Cycle Analyst V3 preview and first beta release

[k.ewin]

Your solution is working, by changing the separator in intl.cpl it's solve the problem.

If the CA setup utility does not work with a comma as decimal separator in Windows settings this is clearly a bug and should be fixed. Many European countries use a comma. I am happy that changing the separator in intl.cpl did help you to work around the issue.

 

[teklektik]

I do not understand why my test drive worked perfectly and the next day the settings were corrupted. I have not changed anything in between.

That was a characteristic of the 3.1 firmware and shouldn't be showing up now... so not the best news. I'll take a look at your files straightaway. I'm thinking this was all part of the comms hangs and incompleted settings updates.

In any case, I have sent you an alpha version of 3.12 that should fix the comms hangs if you get a chance to test it. That said, if you're happy with your CA now and just want to ride instead of test, that's good, too

...I am planning on visiting the LBS that I purchased the CA3 from today. I'll see if he will exchange the cable for another.

I meant newer than 3.1B13 - which by the way is working just fine for what I need. Truly enjoying the Sempu torque sensor and all the PAS parameters available for the user.

It's a little unclear what's up on your machines, but if Setup utility is working for older Firmware then it doesn't look like the cable...
Sent you 3.12a1 as well to see if that helps out the comms problems...

 

[teklektik]

Your solution is working, by changing the separator in intl.cpl it's solve the problem.

If the CA setup utility does not work with a comma as decimal separator in Windows settings this is clearly a bug and should be fixed. Many European countries use a comma. I am happy that changing the separator in intl.cpl did help you to work around the issue.

Yep. I'm writing a ticket about it right now for the Setup Utility guys.
Thanks all for pointing this out and getting a workaround sorted...

 

[BoomerChomsi]

I know I am so stupid......

I have accidentally reverse power input, now CA3 just show blank screen (only light is lit). After opening CA3 I can feel Q1 is very hot! :x So I assume this compoment is broken? Can anyone tell me what kind of transistor is this? also should I look for further components for damaging?

Also is there any possibility to test CA3 without connecting BAT power to V+ connection?

Thanks you!

 

[amberwolf]

does this help?

https://endless-sphere.com/forums/viewtopic.php?f=2&t=19540&hilit=cycle+analyst+repair&start=25#p678269

 

[BoomerChomsi]

Thank you very much Amberwolf! 8) :lol:

So before I will blow more things:

1.
For testing I should supply +12 volts to pin LED+ (that is underneath Q1 transistor?
And supply ground to one of the GND connections and not LED- connection?

2.
Also I can supply battery voltage 58V at V+ and supply 12+ at LED+ at the same time?
This method will shows BAT voltage as before?

3.
When removing this Q1 transistor and using method number 2 (above) this will works.

Aslo can you tell me which is G, D and S at this layout?

Thank you

 

[BoomerChomsi]

Q1 is removed from pcb.
Applied 12V to led+, still blank screen (only screen led is lit).
There is no 5V present at ca3 pcb.

What type is 5V regulator?

 

[BoomerChomsi]

Nevermind, I have throw the CA3 in the bin!

 

[BoomerChomsi]

I have a new CA3, no to prevent my stupid behavior again...

How can I protect CA3 from reverse polarity?
A 3A fuse didn't helped.... :lol:

Thank you!

 

[Silverexpress01]

Hello,

Truly enjoying my new CA3-DP. Over the weekend I installed a Sempu torque sensing BB. The CA3-DP has ver. 3.1B13 of the firmware, and it does not have Sempu listed as a PAS torque sensor. I've been attempting to connect the CA3-DP to a PC, so that I can upgrade the firmware and have been receiving the following message in a popup,

"Unable to communicate with the Cycle Analyst in a timely manner. Please make sure the device is properly connected. Please cycle power to the device and try again.

The FTDI drivers seems like they've installed correctly. No warnings from Windows 10 stating they did not. When I plug the TTL-USB cable into the PC, it get's assigned com3. In the Setup Utility, it shows up as "\\.\com3" in the drop down list. I have the executable CA3 setup utility file (*.exe) and its folder directly residing in C:\.

I've tried three different laptops running - XP, Win 7, and Win 10. Taking care that I installed the correct FTDI drivers. All show the same pop up window. I also went thru the trouble of making my own TTL - USB cable with an FTDI board I was using for an Arduino and I still got the same message.

Did I incorrectly install the software? I run the executable as an Administrator too...still I get the issue.

Thanks in advance.......

by the way, I have a second unit - CA3-DPS and it too cannot connect. Also where can I find the Sempu pre-configured setting values? I'd like to enter them manually for now if possible.

Ok. I just wanted to put closure on this posting of mine from a couple weeks ago. I ended up buying a new FTDI TTL-USB board from an electronics store (known as Microcenter) near my house. What I did differently was swap the TX and RX lines, and it worked! I was able to upgrade one of my CA3's to firmware 3.12a1. Very happy to say the least. I'm also now able to edit and save all my settings. Woohoo!

