Cycle Analyst V3 preview and first beta release

Get real world experience and user feedback on the electric bicycle products.
ebike11
10 kW
10 kW
Posts: 927
Joined: Nov 16, 2013 8:01 pm
Location: Korea

Re: Cycle Analyst V3 preview and first beta release

Post by ebike11 » Oct 05, 2017 10:23 am

teklektik wrote:
ebike11 wrote:im getting a strange flashing battery icon on the CA3.
...
I searched the guide but couldnt see that icon.
CA3_mainStatusScreenTempAlert.png

Again - please do the installation steps....
CA3_basicInstallation.png
Thanks a lot man...you are the king of the CA!

ebike11
10 kW
10 kW
Posts: 927
Joined: Nov 16, 2013 8:01 pm
Location: Korea

Re: Brightening CA Backlight

Post by ebike11 » Oct 08, 2017 2:50 am

teklektik wrote:
ebike11 wrote: actually both CA3s are running on the same battery pack. just one CA3 is brighter than the other straight out of the box.
I am not confident opening the CA3s up.
As the posts in izeman's search revealed, the CA brightness is largely a matter of the +5V current drawn by the CA and it's accessories (throttle, etc). So - if you have a three position switch or ebrakes on one CA but not the other, the brightness might be different. Different LED batches might yield slightly different brightness, but I can't speak to that probability.

As pointed out in post (https://endless-sphere.com/forums/viewt ... 8#p1263588), it's possible to dim the display a bit by bypassing current around the LCD module. Similarly, it's possible to brighten the backlight by adding a load resistor in parallel with the 5V regulator to draw a bit more current through the LCD module. In that case, you would still need to pay attention to the current limitations determined by the pack voltage as called out in the Guide. For instance, since the CA 5V supply is drawing about 10ma, you might add a load resistor to draw maybe 10-50% more current (1-5ma).

Anyhow, since you don't feel comfortable cracking the case, you can go after this by directly increasing the 5V regulator load with an external resistor. This is probably the 'hackiest' strategy, but will work fine if you need a no-messing-with-CA-internals approach - just pay attention to the overall current inventory as mentioned above. So - you might try jumping a 1K resistor across the +5V and Gnd pins of the AuxPot or Throttle connectors. This would boost the current draw 5V/1K = 5ma which will visibly brighten the backlight. You could make up a super short M/F JST pass-through extender that has the power leads jumpered with the resistor of choice (use a small 1/8 or 1/10W part - there's no power to speak of). This would let you do the fabrication off the bike on the bench and then just plug in the gizmo when it's ready with no chance of taking your bike out of service with a fabrication accident.
Hi Teklektik,
Would this be the correct resistor to jump across the POS. and NEG. of the AuxPot? Id like to try it first to increase the the screens brightness a bit. It seems easy enough. Thanks

https://www.aliexpress.com/item/1LOT-10 ... 4d4d39440b

User avatar
izeman
10 GW
10 GW
Posts: 4502
Joined: Jun 21, 2011 8:25 am
Location: vienna, austria
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by izeman » Oct 08, 2017 3:47 am

as tek already stated, different total loads make the backlight brighter or dimmer. as we don't know how bright it is right now, you can't say what load you must add, to make it as bright as the other one.
taking tek's example of a single 1k resistor, the posted parts are correct and fine to use.
the problem you have is, that with 1k the lcd may then by brigher or dimmer than the other one. so you would need a different resistor. buying an assortment of different values gives you the possibilty to play around and find the perfect value.
of course you could put some 1k resistors in parallel and series to build a resistor mesh of any value you like (well 100k max of course, as it's 100 pieces of 1k :wink: )
this may help you: http://www.calculator.net/resistor-calculator.html

