Cycle Analyst V3 preview and first beta release

Using the "2012" updater utility was successful in restoring the CA3's firmware. I used the v10 *.hex file that updates the firmware and also overwrites the configuration settings. Using same laptop, OS, programming cable - only change is to use the 2012 firmware utility. Justin -- let me know if you want me to run any other tests or do any other debugging, since I do have a 'spare' CA3 and test bench setup, and owe you some blood, sweat and tears for past help received from you.

Logging output from 2012 update to "10" firmware on CA3:

Code:
==================
Now searching for device...

Found CA3r1 (16F1938)
 with bootloader version 0202

Now flashing CA3r1 (16F1938)
 with C:/data/ca_2012/CA3-0p10_firmware.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

Flash Write success!
Now writing EEPROM from 0 to f0 word address

EEprom write success!
Now verifying flash from 20 to 3ae0 word address

Verified Flash OK!
Now verifying eeprom from 0 to f0 word address

Verified eeprom OK!
Reset send successful!
Programming Completed Successfully!
 
rowbiker said:
All flash above bldr erased!failed to find input file!Serial Uploader by Grin Technologies (2011)

This is interesting, it says that the problem was a failure to find the input file. Can you try moving the .hex file into the same directory as the V1.3 setup utility is located and see if it works when the .hex file is selected from there?
 
Justin - the *.hex file was already in the same directory as the executable. After doing a bit more testing, I'm beginning to think the real culprit is my WinXP Pro installation/configuration on the (Toshiba Satellite) laptop I'm using. If I double-click on the *.hex file in the file manager listing, the OS loads the file into Notepad (correctly figuring out it's a "text" file), but *also* throws up an error message that it can't find the file -- obviously an incorrect error display, since it's already found and loaded the file (into Notepad, at least). Rather than wasting your time with this, let's see if anybody else out there encounters this type of error using the Win 1.3 app to burn new firmware. If not, I think we'll just assume that something in my Win OS is hosed, and ignore it. I'll also get another Win laptop and load the 1.3 updater and try the firmware update on the same CA3. If it works, I think we can safely assume the error, as typical, lies entirely on the customer side. :D
 
rowbiker said:
If it works, I think we can safely assume the error, as typical, lies entirely on the customer side. :D

OK Cool, thanks for the quick assurances and here's hoping it's all on you! :p
 
I cant remember... Any problem using 32s 134v with CA?

Thank you with your help :!:
 
Okay I have a guide in CA manual how to adjust throttle response.

I have a few questions about gain features....
My English is not 100% :oops: :oops: , so can anyone explain in simple English what they mean by again and wgain..

Thank you!
 
BoomerChomsi said:
Okay I have a guide in CA manual how to adjust throttle response.

I have a few questions about gain features....
My English is not 100% :oops: :oops: , so can anyone explain in simple English what they mean by again and wgain..
Please see the CA V3 page here for a description of all the settings. The AGain and WGain settings are in the 'SETUP POWER LIMS' section of Setup - both on your CA and on the web page.
 
Thank you CA guroe!
I will have a read session :mrgreen:

Another question do I have to connect + and - leads in right order I mean anode and cathode for temp. sensor kty83-210?
 
BoomerChomsi said:
...do I have to connect + and - leads in right order I mean anode and cathode for temp. sensor kty83-210?
The KTY83 is a polarized silicon device so connect end with a stripe (cathode) to Gnd (black) and the other end (anode) to the NTC signal wire (yellow). No requirement as to connection order when connecting to a powered CA.

thermistor.png
KTY83_Pins.png
I couldn't find a KTY83-210 datasheet on a quick search.

The KTY81-210 is a little different - it's non-polarized and you can wire the connections either way. I believe this part is used in the MXUS3000. I did find an indirect reference that suggested that the PTC resistance characteristics of the KTY83-210 and KTY83-210 were similar so if you have a KTY83 and no datasheet you might try calibrating the CA using KTY81 specifications which are readily available.