 

[teklektik]

Ok. I just wanted to put closure on this posting of mine ...
...What I did differently was swap the TX and RX lines, and it worked!

Thanks for getting back on this - glad you were able to resolve this as a hardware issue.
That said, there does appear to be a comms issue when running 3.1 and 3.11 firmware (e.g. hangs doing a CA read/write from Setup Utility 1.54). This is being actively pursued.

 

[danielrlee]

Hello,

Truly enjoying my new CA3-DP. Over the weekend I installed a Sempu torque sensing BB. The CA3-DP has ver. 3.1B13 of the firmware, and it does not have Sempu listed as a PAS torque sensor. I've been attempting to connect the CA3-DP to a PC, so that I can upgrade the firmware and have been receiving the following message in a popup,

"Unable to communicate with the Cycle Analyst in a timely manner. Please make sure the device is properly connected. Please cycle power to the device and try again.

The FTDI drivers seems like they've installed correctly. No warnings from Windows 10 stating they did not. When I plug the TTL-USB cable into the PC, it get's assigned com3. In the Setup Utility, it shows up as "\\.\com3" in the drop down list. I have the executable CA3 setup utility file (*.exe) and its folder directly residing in C:\.

I've tried three different laptops running - XP, Win 7, and Win 10. Taking care that I installed the correct FTDI drivers. All show the same pop up window. I also went thru the trouble of making my own TTL - USB cable with an FTDI board I was using for an Arduino and I still got the same message.

Did I incorrectly install the software? I run the executable as an Administrator too...still I get the issue.

Thanks in advance.......

by the way, I have a second unit - CA3-DPS and it too cannot connect. Also where can I find the Sempu pre-configured setting values? I'd like to enter them manually for now if possible.

Ok. I just wanted to put closure on this posting of mine from a couple weeks ago. I ended up buying a new FTDI TTL-USB board from an electronics store (known as Microcenter) near my house. What I did differently was swap the TX and RX lines, and it worked! I was able to upgrade one of my CA3's to firmware 3.12a1. Very happy to say the least. I'm also now able to edit and save all my settings. Woohoo!

Do you have further details of the FTDI TTL-USB adapter purchased? I've previously tried to make an emergency programming cable for a CAv3, but was never able to successfully communicate with anything other than the official Grin programming cable.

 

[Silverexpress01]

Ok. I just wanted to put closure on this posting of mine from a couple weeks ago. I ended up buying a new FTDI TTL-USB board from an electronics store (known as Microcenter) near my house. What I did differently was swap the TX and RX lines, and it worked! I was able to upgrade one of my CA3's to firmware 3.12a1. Very happy to say the least. I'm also now able to edit and save all my settings. Woohoo!

Do you have further details of the FTDI TTL-USB adapter purchased? I've previously tried to make an emergency programming cable for a CAv3, but was never able to successfully communicate with anything other than the official Grin programming cable.

Sure thing. Attached is the picture of the packaging. At this point the FTDI board is wrapped in electrical tape since I didn't have shrink tubing large enough to cover it. The stereo cable I got from an old headphone. Make sure you set the input voltage to "5v" with the toggle switch. Here is the link to the product - "Inland" is the house brand of Microcenter.

https://tinyurl.com/ybj49ycr

You will also need a usb cable with the following terminating ends - Type A (PC) to a mini B (FTDI Board) and a single row of 5 pin headers.

 

[BoomerChomsi]

Anyone a answer regarding my question about reverse protection?

Thanks 8)

 

[teklektik]

v3.12b1 Released

Thanks to member k.ewin and others, an issue came to light in versions 3.1 and 3.11 that results in the CA hanging with "PC Connection..." on the screen and Setup Utility reporting a timeout. Version 3.12b1 should remedy this problem. There are no other changes so this is identical to 3.11 except that 'Update Firmware', 'Read CA', and 'Write CA' shouldn't hang.

Without getting into details, the tricky part here is that the bug relates to the PC on which the Setup Utility runs so the original 3.1/3.11 hang problem may or may not show up. Soooo - to get this tested on as many different PCs as possible -- even though this new release doesn't really offer any new features or goodies, it would be great if beta testers could load up this new version, try a few CA read/writes/updates after 3.12b1 is initially flashed, and let us know if you see any hangs.

Apologies about this one folks - it slipped by in 3.1/3.11 testing because of the PC dependency...

So - press "Get New Firmware", check both boxes in the pop-up, and the new firmware should appear in your Setup Utility. Flashing it will not change existing settings...