ebike11
10 kW
10 kW
Posts: 927
Joined: Nov 16, 2013 8:01 pm
Location: Korea

Re: Cycle Analyst V3 preview and first beta release

Post by ebike11 » Oct 08, 2017 5:27 am

izeman wrote:as tek already stated, different total loads make the backlight brighter or dimmer. as we don't know how bright it is right now, you can't say what load you must add, to make it as bright as the other one.
taking tek's example of a single 1k resistor, the posted parts are correct and fine to use.
the problem you have is, that with 1k the lcd may then by brigher or dimmer than the other one. so you would need a different resistor. buying an assortment of different values gives you the possibilty to play around and find the perfect value.
of course you could put some 1k resistors in parallel and series to build a resistor mesh of any value you like (well 100k max of course, as it's 100 pieces of 1k :wink: )
this may help you: http://www.calculator.net/resistor-calculator.html
Thx for the reply!
At the moment it seems a bit dimmer than my othe CA that is on the same bike. Id just like to experiment getting the brightness up a bit more without having to open anything up.
There are no accessories or anything connected..just the throttle and the power to the external shunt

ebike11
10 kW
10 kW
Posts: 927
Joined: Nov 16, 2013 8:01 pm
Location: Korea

Re: Cycle Analyst V3 preview and first beta release

Post by ebike11 » Oct 09, 2017 5:58 pm

By the way, i was wanting to confirm, is the Grin external shunt rated at a Maximum 50A ?
I have a 60A controller so I assume it will cut off at 50A?

User avatar
izeman
10 GW
10 GW
Posts: 4502
Joined: Jun 21, 2011 8:25 am
Location: vienna, austria
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by izeman » Oct 10, 2017 9:40 am

ebike11 wrote:By the way, i was wanting to confirm, is the Grin external shunt rated at a Maximum 50A ?
I have a 60A controller so I assume it will cut off at 50A?
a shunt can't cut off. a shunt is just a piece of metal with a stable resistance that doesn't change (too much) while it heats up.
you measure the voltage drop over the shunt. that way you can calculate the current going through it (I=U/R).
that's why it can't cut off, it just gets warmer and warmer, and at some point will just melt ;)

RTIII
1 kW
1 kW
Posts: 436
Joined: Jun 02, 2017 2:52 pm
Location: Oakland, CA, USA

Re: Cycle Analyst V3 preview and first beta release

Post by RTIII » Oct 10, 2017 8:43 pm

Is there any version of Cycle Analyst that captures cadence data?

I'm not interested in control of any kind, so maybe the V3 is the wrong product family for me... I'm hoping to find something that can give me A, V, speed, and cadence in a consistent data set so I can compare between two bikes and also so I can help "map" the performance of my torque-sensing mid-drive...

TIA,
RTIII

User avatar
amberwolf
100 GW
100 GW
Posts: 23948
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by amberwolf » Oct 11, 2017 1:30 am

AFAIK the CA doesn't directly capture the cadence data (not stored in the CA itself) but it does have a serial output that can send various parameters (I don't remember which ones, but it should be in the manual or Grin's site for CA v2.1 (I think) or later) in realtime out at the baud rate you set, so you can use an external serial logger to log that data, then analyze it however you want later.

Grin sells the Cycle Analogger that will do the logging, but you can use anything with a serial input to do it (including a laptop with a terminal program that can write the incoming data to a text file, or arduino, etc).

ebike11
10 kW
10 kW
Posts: 927
Joined: Nov 16, 2013 8:01 pm
Location: Korea

Re: Cycle Analyst V3 preview and first beta release

Post by ebike11 » Oct 11, 2017 9:09 am

izeman wrote:
ebike11 wrote:By the way, i was wanting to confirm, is the Grin external shunt rated at a Maximum 50A ?
I have a 60A controller so I assume it will cut off at 50A?
a shunt can't cut off. a shunt is just a piece of metal with a stable resistance that doesn't change (too much) while it heats up.
you measure the voltage drop over the shunt. that way you can calculate the current going through it (I=U/R).
that's why it can't cut off, it just gets warmer and warmer, and at some point will just melt ;)
Ok thx..im still not too techy when it comes to equations like you posted. I just read 50A continuous on the shunt description on ebikes.ca so i thought that i shouldnt go over that amount for long periods of time
Last edited by ebike11 on Oct 11, 2017 8:24 pm, edited 1 time in total.

User avatar
teklektik
1.21 GW
1.21 GW
Posts: 3525
Joined: Mar 26, 2011 1:15 pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Oct 11, 2017 11:32 am

RTIII wrote:Is there any version of Cycle Analyst that captures cadence data?