KTY81_Pins.png
 
rowbiker said:
Justin - the *.hex file was already in the same directory as the executable. After doing a bit more testing, I'm beginning to think the real culprit is my WinXP Pro installation/configuration on the (Toshiba Satellite) laptop I'm using. If I double-click on the *.hex file in the file manager listing, the OS loads the file into Notepad (correctly figuring out it's a "text" file), but *also* throws up an error message that it can't find the file -- obviously an incorrect error display, since it's already found and loaded the file (into Notepad, at least). Rather than wasting your time with this, let's see if anybody else out there encounters this type of error using the Win 1.3 app to burn new firmware. If not, I think we'll just assume that something in my Win OS is hosed, and ignore it. I'll also get another Win laptop and load the 1.3 updater and try the firmware update on the same CA3. If it works, I think we can safely assume the error, as typical, lies entirely on the customer side. :D
Try putting the program and config file in your user directory- there were some issues with other versions of Windows at one time to do with 'security' at one time. You should also reset that .hex file association- it may have changed or become corrupt.
 
I got a chance this past weekend to ride with the newly-uploaded "P10" firmware.

The good news is that every aspect I use on my bike works as expected. Functionally there were no issues. If anything there was a slight improvement in that it felt as if my drivetrain was not slammed quite as hard when resuming pedaling under PAS-mediated throttle.

But, I did see one persistent bug.

When adjusting the Aux Level on the Watt-Hour, Amp-Hour or Trip/Total Distance screen (the latter when stopped), the Aux Level characters persist when the display returns to the given screen. I'm pretty sure I submitted a bug report about this on the last go 'round, so it looks like this fix didn't make it into P10.

The watt-hour/mi calculation appears to occur when watt-hours > 1 and/or when distance >0.1 mile, or both. So, the blank or incorrect wh/mi display problem for the first 83.333 wh appears to have been fixed.

Suggestion:
Re: the missing "3 dot" battery capacity indication - I'd prefer the missing dot be the top dot at the tip. I.e. Go from full capacity (all dots) to missing the first two dots. At the beginning of discharge I'm not too concerned about remaining capacity, but as I near full discharge I'm checking the remaining capacity more frequently. It appears that the capacity that would otherwise have been assigned to the third dot is added to the second dot (2 dots remaining capacity), with the result that the 2 dots show for 1/16 - 3/16 remaining capacity (at least with my Samsung 29E batteries when using the "LiPo" profile). At a time when I'd like a bit more resolution, the display gives less.
 
mrbill said:
When adjusting the Aux Level on the Watt-Hour, Amp-Hour or Trip/Total Distance screen (the latter when stopped), the Aux Level characters persist when the display returns to the given screen. I'm pretty sure I submitted a bug report about this on the last go 'round, so it looks like this fix didn't make it into P10.
Yep - I can verify the issue on the 'Odometer' (Trip Distance) screen. I can't seem to invoke the other failures at rest or underway. Can you please use the screen names from the CA V3 page to ensure we are testing the proper screens? Also - do these last two cases fail only when underway?

I can't locate the earlier bug report but will submit a new one on this issue. (We are changing problem tracking procedures but this item will go directly to the new tracker.)

mrbill said:
Suggestion:
Re: the missing "3 dot" battery capacity indication - I'd prefer the missing dot be the top dot at the tip.
...
At a time when I'd like a bit more resolution, the display gives less.
This isn't really a loss of resolution, it's an artifact of meshing the available number of discrete SoC states to the graphic. Changing the locale of the 'dot jump' would not really improve resolution but might avoid misinterpreting the rate of fall in the region that is arguably more important. Will discuss...

Thanks - a detailed beta review - as always...
 
teklektik said:
mrbill said:
When adjusting the Aux Level on the Watt-Hour, Amp-Hour or Trip/Total Distance screen (the latter when stopped), the Aux Level characters persist when the display returns to the given screen. I'm pretty sure I submitted a bug report about this on the last go 'round, so it looks like this fix didn't make it into P10.
Yep - I can verify the issue on the 'Odometer' (Trip Distance) screen. I can't seem to invoke the other failures at rest or underway. Can you please use the screen names from the CA V3 page to ensure we are testing the proper screens? Also - do these last two cases fail only when underway?

