Cycle Analyst V3 preview and first beta release

A really big 'Thanks!' to Merlin for a flock of PMs and excellent video to get this issue localized and fixed quickly! It's super to have beta testers who not only load up this stuff, but make the special effort to critique the designs and document the issues. :D

So - a revised 3.1b2 version has been packaged up and will get posted here straightaway after a round of bench and road testing.

Also - as Merlin noted in his PMs - ThrO->DownRate also has been made into a global setting. I added it as item 'H' in the release post above so the noteworthy changes are all in one place...
 
v3.1 Beta 2 Released

Okay - One More Time - "We're back in 'beta mode' again!"

Upgrading to v3.1b2

Apologies for the false start with b1 - here's the revised b2 version:

Attached below is v3.1b2 in zip file format. If you unpack this into the Software Setup Utility installation directory, the new firmware files will be installed as well as files updating the Setup utility to be able to configure new features.
See the included README_Install.txt file.

  • IMPORTANT: This firmware has reorganized internals and any existing 3.0 setup or previously saved 3.0 Setup Files will not work. This means you will need to flash 'CA3-1b1_firmware.hex' to get started with a clean slate. After that, a fresh configuration can be built using the console buttons or the Software Setup Utility as before.

    IMPORTANT: Because the firmware flash will load the default settings, the throttle input/output configuration will not be correct. Treat this as an initial setup and place the bike on a stand or otherwise elevate the drive wheel to prevent a runaway when the flash completes and before a proper configuration can be restored.


    To use the Software Setup utility to upgrade:
    1. use the Setup Utility to suck out the existing CA 3.0 configuration and save it to a file,
    2. either enable 'Preference/Multiple Files' on the menu bar or open another Setup Utility instance and create a new 3.10 setup file
    3. manually copy the fields from the original setup to the new one and save it to a file
    4. update the CA with 'CA3-1b1_firmware.hex'
    5. download the newly created settings file to the CA
    This is sort of a PITA, but a much-improved setup scheme was designed and went into the Satiator. It hasn't yet been back-ported to the CA, so for now... :)
What's New

There are few material functional changes from 3.0 - the focus for this release is to improve the usability of what is already there by upgrading the display and user interface. A file 'ReleaseNotes.txt' is installed from the package and describes the new features, but here's the highlights:

  • Average Cell Voltage Display: The voltage displays can be configured for pack voltage as in 3.0 or an alternating pack & Avg Cell Voltage display.

    battVoltsDisplay.png
  • New EBrake Setup Menu with Ebrake Signal Level: EBrake signals can now be configured to work with either standard Active-Low or Active-High signals.

    View attachment 8
    ebk_signalLevel.png
  • Diagnostic Flags for xxRate limiting: Three new flag indicators have been added to the Diagnostic Screen after the 'Out' label to show when Fast/Up/Down-Rate are in play. These can help understand what is going on and make rate tuning a bit easier.

    diagnostic_allRate.png
  • Customizable Display Field on the Main Screen: The old "mi, Ah, Temp' field in 3.0 has been replaced with a field with 11 user-selectable data views which alternate sequentially.
    • 01 - distance (mi or km)
    • 02 - AH
    • 03 - temperature
    • 04 - temperature alert (displays temperature only if within 5 degC of threshold temp)
    • 05 - RPM (cadence)
    • 06 - Human Wt
    • 07 - ThrO
    • 08 - Aux Limit
    • 09 - Gear-inch (effective wheel diam for present pedal-to-road gearing)
    • 10 - Total Wh/mi or Wh/km
    • 11 - Instantaneous Wh/mi or Wh/km
    Since the Volts display may be alternating as well, the two rightmost fields have been interchanged to place all alternating display fields on the top line. This places the custom fields prominently in the top row which seems to be easier to read at a glance. Here is a sample screen showing the cell voltage and the instantaneous Wh/mi energy consumption. This last custom field makes it easy to identify power-eating driving behavior... Select only a single custom field to lock in that display without alternation.

    mainWhPerMiInstant.png
  • 'Flagword' Parameters have Submenus with Prompts: The new customizable fields joined the existing 3.0 'Still' and 'Moving' display screen features as defined by strings of 0/1 flags. To simplify configuration, these three parameters now have submenus where each bit has a separate screen with a brief prompt. A leading checkbox on each screen indicates if the feature is selected. A long left press will escape the submenu and return to the parameter screen; a second long left press escapes Setup.

