Cycle Analyst V3 preview and first beta release

John Bozi said:
Is the cave plug and play w/ the lyen eb3xx boards/controllers?


throttle doesn't work through cav3 (as it did when running grinfineon controller)
Click to expand...

I guess it depends on how old your controller is. and if you followed the guide or not when setting up

I have a Lyen 6 FET, about 6 months old and it works just fine direct plug in with the CA v3
I determined it worked OK first by following the test that you run to determine throttle wiring, as per telektiks User Guide
The Unofficial User Guide is available for download here.


Extract from version 3-0p6- j of the guide

1. If the CA interface type is unknown (the controller is not of recent manufacture or is not clearly labeled 'Large Screen Compatible') then perform the following test to ascertain the type:
a. place the bike on a stand or invert it onto handlebar/seat so that the motor can run safely,
b. plug the CA into the controller CA-DP interface and a throttle into the controller,
c. power up the controller and CA,
d. enter Setup, use all default CA settings, and configure:
e. ThrO->MaxOutput = 2V and ThrI->CntrlMode = Off(WOT),
f. exit CA setup and be prepared to quickly power down the controller if the motor starts.
g. Power down the controller.
h. If the motor powers up spontaneously leaving Setup without manually applying throttle, then the controller interface is Large Screen Compatible, otherwise it is V2 Compatible.
i. Power up the controller and restore ThrI->CntrlMode = Passthru and ThrI->CntrlMode = 0.8v
Click to expand...





throttle does work through controller but CA won't show amps etc.

thanks
Click to expand...

If the CA is not reading amps then there is definitely an issue.....from memory, there is no difference between all the version CA plug where the shunt monitor wiring is concerned....middle pins...3 and 4 on all the 6 pin CA-DP plugs...So if the current monitoring works with a V2 it should work exactly eh same with a v3.

if not then there is a problem.
 
Fudge - another doc bug. Proofreading your own stuff is never a good idea... :(

Here's the way that section should read:
<deleted>
Will re-post a revised doc ASAP.

EDIT - okay - that was not a bug - I was dazed and confused. There was an error there as mentioned below, but it was harmless...
 
What have you had to change? can't see a difference...

I used those instruction and it all worked fine.
 
teklektik said:
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.
Click to expand...


Have tried all of the following .HEX files multiple times.... still the same error....

CA3_BootFix_V01.hex
CA3_BootloadFix_V01.hex
CA3_Prelim6_NoCal.hex
CA3_Prelim6_NoEeprom.hex

Any more ideas?
 
izeman said:
try flashing b19 firmware first. i had troubles to flash any firmware. to my findings some combinations work. some don't.
good luck! it WILL work.
Click to expand...


Can't find the b19 firmware file - any idea where I can download it?

Still no luck with the other firmware files...
 
teklektik has a link on his post back at page 67 of this thread..but clicking on it takes you to the Grin Tech page, but no firmware.looks like the linked page has been changed...sorry thought I had found it for you .

Lots of Google advanced links, but th slinks all end up as dead ends.
sorry
 
NeilP said:
teklektik has a link on his post back at page 67 of this thread..but clicking on it takes you to the Grin Tech page, but no firmware.looks like the linked page has been changed...sorry thought I had found it for you .

Lots of Google advanced links, but th slinks all end up as dead ends.
sorry
Click to expand...


Thanks for looking. Yeah I've been hitting the same dead links.... Lots of references to the b19 firmware but no working links....
 
NeilP said:
What have you had to change? can't see a difference...

I used those instruction and it all worked fine.
Click to expand...
  • EDIT -
    Ya - you are 100% correct. I was confused. Sigh....
There's also an error in restoring the default settings in the last step.

This section and many many others were reworked in the last release to try to make things clearer and to make the setup more bullet-proof for folks without much ebike experience. So - as usual - if it gets touched, it can get broken....
 
ummm..I'll have to go back and test again.
but it all seems to work correcly with V3 and my controller. I guess that using the wrong setting, just causes the Th out to go to max as opposed to 2 volts..as I guess is intended.
 