I'm not interested in control of any kind, so maybe the V3 is the wrong product family for me... I'm hoping to find something that can give me A, V, speed, and cadence in a consistent data set so I can compare between two bikes and also so I can help "map" the performance of my torque-sensing mid-drive...
Since the CA3 is the only CA with PAS inputs, it's the only candidate :D

It can do what you request and has both live cadence displays and well as serial log data that is in csv format, ready for import to Excel. The newer 3.1beta firmware has a slightly enhanced data stream but the basic crank rpm, torque, human watts stuff is still there along with the expected electrical and speed/distance stuff.

Please download the Unofficial Guide. This lists the available serial data (see below) and describes a means to log the data on your phone using the standard CA programming cable + adapter cable. Alternatively, the Cycle Analogger from Grin can be had with a GPS option so that you get positional information as well. That enhanced log data can be uploaded to the Grin Trip Analyzer and you can see the map and plot the various parameters to see what's happening on hills, flats, etc.
CA3.0_serialData.png
CA3.0_serialData.png (26.09 KiB) Viewed 720 times
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

RTIII
1 kW
1 kW
Posts: 436
Joined: Jun 02, 2017 2:52 pm
Location: Oakland, CA, USA

Re: Cycle Analyst V3 preview and first beta release

Post by RTIII » Oct 11, 2017 3:26 pm

teklektik wrote:
RTIII wrote:Is there any version of Cycle Analyst that captures cadence data?
Since the CA3 is the only CA with PAS inputs, it's the only candidate :D
Thanks, teklektik, I'll be reading up!

cerberpeter
1 mW
1 mW
Posts: 14
Joined: Nov 20, 2016 11:56 pm
Location: Newark-on-Trent, UK

Re: Cycle Analyst V3 preview and first beta release

Post by cerberpeter » Oct 14, 2017 6:46 pm

I'm digging the forum for few hours now, but I could not find what is the lowest shunt resistance that I can set. I use modified Lyen controller that appears to have shunt resistance of about 0.55 mOhm, but when I set anything below 0.755 mOhm in CA V3, it saves the setting with that value (0.755). Do I have to use the Hi settings (for KW) or is there a way around this?

User avatar
teklektik
1.21 GW
1.21 GW
Posts: 3525
Joined: Mar 26, 2011 1:15 pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Oct 14, 2017 8:34 pm

cerberpeter wrote:Do I have to use the Hi settings (for KW) or is there a way around this?
See Unofficial Guide:
CA3_UUG_shuntSnippet.png
CA3_UUG_shuntSnippet.png (77.21 KiB) Viewed 639 times
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
izeman
10 GW
10 GW
Posts: 4502
Joined: Jun 21, 2011 8:25 am
Location: vienna, austria
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by izeman » Oct 15, 2017 3:26 am

Conclusio: NO way around. You have to use the kW setting which is some kind of strange if you have a bike in the x00 W range, showing eg 0.7kW instead of 753W ;).

cerberpeter
1 mW
1 mW
Posts: 14
Joined: Nov 20, 2016 11:56 pm
Location: Newark-on-Trent, UK

Re: Cycle Analyst V3 preview and first beta release

Post by cerberpeter » Oct 15, 2017 10:14 am

Thanks guys! I switched to KW and it is alright for me. I guess the other option would be to use some external shunt, but I will be over the 10 kw sooner or later anyway :D

User avatar
amberwolf
100 GW
100 GW
Posts: 23948
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by amberwolf » Oct 15, 2017 10:27 pm

Yeah, I recently changed SB Cruiser to a parallel pair of Grin's SA-CA shunts, for 0.500milliohm, and had to change teh CA2.3 mode to high-range.

Doesn't affect anything for me other than getting used to the slight change in the way watts are displayed. :)



(with peaks of 3x what the shunt is rated for, I figured it might help to decrease voltage drop there, and heating, if I halved the shunt resistance. Under heavy hauling loads up any inclines, it could sustain those kinds of currents for a while, and a single standard SA shunt isn't designed for that).

interrupted
1 µW
1 µW
Posts: 2
Joined: Nov 03, 2017 10:17 am

Re: Cycle Analyst V3 preview and first beta release

Post by interrupted » Nov 03, 2017 10:20 am

Heyho :)

can i somehow hide the display output when using the analog poti + pas factor? i dont want the ca to display tons of watts while turning the pot, this could be very hard to explain in certain situations (EU 250w Limit).

