aux Pot setup. You seem to be using three presets and have the Aux Pot limits set to (1V, 4V) as if you want the 3-position switch to handle presets - but the AuxPot Function is set to 'Limits' not 'Presets' so the switch is doing something wonky with the power limit in the present preset instead of switching presets. This is not the cause of your problem, but it's confusing...

    (It may be that you have resistors, etc for your switch that make this work the way you want, but the (1V,4V) settings for min,max look suspicious if you know what I mean...)
Truth be told, there does not appear to be a lot of experience with the V3 and high powered mid-drives. As mentioned in an earlier post, this looks like an area where the varying torque and fixed limit/gain settings may not get along too well as the PI controller characteristics have issues addressing vastly changing drivetrain characteristics - dunno. Justin is successful with the StokeMonkey drive but it isn't as powerful as your setup.

So - first steps - see if the hints above help clarify the situation. With a little more information on the table a clearer direction for investigation might be possible.

BTW - your setup looks fine except for the strangeness with Aux Pot. I fully expect that the 'W' Limit is the issue and your WGain is the adjustment that is needed (I think you have been adjusting this and just calling it Pgain instead.) I'm puzzled by your report of surging even when WGain is as low as 3 (!) which is why I want to see if other flags are in play for some reason....
 
thanks for your support. this will be my last post about this issue in this thread as i think it's not a CA issue. if you may want to help i will continue in my build thread http://endless-sphere.com/forums/viewtopic.php?t=55215
i had a broken fet (had one before) and replaced it with a 3077. only then i realized that all (but the one replaced before and the now replaced one) other fets are 3307. i don't know about those fets but i could imagine that this could cause the issues i see.
i also see a no load current of 7a (if my CA reads correctly - will check with a turnigy watt meter as well). i can't imagine that there is a hall/phase combination issue as the bike runs nicely at very low throttle and nicely on wot.
so if you like to support me please join my other thread and thanks for everyones help so far!
 
Something seems to have gone wrong when I tried to flash my CAV3 with the latest firmware. I was using my Macbook running Parallels with Windows 8.1 to program it. The connection seemed to drop during the update process and now the Cycle Analyst powers on (back light come on) but the screen is completely blank and the Firmware Update Tool isn't detecting it.... Is my Cycle Analyst dead???

Any help would be GREATLY appreciated!
 
izeman said:
you should be able to solve that with the old updater program. it can flash the ca even if it's not recognized by v1.2 updater.

That's sounds very promising!! Thanks Izeman. Any idea where I would get the old updater program?
 
izeman said:
contact ebikes.ca. i'm out of office so i can't send the file now. i could do it tommorrow.

Justin (ebikes.ca) got back to me and seems to think the same as you Izeman - good advice. He wasn't in the office so told me to go through the early forum posts in this thread for the software - it should be here somewhere.

I'll post it up here again if I find it but if I don't (and you had it handy) it would be great to get it from you tomorrow.
 
liamcaff said:
Justin (ebikes.ca) got back to me and seems to think the same as you Izeman - good advice. He wasn't in the office so told me to go through the early forum posts in this thread for the software - it should be here somewhere.
This is a well-known issue that has come up here before. For certain laptops, I have never been able to use the any of the later updaters that are automated and lack the extra operator buttons to sequence CA discovery and download separately. The timing seems too tight in the later automated versions and the comms appear to get confused and hang. Since you seem to be VMing Windows on a MAC, you may be encountering similar I/O timing issues. I use the last non-automated version that was released back in Oct 2012.

Those original downloaders were never posted in the this thread - only on the ebikes.ca site and are long gone. So - here's the one I use:

View attachment CA3_Update_Tool_20121016.zip
Enjoy!
 
teklektik said:
This is a well-known issue that has come up here before. I have never been able to use the any of the later updaters that are automated and lack the extra operator buttons to sequence CA discovery and download separately. The timing seems too tight in the later automated versions and the comms appear to get confused and hang. Since you seem to be VMing Windows on a MAC, you may be encountering similar I/O timing issues. I use the last non-automated version that was released back in Oct 2012.

Those original downloaders were never posted in the this thread - only on the ebikes.ca site and are long gone. So - here's the one I use:


Enjoy!

Thanks a million! Have visitors for the rest of the weekend but looking forward to testing this out asap! I'll let ye know how it goes.

Cheers!!!

L
 
An updated version of the Unofficial User Guide is available for download here.

The primary revisions reflect issues discussed in recent posts:
  • use of DC/DC converters to power the CA (voltage ranges, settings) and
  • CA RShunt calibration using a constant current source.
 
teklektik said:
Those original downloaders were never posted in the this thread - only on the ebikes.ca site and are long gone. So - here's the one I use:

Enjoy!

Thanks for posting this Teklektik. I've been having a hard time actually trying to find where we had this earlier build archived on our servers so this is a bit of a saver. I've updated the ebikes.ca page to include a permanent link to this 2012 version of the firmware update tool too to help with others in the same situation:
http://www.ebikes.ca/downloads/CA_Update_Tool_2012.zip

izeman said:
you can program the v2? i thought only grin can do that ...

To clarify, any of the V2.3 CA's can be reflashed with the same utility as the V3 devices, but a V2.2X CA or earlier uses a different microchip and lacks an onboard bootloader, so they can only be reflashed with a dedicated microchip programmer. Even though they can be reflashed with the same tool, the V2.3 CA chips have half the memory as the V3 CAs, so you can't succesfully load the V3 CA software onto a V2.3 device. It won't work after that, but if you do it you can reload the V2.3 firmware and all will be fine again.
 
Justin, where can I download the 2.3 firmware?

I've attached it here. We haven't yet updated the V2 CA webpage to have a list of downloads, as right now there is only one firmware version in place. But in the future we will aim to produce a V2.4 and then V2.5 firmware build for these parts that will incorporate some of the CA3 features, and which will also allow the use of the CA setup utility for configuring all the parameters from a computer.
 

Attachments

  • CA-DPS V23 (chng to .hex).txt
    34.2 KB · Views: 73
justin_le said:
Justin, where can I download the 2.3 firmware?

I've attached it here. We haven't yet updated the V2 CA webpage to have a list of downloads, as right now there is only one firmware version in place. But in the future we will aim to produce a V2.4 and then V2.5 firmware build for these parts that will incorporate some of the CA3 features, and which will also allow the use of the CA setup utility for configuring all the parameters from a computer.


Thanks Justin. My CA is back to normal!
 
Hi guys,

So I finally managed to try out the older version of the Cycle Analyst Update tool... My CAV3 is being detected but I keep getting the following error.
This is a copy and paste of what happens: (Any ideas??)

Found CA3r1 (16F1938)

with bootloader version 0202

Now flashing CA3r1 (16F1938)

with C:/Users/liamcaff/Documents/CAV3 Software/CA Firmware Flash Tool V1.2/CA3_Prelim6_NoCal.hex...

Erasing ff pages from 20 to 2000

Erasing d7 pages from 2000 to 3ae0

All flash above bldr erased!

file opened

buffers cleared

hex file loaded

Hex file defines memory unreachable. Truncating at address 0x75c0

Now writing from 20 to 3ae0 word address

Timeout!Flash write command receive failed!

write fail!
 
My flashes sometimes fail but a retry usually does the job. Problem is that there is no hardware handshake for synchronization so if a byte gets missed in whole or in part, the download fails. Since we know you are already having timing issues, this problem isn't really too surprising. On the good side, you have gotten around you original failure... :)

Give it another try or two.
 
Back
Top