An updated version of the Unofficial User Guide is available for download here.

The error discussed in the posts above is pretty important since it can alter the whole installation process - so yet another doc release... :roll:

There is also a bit of re-wording in the section about using DC/DC converters to power the CA. Nothing incorrect there, just pretty poor form.
 
izeman said:
b19 here you go:
Click to expand...

Thanks Izeman,

Still no luck I'm afriad.... :(

Here's the text from the Firmware Updater Software:

Found CA3r1 (16F1938)
with bootloader version 0202

Now flashing CA3r1 (16F1938)
with C:/Users/liamcaff/Documents/Old CAV3 Software/CA_Update_Tool_2012 - Justin/Firmware/CA3B19.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
Now writing from 20 to 35e0 word address

Timeout!Flash write command receive failed!
write fail!




After I power cycle the Cycle Analyst I get a slightly different error at the end:

Found CA3r1 (16F1938)
with bootloader version 0202

Now flashing CA3r1 (16F1938)
with C:/Users/liamcaff/Documents/Old CAV3 Software/CA_Update_Tool_2012 - Justin/Firmware/CA3B19.hex...

Erasing ff pages from 20 to 2000
Timeout!Erase command receive failed!
Erase failed!



Am feeling pretty stuck now... I tried uploading firmware b19 about 10 times there
 
liamcaff said:
teklektik said:
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.
Click to expand...
Have tried all of the following .HEX files multiple times.... still the same error....

CA3_BootFix_V01.hex
CA3_BootloadFix_V01.hex
CA3_Prelim6_NoCal.hex
CA3_Prelim6_NoEeprom.hex

Any more ideas?
Click to expand...
Sadly, no.
You are failing with this error:

liamcaff's updater said:
Timeout! Flash write command receive failed!
Click to expand...
This indicates a comms or sync error. The older updater is really the only available option to address this problem. You will need to use a different host for the updater to get the app and CA talking. I suspect the VM environment is to blame - you probably need more (virtual) processor speed.

Trying to load different hex files does nothing to address the underlying problem which is with the updater/bootloader NOT the program being loaded.

Time to run the updater natively - borrow a laptop or bring your CA to a friend with a Windows PC...

Also - it's hard to tell what's in your CA now, so in the end you will need to either do the classic 'double flash' or use the newer 'CalOverWrite' procedure with the CA Settings Setup App as mentioned on the Grin Tech V3 page and described in this post. This accomplishes the same thing by forcing update of the settings that are normally not updated using the Settings tool (i.e. no need for double flash with B19).
 
teklektik said:
...I suspect the VM environment is to blame - you probably need more (virtual) processor speed....
Click to expand...
that may be the best way to solve it. i can confirm that it DOES work. i tried several usb2com converter and different computers until it worked for me.
 
Adjusting Firmware Updater Max Baud Rate

izeman said:
what did you set your com ports (max baud in old updater) speed? i was told by grin that it could help to reduce it to 9600 baud as more complex commands sent to the bootloader my cause problems at higher speed.
Click to expand...
Ya - reducing the baud rate is a good strategy in these cases, but I was under the mistaken impression that the CA was locked in at 9600 baud.

I just loaded up the old updater and saw the (forgotten) Max Baud field. Oops, My Bad.
So - I tested this and it definitely slows the transfers when reduced to 9600 baud so this may indeed do the trick. :D

OldUpdaterMaxBaud.png
Similar max settings are available with the new updater utility in the device.txt file (same directory as app). If this was the same error you were getting originally, then adjusting that file would be a better alternative than resorting to the older updater. The need to use the old updater is usually signalled by an app hang (hourglass) with no message.