Display #4 – Energy Mileage
Display #6 – Regenerative Braking

The overwriting of the Aux Level on these screens is subtle. On Display #4, the "l" (el, last character in the word) "Level" (in the phrase, "PAS Level") from the Aux Level screen persists and is juxtaposed to the right of the watt-hours figure, looking like a 1 (one). Easy to miss this one.

On Display #6 the characters "vel" from the word "Level" follow the first "%" sign. This one should be more obvious.

I admit these are cosmetic quibbles, but it should be straightforward to clear the entire display before returning to the underlying screen after making an "Aux Level" change.

mrbill said:
Suggestion:
Re: the missing "3 dot" battery capacity indication - I'd prefer the missing dot be the top dot at the tip.
...
At a time when I'd like a bit more resolution, the display gives less.

... but might avoid misinterpreting the rate of fall in the region that is arguably more important. Will discuss...

When I still have about 1/4 of battery capacity (4 dots) the indicator jumps suddenly to show only 1/8 capacity (2 dots) remaining. This was alarming until I understood why the 3rd from last dot does not show.

Thanks for following up so quickly.
 
mrbill said:
On Display #4, the "l" (el, last character in the word) "Level" (in the phrase, "PAS Level") from the Aux Level screen persists and is juxtaposed to the right of the watt-hours figure, looking like a 1 (one).

On Display #6 the characters "vel" from the word "Level" follow the first "%" sign. This one should be more obvious.
Ahh - got 'em. The failure is directly related to using 'PAS Level' limiting which has a particular message format.

Thanks! On the list for p11. :D
 
Just been setting up the bike..again for the first time..an on going project that I have not really finished.

So started again from the beginning. Flashed latest firmware, Prelim 10 and calibrated throttle, bottom bracket torque figures etc again.


Issue is rapidly changing ( to quick to read) figures on the decimal side of the in and out throttle screen.
This is so great that the Auto throttle cruise function wont stay locked in , even with it set to its maximum variation of 0.5 volts
But this only occurs when the throttle is Open. cThrottle Closed it sits at its resting input of 0.85 volts.
If I go in to Setup and Throttle Input setup screen, the display is stable, so that rules out a fluctuating throttle hall output



Running off separate power for the CA, since i am running 84-100 volt traction pack. So I suspected maybe a noisy output on that , so I have been running the CA off a straight 5s battery pack...no joy
Tried removing the DC-DC power convertor to run the lights...no joy, throttle still fluctuates
I had GPS data logger running too, so disconnected that.
Tried changing data output to 1 or 5hz

Also adjusted the averaging of the data from 0.2 seconds all the way up to 1.2 seconds...again no joy

What ever I try, the 'Open throttle ' figure (as viewed on the diagnostic screen, (one left of main screen)) ..always shows the decimal figures changing too rapidly to read...and hence not stable enough to 'Hold Auto Cruise ' throttle...well not for more than 2-3 seconds. It will often latch then drops out almost instantly.

If I roll along with no throttle applied and motor NOT running...the throttle in figure is stable and slowly rises..the decimals can be seen slowly increasign as the throttle is opened.. Until the throttle Out stats increasing and the motor starts running.

At this point the throttle input decimal figures start to go wild.

Also with throttle closed, and again viewing the Diagnostic screen(one left of main) , the throttle input figures go wild as soon as the e-brake input is activated.

E-brake levers is wired so as to activate CA brake, and also the controller brake to give regen braking.