    Pref-CustView.pngPref-Cust_degCAlert.png
  • Copy 'Per-Preset Parameters' Between Presets: A new option in the 'Preset' group allows either of the other two presets to be 'imported' to the current preset - even if the other preset is not enabled. This affords a simple means to copy working settings to a new preset for experimentation.

    PrSt-Import.png
  • Aux->ScaleLim is Now a Global Parameter: This is actually a downgrade in the status of this parameter so that it is no longer possible to have the AuxPot input control different Limiting in different presets (eg Amps in #1, PAS in #2) - all presets must have the same AuxPot Limiting. This was a difficult decision, but it is done as a precursor to cool new AuxPot features in the next beta release that would otherwise have used a great deal of valuable EEPROM configuration memory and been needlessly complicated for the vast majority of users.

  • ThrO->DownRate is Now a Global Parameter: As with Aux->ScaleLim, this single setting now applies in all presets.

  • PAS->PASWatts and Trq->AsstFactr are now a pre-preset parameters: These parameters have been expanded from global to per-preset so that the AutoPAS and Torque assist levels can be set on a per-preset basis.

  • Serial Data Log Changes: The inclusion of the AuxPot voltage allows proper interpretation of other data if limiting is in play. Torque data is now logged even if there is no pedaling. This doesn't affect PAS data much but is mainly for folks wishing to use the CA on a dyno to measure RPM via the speed input and a torque-based strain gauge signal via the PAS torque input.

    CA3-1b2_logChanges.png
Have at it and post up any feedback! :D

Download!
View attachment CA3-1b2.zip
(Newer 3.1b3 beta release available here!)
 
izeman said:
can we still use the "old" setup utility v1.31 to handle all new settings?
as i was unsure i did a clean install and set all values by hand :)
Yep - If unpacked according to the README_install.txt file, a couple of new files are made available to the Setup Utility so it knows about the new memory layout, new parameters, and default values. It can then do either 3.0 or 3.1 style configurations (it just can't convert one to the other).

izeman said:
after a short ride around the block all seems to work as expected.
Excellent!
I have been riding with it in evolving forms for months and do like the features. The 'instantaneous Wh/mi' is pretty interesting - watching the numbers shoot up to 140Wh/mi blasting up a hill was a slap in the face - sort of like watching the instantaneous mpg in the car in the same situation. I also like the 'temp alert' option that hides the temp display until it gets meaningful.

I frankly didn't think the per-cell-voltage would be that useful for me, I had a few pre-computed pack-voltage numbers in my head. But now I use it all the time - it made me realize the calculation guesstimates that I was doing on the fly. Interestingly, the reason that it alternates is that we realized that with a steady per-cell voltage display we were doing the opposite calculations to get pack voltage! So - it's a steady pack voltage for the non-techie types and alternates to give both values to folks more schooled in battery stuff...

Anyhow, we tried to hit popular suggestions from earlier in the thread - and some new ideas. Hope folks find them useful.
 
izeman said:
i still have the "no speed reading under 4-5km/h" problem. even with a completely new setup the problem is there. even though i have 3 magnets for the speed sensor. any ideas?
This is the way it should work. Please see post here.
I examined that code and did the math - 2.5mph/4kph look to be the minimum value. I ran a test on my bike (with 3 magnets) and got the same minimum speed.
As far as I can see, nothing has changed in this code section for awhile, so I'm a little at a loss for this change in behavior on your bike.
 
oh sorry. i overlooked your previous post on that issue. yes. 2.5mph or 4km/h seems to be the lowest value. i wish it could be smaller, but it's more a "like to have" than essential. i'm almost 100% sure that this was changed sometime in the past, and there were smaller values shown.
the strange thing is, that you can set "speed poles" to eg 20 - something way too high for a speed pickup sensor, and then you have to go WOT to get a speed reading of 4km/h, with the wheel revving at 400rpm. so for me it would make more sense to remove this limit for low pole count speed pickup sensors. there are very little speed signals sent to the CA, compared to a DD motor. just my 2c. :)
 
Having a few troubles with my CA(i'm no expert and have been through the manual). i cannot get an Amp/watts reading from it. Would this have something to do with running it in legacy mode?, Also i'm running a 50mv 100amp external shunt (previous owner), so should my R-shunt value be 0.500??
 
millzy555 said:
Having a few troubles with my CA(i'm no expert and have been through the manual). i cannot get an Amp/watts reading from it. Would this have something to do with running it in legacy mode?, Also i'm running a 50mv 100amp external shunt (previous owner), so should my R-shunt value be 0.500??
as long as you got a shunt connected the CA WILL show current readings. they may not be correct, but it will always show something. it just measures the voltage drop between the pins #3 and #4 of the 6pin CA-DP connector. and yes, 0.5mOhm should be the correct value to set.
 