NewUpdaterMaxBaud.png
 
NeilP said:
ummm..I'll have to go back and test again.
but it all seems to work correcly with V3 and my controller. I guess that using the wrong setting, just causes the Th out to go to max as opposed to 2 volts..as I guess is intended.
Click to expand...
Don't bother re-testing - I was mistaken - the original procedure was okay.

I edited my original posts above. The original Guide text was correct except for the new addition at the end about restoring the defaults. Sadly, that new last part led me down a line of confused thinking. Thanks for posting back questioning the error - your remarks nagged me into thinking about this again. :D
(So much CA poo to keep straight...)

Apologies for confusion that leaked from my brain into this thread....


  • I have undone my bad 'fix' and yet again re-posted an updated version of the Unofficial User Guide here.
    This retains the other minor revisions.
 
NeilP said:
John Bozi said:
Is the cave plug and play w/ the lyen eb3xx boards/controllers?


throttle doesn't work through cav3 (as it did when running grinfineon controller)
Click to expand...

I guess it depends on how old your controller is. and if you followed the guide or not when setting up

I have a Lyen 6 FET, about 6 months old and it works just fine direct plug in with the CA v3
I determined it worked OK first by following the test that you run to determine throttle wiring, as per telektiks User Guide
The Unofficial User Guide is available for download here.


Extract from version 3-0p6- j of the guide

1. If the CA interface type is unknown (the controller is not of recent manufacture or is not clearly labeled 'Large Screen Compatible') then perform the following test to ascertain the type:
a. place the bike on a stand or invert it onto handlebar/seat so that the motor can run safely,
b. plug the CA into the controller CA-DP interface and a throttle into the controller,
c. power up the controller and CA,
d. enter Setup, use all default CA settings, and configure:
e. ThrO->MaxOutput = 2V and ThrI->CntrlMode = Off(WOT),
f. exit CA setup and be prepared to quickly power down the controller if the motor starts.
g. Power down the controller.
h. If the motor powers up spontaneously leaving Setup without manually applying throttle, then the controller interface is Large Screen Compatible, otherwise it is V2 Compatible.
i. Power up the controller and restore ThrI->CntrlMode = Passthru and ThrI->CntrlMode = 0.8v
Click to expand...





throttle does work through controller but CA won't show amps etc.

thanks
Click to expand...

If the CA is not reading amps then there is definitely an issue.....from memory, there is no difference between all the version CA plug where the shunt monitor wiring is concerned....middle pins...3 and 4 on all the 6 pin CA-DP plugs...So if the current monitoring works with a V2 it should work exactly eh same with a v3.

if not then there is a problem.
Click to expand...

I lied, I was babying the throttle because Im still scared i may have the wrong hall/phase connection. In Off WOT mode (throttle to controller) I do get kmph now up to just over 40kmh (18s 20" wheel) but the amp readings are all in 0.1, only full throttle i got to 1.3amp... I have never looked at my amps before when the bike is unloaded, but I suspect I should be pulling much more than this....

this is my third controller but my first time needing to find the right hall/phase combinations... Does this incorrect amp reading mean I have the wrong combination?

Im using clyte 4065 so followed the most common exampe on his manual: yy gb bg halls - gg by yb phase.

EDIT: Pendragon =

1k resistor in place of this diode works

file.php


EDIT 2:
to confirm, removing pin 6 (throttle) from dp plug makes the right speed but my amps or watt readings are in decimals.

The bike is at least ridable.
 
teklektik said:
Adjusting Firmware Updater Max Baud Rate

izeman said:
what did you set your com ports (max baud in old updater) speed? i was told by grin that it could help to reduce it to 9600 baud as more complex commands sent to the bootloader my cause problems at higher speed.
Click to expand...
Ya - reducing the baud rate is a good strategy in these cases, but I was under the mistaken impression that the CA was locked in at 9600 baud.