If I remove the e-brake wire from the controller, so the brake levers now only operate the CA, then I the issue dissappears when throttle is closed, but re appears as soon as throttle is opened, even if the e-brake is still held on via the levers


So interference from the controller? caused by wheel rotation. if the controller is not driving the wheel, so you are rolling, and activate the e-brake line, the the throttle voltage decimals go wild, If the CA is powered on via V+, CA is in Setup Mode, but the controller is powered OFF, and there is no power either on Vex, there is NO interference on the voltage figures, but as soon the controller is powered on, the interference reappears.
it is wheel speed related, more precisely halls, as if I disconnect the halls, then the problem also goes.

I am using a CA-DP with speed pickup from the halls on the controller..not an external wheel speed pickup. so I thought that maybe a wheel sensor rather than hall speed sensor from CA-DP plug would help. I have removed the yellow speed signal wire (pin 5 ) on the CA-DP plug, and the problem of the unstable Throttle Input decimal figures is still there.

Seems the CA is overly sensitive to speed related interference from the controller.

Have tried with the CA power input and controller input and Vex grounds, and DC-DC PSU for the CA V+ power, alternately common ground and isolated and it makes no difference either.



Any idea on how to solve this?
 
Not sure if this is related to my previous post...but when using external power to run the CA, it wont power on until the monitored power is also applied.

I can apply power to the V+ and ground to power the CA, but the CA display won't come one, until Vex too has power. Boot screen comes on, but as soon as it is about to go on to the first screen, the screen goes blank and a re boot occurs.

if I apply traction/ monitored power to Vex, enter Setup , I can then remove power from Vex and the CA stays powered up. I seem to remember something about this previously , but before the screen was scrambled
 
Hey Neil-
Sounds like some annoyingly stubborn issues...

First - the one bright note: the scrolling screen garbage was a bug fixed in p10 - see the Release Notes.

The reboot stuff needs a bit of code drilling...

The throttle stuff needs more attention. I located the wiring diagram we had from earlier discussions about your external supply hookup. This looks to me like a ground reference issue, but that's a knee-jerk reaction and a detailed wiring diagram looks to be needed to sort out what is going on because of the external monitoring, external power supply and grounding stuff. The throttle jitter under high motor current is making me think the CA ground is somehow seeing the voltage fluctuations impressed across the shunt by high motor Amps....

So - I'll PM you the last wiring diagram I have of your rig so you can make or describe any changes - both for the DC/DC converter case and the 'extra battery' case.
If you can get something back to me, it would clear things up a lot. It looks like the 'other side of the shunt' Gnd connection may be doing you in if that design is still in play.
 
As a follow on to the pm i replied to you,

The voltage jitter even occurs with NO controller current flow ... It happens with the controller powered ON but not supplying power,

It will occur if the wheel is turning, ..... umm and brake lever pulled....umm so that means some regen current, but very little, it ocvurs even at slow walking speed.

Will confirm wiring when off the phone, and with regen dissbled
 
I got a chance today to ride with P10 on a bike with a DD hub motor that gets hot when regen braking down a long hill.

I like the new thermometer icon that appears intermittently with the battery icon when the thermistor temperature crosses the lower threshold.

I have a suggestion for improvement: Showing the "mercury" rise in the thermometer icon is a nice touch, but it would be easier to see the effect if the cross-bar rose with the top of the "mercury" column. The one pixel width "mercury" column is too small to see clearly when glancing quickly at the display, and when I'm likely to be getting the motor hot, my eyes are best on the road most of the time, glancing at the display only occasionally.
 
Fixed with help of Justin and Tek

Phase interference on throttle line, fixed with 1 uF cap across throttle and ground .
 
I need help.

My CA3DP is reading -11.5Ah on the main screen. Regen% is at 999.9 and 11.5Ah. The regen on my controller isn't even plugged together. I was running great. Then I cut all my connections and rewired the bike. Everything else seems correct, but the regen problem popped up from nowhere. The CA wire colors matched the colors from my Lyen controller. I think it's the MK. It's the one for the LR small block. The only other difference I can find is at rest I reed zero Watts. It used to read 5W.

Could I have reversed the S+ and Sp connections?
 
Back
Top