A slightly updated version of the Linear Temp Sensor Setup Calculator XLS file has been posted up here.

There were a couple of transcription errors from the data sheets for two of the pre-configured parts. These were some odder parts that AFAIK are not in use for motor monitoring, but fixed now anyhow. Also added a leading index column so it is easier to see if the data is scrolled to other than the top row.

This affected only pre-configured data - calculations are unchanged, so any custom parts you may have configured with the previous sheet are correct - this version will give the same results.
 
I am building my kid Oset 12.5/16 mod with 10kw revolt motor. I would need these two screnarios to happen same time with bike in order my little kid to tame it:

1.) When gently turning the throttle it accelerates slow and steady
2.) When hitting the throttle full it accelerates like a beast

Will I achieve this with Kelly KLS and cycle analyst v3?

Or would there be another controller sub $300 which would give me 300 phase amps and 80 volts. Or is this my best bet:
http://kellycontroller.com/kls7240s24v-72v350asinusoidal-brushless-motor-controller-p-1344.html
 
Hi I have a CA v3 and want to connect a motor temperature sensor , its already got a sensor built in its a KTY84-130 so don't want to run another sensor.

I noticed you made a lookup table of values for this so if I just put it inline exactly as the plug in the CAv3 what is the pullup voltage is it 5v ?

So all i need to do is set the CA v3 to 0.45 and 262 and wire the Sensor direct to the 2 CA wires ?
 
So I put the pullup voltage as 4.96 ? or do I need to measure the voltage on the pin of the CAv3 temp plug with a multimeter ?

This motor is an old one I got shand someone else put this sensor in.
 
The pull-up is 5K and doesn't need adjustment.

The voltage is usually a bit under 5V so the default is fine. If you are really picky you can measure it and plug in the value, but the difference won't be very noticeable - particularly since the cutoff temps that you select are just ballpark numbers anyhow.
 
I've been riding with the latest version (3.10), and I like what I see, esp. that I can now observe an instantaneous wh/mi (or wh/km) statistic.

I notice that one of my complaints has been fixed, although I do not recall seeing it listed on Grin's "Feature and Bug List" at the bottom of this page:

http://www.ebikes.ca/product-info/cycle-analyst-3.html

When in AutoPAS mode with throttle I like to use the throttle to give a burst of power when starting from a stop, especially when starting on a hill. As I start to move, I immediately start to pedal and then I release the throttle while I'm pedaling. With version 3.0 (and earlier), the CA3 would allow the throttle out signal to go to zero then begin anew its ramp up according to the throttle ramp parameters and the PAS setting. This had the net effect of a power dropout while the user transitioned from throttle to PAS control.

With version 3.10 I see that this transition has been made seamless. Releasing the throttle (allowing it to go to "zero") while pedaling has the CA3 switch to the PAS signal seamlessly. There is no longer a power dropout.

Unfortunately, I did find a remaining bug that I did enter into Grin's bug fix page some time ago.

If I uncheck "Set Max Speed" box under "Speed Lims Menu Mask" in the Setup Utility, then write the configuration to a CA3, I can still set the speed limit on the device itself. Moreover, the parameter "IntSGain" is no longer accessible on the device. It appears the mask is not aligned with the memory array in the CA3 firmware.

Also, if in "Power Lims Menu Mask" I uncheck "Set Max Watts", then write the configuration to the CA3, I can still set Max Power on the CA3 itself. If I uncheck both "Set Max Watts" and "Set Max Amps", then I can still set these parameters on the CA3, but I cannot set AGain. I suspect there is a similar mis-alignment of the mask with the array in the firmware.

I would like to disable the setting of these parameters on the CA3 so that the operator of the bike cannot change speed or power limits of the bike while operating the bike.
 
Hey MrBill!
Thanks for your review. You're probably correct that the menu mask has become misaligned - there was a lot of internal restructuring in 3.1 and this may have gotten out of sync with the various menu changes. This issue is now in the task list and we'll be sure to get it verified and squared away in b3.