 

[interrupted]

so, this might be a stupid question but i am a bit confused about the pas configuration possibilities with "stupid" magnet pas discs

currently i am using a 12 pole pas device in basic mode with rpm downscaling, which works but i think this maybe can be improved. i stumbled across the pas mode torque within the firmwares help file, without further knowledge i would guess that this mode will only work with real torque sensor devices? but since there is no special hint in the docs that this is only usable with torque sensors i am curious if its compatible with stupid pas dics, too?

i tried enabling torque mode, switching to torqpas and than i was lost because it made no sense for me to stay with pas basic scaling and for torquepas scaling its required to measure the human watts as far as i understand?

anyway, maybe someone can give me an advice (probably even with some known setting values to start with) how i could improve the pas behaviour. my target would be decreasing power with increasing speed, perfect situation would be to accelerate up to the speed limit and stay within that speed limit range (+-1-2kmh) while still pushing some low power levels to the motor (80-120w should be plenty for me)

my configuration:

12 FET Infineon Controller
12T Rear Mac Hub

i attached my current configuration, would be great if someone with more knowledge about the CA could review it. currently i am running the latest 3.11 firmware

another issue i am facing is, that even when i hide all configuration options via the windows tool the CA still shows all the topics of the hidden options. its just removing the options in the sub menus. can i somehow make them completly invisible? maybe i remember wrong, but i somehow think that this was possible in older firmware releases

 

[vmbray]

so, this might be a stupid question but i am a bit confused about the pas configuration possibilities with "stupid" magnet pas discs

You can use a cadence sensor alone with pas but not in torque mode. The settings are a little confusing and somewhere I know Grin answered me about this but there is a way to have more assist at lower cadence. Try the throt pas mode and play with the settings. Throt pas will give more power at higher cadence with the default settings so it's like launch mode, you pedal, it helps, you pedal faster, it helps more, etc until you're at max throttle at least for me that's what happened. I got a torque sensor working finally so I didn't keep working on the cadence setup.

 

[vmbray]

Having a weird issue with the trike setup. It's the prodecotech sensor and a cadence sensor wired together to give a valid torque reading and cadence and has been working great. Yesterday it started dropping out at times and it definitely had valid human watts and cadence signals while not working. The battery is a 52 volt em3ev pack and was not fully charged but was showing around 52-52v under load and 56 at idle. The low volt cutoff is set at the default 19v because I overlooked that the first go around. What's weird is that charging the battery greatly improved this, I only saw one cutout for a few seconds while testing (could have just been laggy pas but it was a few pedal rotations and I had time to start down shifting so not sure) and my wife had no issues.

This has to be in the CA because while it was not working the throttle output showed .99v even while I had valid human watts and cadence showing. Also it was very dramatically better after charging but the low volt cutoff is set so low that it doesn't make sense.

 

[teklektik]

...i would guess that this mode will only work with real torque sensor devices? but since there is no special hint in the docs that this is only usable with torque sensors i am curious if its compatible with stupid pas dics, too?

i tried enabling torque mode, switching to torqpas and than i was lost because it made no sense for me to stay with pas basic scaling and for torquepas scaling its required to measure the human watts as far as i understand?

Human Watts is equal to RPM x TORQUE so with no torque device on the TRQ input the CA will read 5V and calculate an imaginary torque reading. Applying this to your cadence will result in some value for Human Watts. Is this 'working'? Well, yes and no. The calculation and any Aux PAS Assist level scaling are still being computed correctly, but the 'torque' is just a fixed value.

If you examine "6.6 PAS Assist Control without AUX Pot – DIY Virtual Torque Sensor" of the Unofficial Guide, this is explained in greater detail and you will see how a 'torque' level pot might be used with a simple PAS wheel to obtain variable PAS level assist without using the AUX input. (That strategy was devised before the advent of DigiAux allowing a second AUX input and so is a little outmoded these days...)

my target would be decreasing power with increasing speed, perfect situation would be to accelerate up to the speed limit and stay within that speed limit range (+-1-2kmh) while still pushing some low power levels to the motor (80-120w should be plenty for me)

You are actually asking about two different things here....

1. PAS with decreasing assist using a simple PAS wheel:
   This is pretty easy. Set up a high Start level and use a negative scale factor. The CA assumes a minimum or 50rpm and scales for every rpm above that cadence, so the following example will give you 600W off the line reducing to no assist at 80rpm. Adjust settings for your application and fiddle PASRate to get a smooth getaway :
   
   
   
2. PAS with Speed limiting:
   This is really just a tweak on whatever PAS setup you want to use and is essentially PAS-activated cruise control. The idea is to set up your PAS for a surplus of power so that the CA is trying to assist you to a high rate of speed. Then use the AUX input to limit either speed or ThrO to limit what PAS can actually achieve. Digital buttons are handy here since you can just long-press the UP button to quickly remove the speed/ThrO limit. Although using Speed Limiting for the control will set/maintain your speed accurately much like an auto cruise control, you may need to adjust the PID controller a bit to get smooth speed limiting. ThrO limiting is the lazy-man's approach and simply limits the applied throttle (no PID adjustment required). This works pretty well for generally flat terrain but is really a sort of throttle lock so it doesn't do hills that well. Depending on how you set up the PAS, you can pedal as hard as you want - or not at all, and the PAS will make up the difference up to the speed limit.

another issue i am facing is, that even when i hide all configuration options via the windows tool the CA still shows all the topics of the hidden options. its just removing the options in the sub menus. can i somehow make them completly invisible?