I just loaded up the old updater and saw the (forgotten) Max Baud field. Oops, My Bad.
So - I tested this and it definitely slows the transfers when reduced to 9600 baud so this may indeed do the trick. :D

View attachment 1
Similar max settings are available with the new updater utility in the device.txt file (same directory as app). If this was the same error you were getting originally, then adjusting that file would be a better alternative than resorting to the older updater. The need to use the old updater is usually signalled by an app hang (hourglass) with no message.

Click to expand...




Communication error! Please ensure all cables are secure and in good repair!
Upload Progress 6%
Cycle Analyst 3 detected with bootloader version 0202
Now flashing with CA3_Prelim6_NoCal.hex
Please cycle power to the device....


I had 'some' progress when I changed the Baud Rate in the device.txt file to 9600 like suggested. The Upload Progress reached 6% before I received the 'Communication error!'.

I don't have a PC here so I'll see if I can source one to borrow tomorrow. Will let you know how I get on - I REALLY appreciate the continued help.

L
 
John Bozi said:
I lied, I was babying the throttle because Im still scared i may have the wrong hall/phase connection. In Off WOT mode (throttle to controller) I do get kmph now up to just over 40kmh (18s 20" wheel) but the amp readings are all in 0.1, only full throttle i got to 1.3amp... I have never looked at my amps before when the bike is unloaded, but I suspect I should be pulling much more than this....
Click to expand...
Unloaded it will be very low. I'd be worried if it was unloaded speed and high current. that is probably about correct unloaded.

John Bozi said:
this is my third controller but my first time needing to find the right hall/phase combinations... Does this incorrect amp reading mean I have the wrong combination?
Click to expand...
Only if current is very high



John Bozi said:
Im using clyte 4065 so followed the most common exampe on his manual: yy gb bg halls - gg by yb phase.

EDIT: Pendragon =

1k resistor in place of this diode works

EDIT 2:
to confirm, removing pin 6 (throttle) from dp plug makes the right speed but my amps or watt readings are in decimals.

The bike is at least ridable.
Click to expand...

Not sure about the above,
If you are removing pin 6, that is the throttle line ..so you must then have the throttle connected directly to the controller, so the CA will not be having any override function, so that should work fine if you don't want the new CA v3 functions.

If current is still low when riding the bike, and all else seems OK, then it could be a shunt calibration issue. Lyen often labels his controllers with the shunt value...6 get probably 2.0 milliohm. Or if you have an iCharger, follow the UUG and use the calibration procedure in there, using the iCharger in Foam cutting mode.

Edit 1 ...might be worth starting a new thread, outlining what kit you have and what you have done so far.

edit 2 Since you have the controller open, then coin the shunt wires, each one is normally 'about' 4 milliohms. Also check the back of the board. The latest mod that has been com in from keywin labs/ Lyen is to solder a second SMD shunt to the back of the board..so it appears you have one shunt when in fact you have two, a front of board wire and a rear SMD)..They will be in parallel...so two 4 milliohm shunts in parallel, is 2 milliohms.thereabouts
 
teklektik said:
Don't bother re-testing - I was mistaken - the original procedure was okay.


Apologies for confusion that leaked from my brain into this thread....
Click to expand...

Hurray, it was not me !! :p
SO, to confirm to to others reading this, the idea of the procedure is to send a constant 2 volt max out down pin 6 of the CA-DP plug to the controller. If motor runs, controller is v3 compatible, if it does not run it is not V3 compatible
 
NeilP said:
SO, to confirm to to others reading this, the idea of the procedure is to send a constant 2 volt max out down pin 6 of the CA-DP plug to the controller. If motor runs, controller is v3 compatible, if it does not run it is not V3 compatible
Click to expand...
Spot on.

The Settings:
The idea is that Off(WOT) ensures that almost all CA settings and logic are out of the picture and the CA will shoot MaxOut voltage on the Throttle Out (ThO) CA-DP connection. MaxOut limits the WOT condition - the 2v limit might arguably be reduced, but it pretty much guarantees that the throttle signal is high enough to have an effect regardless of the controller design. A real throttle is not required (hooking one up actually comes in a later step and we don't want to jump ahead at this point).