On a related note,we need to revisit maintenance of the bug list on the Grin V3 page. This was remapped into an different bug tracking scheme last year - which is where the task numbers come from that are listed in the Release Notes. Although the pre-existing items in the web list also exist in the new system, there is presently no backflow of new items from the new system to the web list...
 
redline2097 said:
I am building my kid Oset 12.5/16 mod with 10kw revolt motor. I would need these two screnarios to happen same time with bike in order my little kid to tame it:

1.) When gently turning the throttle it accelerates slow and steady
2.) When hitting the throttle full it accelerates like a beast

Will I achieve this with Kelly KLS and cycle analyst v3?

Or would there be another controller sub $300 which would give me 300 phase amps and 80 volts. Or is this my best bet:
http://kellycontroller.com/kls7240s24v-72v350asinusoidal-brushless-motor-controller-p-1344.html

anyone?
 
redline2097 said:
redline2097 said:
I am building my kid Oset 12.5/16 mod with 10kw revolt motor. I would need these two screnarios to happen same time with bike in order my little kid to tame it:

1.) When gently turning the throttle it accelerates slow and steady
2.) When hitting the throttle full it accelerates like a beast

Will I achieve this with Kelly KLS and cycle analyst v3?

Or would there be another controller sub $300 which would give me 300 phase amps and 80 volts. Or is this my best bet:
http://kellycontroller.com/kls7240s24v-72v350asinusoidal-brushless-motor-controller-p-1344.html

anyone?

With a given controller you can set up the CA3 with two different presets, each preset offering a different throttle up-ramp behavior. But, that's not quite what you want. To get (1) and (2) for a single preset isn't possible with current firmware as it would require the CA3 to give different treatment for different rates of change of throttle input voltage. In theory this could be done, but I suspect it requires too many computing resources for a feature that is not in high demand.
 
I discovered a couple more bugs in version 3.1b2 after my long ride today. (https://www.strava.com/activities/556065564)

1) I use three mode presets, the first for "Legal", the second for "Pdl Asst", and the third for "Unlimited". For each of these modes I set my PAS Watts to 750, 1000, and 2000, respectively and use the Aux Pot (set to adjust PAS Level) dial to dial-in the exact power. Everything works fine when in "legal" mode, Mode 1: the Aux Pot dial varies from 0 to 750. But, when I switch to "Pdl Asst" (Mode 2), the Aux Pot dial varies power from 0 to 750 watts, not 0 to 1000 watts. Same thing occurs when in "Unlimited" (Mode 3). It appears that the PAS Watts is fixed to that defined for Mode 1, even though this is settable by Mode.

I also observed another oddity that I am unable to consistently observe. The upper limit of the Aux Pot range is occasionally less than PAS Watts defined for Mode 1. I saw "740", "735", "699" watts and other figures slightly less than the correct figure as the upper bound, even though I never changed PAS Watts all day. The upper bound seemed to change based on other performance characteristics at the time, such as my speed. And, I confirmed that my Aux Pot voltage limits are consistent and unchanging--I don't have a dirty pot. I suspect there may be an addressing problem in the code. Maybe a pointer pointing to the wrong register? Might be a good idea to review all code that touches PAS Watts.

2) When I am regenerating more than 1000 watts, the minus sign "-" disappears, creating a moment of cognitive dissonance. I would prefer to see a minus sign in front of four-figure regeneration power levels instead of the PAS activity symbol as I am typically not pedaling while regenerating this much power.

Other than that everything worked well.
 
mrbill said:
redline2097 said:
redline2097 said:
I am building my kid Oset 12.5/16 mod with 10kw revolt motor. I would need these two screnarios to happen same time with bike in order my little kid to tame it:

1.) When gently turning the throttle it accelerates slow and steady
2.) When hitting the throttle full it accelerates like a beast

Will I achieve this with Kelly KLS and cycle analyst v3?

Or would there be another controller sub $300 which would give me 300 phase amps and 80 volts. Or is this my best bet:
http://kellycontroller.com/kls7240s24v-72v350asinusoidal-brushless-motor-controller-p-1344.html

anyone?

With a given controller you can set up the CA3 with two different presets, each preset offering a different throttle up-ramp behavior. But, that's not quite what you want. To get (1) and (2) for a single preset isn't possible with current firmware as it would require the CA3 to give different treatment for different rates of change of throttle input voltage. In theory this could be done, but I suspect it requires too many computing resources for a feature that is not in high demand.