This is a pending task that will likely appear in 3.2.

 

[teklektik]

Yesterday it started dropping out at times...
...my wife had no issues.

My initial guess is that your Zero Throttle Threshold is too small. A small motion of the throttle is kicking in the throttle the tiniest of bits which is enough to kill PAS. Perhaps your wife didn't have problems because she has a death grip on the throttle holding it firmly off...

Verify that the ZERO threshold is 0.15V or so higher than the actual minimum throttle voltage as displayed on the THROT IN prefix screen.

 

[teklektik]

v3.12b1 Released

Hey beta testers!
We've pretty much finished testing 3.12b1 in-house -- with good results . But as mentioned above, the comms issue addressed by 3.12b1 is related to the platform on which Setup Utility runs so all the different machines you guys have gets a lot more test scenarios in the mix. If you've been thinking about testing and haven't gotten to it, load up 3.12b1 and bang away for a bunch of CA reads and writes - no need to actually change any settings. You might also try re-updating 3.12b1 over itself repetitively as well. If it doesn't hang all is good (typically "PC Connection..." on the screen and a Setup Utility timeout).

Anyhow, this is a last call for hang reports. If nothing gets reported by Tuesday, this will go live as the production release.

As always - thanks for your support!

 

[Adamlivi]

I just tried to update firmware to 3.12b1 and failed. Red screen cape up saying write failure at 2% with connection issue. left as is fr 5 min and then disconnected and restarted CA, it seems like it is now "bricked" with only a lit and textless screen when powerd up.

Any help or is this a "send in" repair?

uggghhhh!!!!

Thanks

 

[Hellemyr]

Hi.

New guy here. Having finally bought and put together my first ebike last weekend, I am having some issues I need som help with.

I got a CA3, Phaserunner and a mxus motor bought at ebike.ca, powered by a 18S3P 18650 battery.

I am having some of the same issues as Silverexpress had, just some post above here, "Unable to communicate with the Cycle Analyst in a timely manner. Please make sure the device is properly connected. Please cycle power to the device and try again.

I dont have a USB -> TTL cable, but using a arduino as a TTL converter, by connecting reset pin to ground.

In the CA tool on my computer, if I try to READ CA, the CA goes into PC CONNECTION.. and stops there, before i get the message unable to connecet on my pc. CA stays in PC CONNECTION... mode until i cycle power to it.

If i try to click on the update firmware, nothing happens on the CA display it just sits in the normal display mode, and i get this message .

checking the reset setting to default, as suggested gets me this

When i cycle power to the CA by unplugging the battery, and plugging it back in, the CA display lights dimly up, as the display gets power, but no text is displayed, and it stays this way until i cycle power it it again.

Does anyone know whats going on, and what i can do to fix this?

I may also add that I had no luck communicating with the phaserunner aswell. Seeing that it solved the problem for Silverexpress to buy a new adapter, i went out to my local electronic gadget store to get a USB -> TTL cable, but they did not have any in stock. But i have previously used the arduino method to communicate with a RC drone controller without any problems, so the arduino adapter shudent be the problem.

 

[teklektik]

...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.

 

[teklektik]

I dont have a USB -> TTL cable, but using a arduino as a TTL converter, by connecting reset pin to ground.
...
I may also add that I had no luck communicating with the phaserunner aswell. Seeing that it solved the problem for Silverexpress to buy a new adapter, i went out to my local electronic gadget store to get a USB -> TTL cable, but they did not have any in stock. But i have previously used the arduino method to communicate with a RC drone controller without any problems, so the arduino adapter shudent be the problem.

As mentioned in the recent posts, there is some timing issue in play with the Setup Utility and it's quite possible that your non-standard comms solution is upsetting things - your DIY solution seems to fail with two different products.

Discussing details of how to avoid purchasing the proper Grin cable isn't really the topic of this thread and I can only recommend that you get the proper cable or settle in and pound away on the usual DIY cable solution Science Project of mismatched Rx/Tx wiring and peculiar driver anomalies... Barring this recent timing issue, the CA stuff isn't really that fussy - I have flashed CAs using a modified Lyen/EM3EV controller cable with a Prolific chip set. Other non-Grin cable solutions can clearly be made to work with a dose of experimentation if saving a few bucks is important...