Why it Works:
A 'V2 Interface' has a diode that blocks the CA ThO voltage so nothing happens, a 'Large Screen Interface' instead has a resistor that routes the CA ThO voltage to the throttle input causing the wheel to spin.

Rationale:
In short, this procedure eliminates unknowns of controller design, test equipment availability or quality, ancillary test wiring, throttle, connectors, and the technical skill of the builder. It relies only on the settings in the CA and uses only the existing CA-DP connection in the normal way. Think: lowest common denominator builder and fewest variables.

  • There are a number of means to go about this, including hooking up external voltages or using meters. This 'operational' strategy takes the many possible controller circuit and component value variations off the table - all must simply 'work' although they may have varying electrical characteristics if we examine the controller throttle/Ground connections with test equipment. This also eliminates issues with cheap meters and inexperienced users fumbling around the CA-DP connector with probes - the live Vbatt+ voltage there can do damage if accidentally routed into the CA or controller throttle pins by bungled probing.

    Alternative suggestions are welcome, but that was the thinking behind the present test...
Truth be told, this test is only used by folks with old controllers of unknown origin - everything after summer of 2013 is 'large screen compatible' and so there is an expectation that 99% of the cases will have new controllers and can simply skip this step.
 
teklektik said:
Adjusting Firmware Updater Max Baud Rate

izeman said:
what did you set your com ports (max baud in old updater) speed? i was told by grin that it could help to reduce it to 9600 baud as more complex commands sent to the bootloader my cause problems at higher speed.
Click to expand...
Ya - reducing the baud rate is a good strategy in these cases, but I was under the mistaken impression that the CA was locked in at 9600 baud.

I just loaded up the old updater and saw the (forgotten) Max Baud field. Oops, My Bad.
So - I tested this and it definitely slows the transfers when reduced to 9600 baud so this may indeed do the trick. :D

View attachment 1
Similar max settings are available with the new updater utility in the device.txt file (same directory as app). If this was the same error you were getting originally, then adjusting that file would be a better alternative than resorting to the older updater. The need to use the old updater is usually signalled by an app hang (hourglass) with no message.

Click to expand...

Hi guys,

So all good!! Managed to borrow a PC laptop from work. The upload of the latest firmware worked a treat first time.

Note to other Mac users - you'll need a PC to update firmware on Cycle Analysy V3 and/or to upload new .HEX setup file.

Thanks for all the help guys - much appreciated! Looking forward to giving this bike a good test run tomorrow!!

L
 
Anyone using one of Lyens Thumb Throttles with the Cycle Analyst V3-DP?

I have Lyens thumb throttle which works when plugged directly into my 12FET Lyen controller. When I moved the throttle connections to go straight into the cycle analyst V3 I get nothing from the motor. The throttle graphic on the cycle analyst display moves up and down in response to the throttle position but still no motor action....

I thought I had followed the instructions as per the UUG:
Screen Shot 2014-08-28 at 23.38.11.png


Wiring going from CA to Throttle:
Red to Red
Black to Black
Green to White
Throttle wiring.jpg

Any ideas?
 
You must log in or register to reply here.

Similar threads

T
CA v3 limiting power even when I raise the limits.
Replies
3
Views
408
amberwolf
amberwolf
Twelvebears
Phaserunner v6 + Cycle Analyst CA3-WP, variable regen braking
Replies
0
Views
277
Twelvebears
Twelvebears
J
  • Question
Grin Cycle Analyst with CA 3_switch not displaying Mode 1/3 ( but shows 2/3 and 3/3 )
Replies
5
Views
265
jimmythebike
J
B
  • Question
3 pin PAS not working with Cycle Analyst V3
Replies
2
Views
580
E-HP
E-HP
M
  • Question
SOLVED -- Throttle lag - Sabvoton 72150 & CA V3
Replies
9
Views
2,049
MartyMcfly1
M
Back
Top