Is there any controller manufactor with controller which works like that? Or can I overcome this problem with Kelly controller + throttle tamer?
 
v3.1 Beta 3 Released

Thanks to some excellent testing and bug reporting by dedicated beta testers (see mrbill's posts above - and other email contributors), we have a new beta release. :D

This release primarily addresses issues discovered in 3.1b2, but a few new features are included to polish up screens to improve usability or existing content. All changes are listed in the Release Notes (see summary below).

One feature of interest is a new Lifetime Statistics category with parameters relocated from the Speedometer and Battery categories to make it easier to review/save this historical information.


  • Unfortunately, the EEPROM parameter map has changed and any existing 3.1 alpha or beta setup is incompatible and cannot be used.

    Unlike the situation with the many 3.0 beta releases where EEPROM did not change for a long time, some changes are inevitable with 3.1 where existing features are modified to fit in with the new feature set. That said, the planned b4 release will be compatible with b3 so an easy 'NoEeprom' upgrade will work.
    Meanwhile - apologies for the PITA with b3....
The required upgrade procedure is detailed in the Release Notes which you are strongly encouraged to read to minimize installation difficulties. Here's an excerpt with upgrade instructions and a summary of changes:

3.1b3 Release Notes said:
Note: V3.1b3 Setup files are **incompatible** with firmware or Setup files of any earlier releases including any earlier 3.1 version.

IMPORTANT: Please install this package according to README_install.txt. Failure to do so may result in improper operation of the Setup Utility.

IMPORTANT: Once this zip package is installed on a PC, the Setup utility will no longer be able to properly utilize Setup files prepared for previous 3.1 beta versions. Settings in older 3.1 files will not display or edit properly and will download to the CA incorrectly resulting in improper CA operation. Similarly, setup files prepared with this release cannot be used with earlier 3.1 firmware versions. 3.0 Setup files are unaffected.

>> Any incompatible existing CA setup must be recorded and manually reconfigured over a fresh 'CA3-1b3_firmware.hex' installation.
>> In particular, existing 3.1 beta CA Setups must either be:
  1. exported manually from the CA console (at least the statistics)
  2. OR read by the Setup Utility and preserved in screen snapshots PRIOR to installing this package. Saving to a 3.1 Setup file and reading that file after installing the zip package will not work.
IMPORTANT: Because the firmware flash will load default settings, throttle voltage configuration will be incorrect. Treat this as an initial CA installation and place the bike on a stand or otherwise elevate the drive wheel to prevent a runaway when the flash completes and before a proper configuration can be restored.

  1. 2228 - PassThru Throttle has initial 200ms delay
  2. 2252 - Menu mask bits affect the wrong setup parameters
  3. 2260 - PAS->PASWatts operates globally instead of per-preset
  4. 2264 - Menu mask items in Setup Utility are in reverse order
  5. 2262 - (new) Lifetime statistics moved to new 'Statistics' category
  6. 2268 - (New) Maximum PAS Pole Count has been restored to 24
  7. 2258 - Main Screen displays regen Watts w/o (-) sign if more than -999W
  8. 2272 - Reducing Preset Count can leave Current Preset set to an unenabled preset
  9. 1730 - (new) Simplify Aux Input Setup Menu
  10. 2274 - (new) Enhance the Preset and Preferences setup preview screens and menus
    • Add 'Data Rate' and 'Averaging' to the Preferences Preview screen
    • Revise Presets Preview screen to show current preset as '#n PresetName'
    • Center line 2 display text of Preview Screens
    • Reorder the Preset Submenu so that PrSt->CrntPrset is the first item to facilitate changing the preset within Setup.
    • Fix PrSt->CurPrSet to decrement preset on short left press to make operation consistent with all other submenu edit operations
    • Rename PAS->SensrType to PAS->SignlType to avoid conflict with Trq->SensrType

Again, thanks for the excellent testing and reports!

View attachment 1
(Previously implemented 3.1b2 beta features (part of this build) described here and in the Release Notes.)
(Newer 3.1b4 beta release available here!)
 
The rubber button strip on my CAv3 came off during a ride the other day, not sure why as I never noticed that they were loose. What are my options to get a replacement?

I emailed info@ebikes.ca a few days ago but did not get a reply.
 
Back
Top