Cheers!

User avatar
teklektik
1.21 GW
1.21 GW
Posts: 3525
Joined: Mar 26, 2011 1:15 pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Nov 03, 2017 11:29 am

interrupted wrote:can i somehow hide the display output when using the analog poti + pas factor?
No. This has been discussed in-house but hasn't made it to the feature list...

One thought has been to make it an option to have Aux changes force the main screen to immediately switch to the AuxA or AuxD custom field if that field was enabled instead of bringing up the full screen Aux change pop-up. But with so many other meaningful features, this one didn't get much traction since it didn't really add a lot of new functionality...
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

interrupted
1 µW
1 µW
Posts: 2
Joined: Nov 03, 2017 10:17 am

Re: Cycle Analyst V3 preview and first beta release

Post by interrupted » Nov 03, 2017 1:19 pm

teklektik wrote:
interrupted wrote:can i somehow hide the display output when using the analog poti + pas factor?
No. This has been discussed in-house but hasn't made it to the feature list...

One thought has been to make it an option to have Aux changes force the main screen to immediately switch to the AuxA or AuxD custom field if that field was enabled instead of bringing up the full screen Aux change pop-up. But with so many other meaningful features, this one didn't get much traction since it didn't really add a lot of new functionality...
:( but yes, i understand your point. in one of the older firmware releases the CA showed a factor 1x 1.25x and so on, do you remember in which version this has been changed? maybe i am going to downgrade to that one

User avatar
teklektik
1.21 GW
1.21 GW
Posts: 3525
Joined: Mar 26, 2011 1:15 pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Nov 03, 2017 2:00 pm

interrupted wrote:do you remember in which version this has been changed? maybe i am going to downgrade to that one
No, not off the top of my head (AutoPAS used to show percentage of the Aux adjustment for the Aux change pop-up) - I'd have to search for it as well. I also think that may have been a 'minor change' and so not called out in the Release Notes, so they may not be of much help.

We have been working on new PAS features which have in part reflected into the Aux Change displays, but this question of 'percentage' vs 'Watts' units has always remained with the 'Watts' conclusion....

That said, the Torque Aux Change is in terms of Assist Factor (e.g. 3.5x). If this is really making you unhappy, you could go to the Guide and look at the section on AutoTorqPas and use that instead of AutoPAS mode. You would get the same PAS assist but you would have the 'factor' display of the torque sensor PAS systems instead. Sort of the long way 'round the barn, but....

(I hope I've gotten the various displays and versions correct - we've had a lot of PAS versions and changes since b20 and they all sort of blur together...)
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
haulincolin
10 W
10 W
Posts: 98
Joined: Apr 28, 2013 8:37 pm
Location: Seattle, WA
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by haulincolin » Nov 12, 2017 7:33 pm

The speedometer output reading has stopped working on my Cycle Analyst, and I'm trying to track down how to fix it. This is actually a CA 2.4, but it's relatively new, so I suspect the board layout is similar to a CA3.

Recently I had an issue where the side plate rubbed the housing off of a phase wire and a hall sensor ground inside my hub. This blew up the hall sensor, which I replaced. Ever since then, my speedo has read 0, so I suspect this has something to do with it. I've checked everything all the way back to the pads on the CA board, and it is getting a signal at that point, so something is wrong past there.

I've searched this thread as best I can for speedometer-related issues and haven't found anything useful. I'm comfortable checking individual circuit board components if I need to, but without a detailed schematic, I don't know where that signal goes after it leaves the SPD pad.

User avatar
amberwolf
100 GW
100 GW
Posts: 23948
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by amberwolf » Nov 12, 2017 9:29 pm

At least yours only took out the speed sensor...a few years ago, I had a short at the standalone shunt's cable where it exits the shunt body that seems to ahve taken out the CAv3 MCU itself (and also took out the hall in the motor, that was hooked to the sensor line at the time).

I haven't gotten to repair that CA yet, so I don't have any detail info to help--except that IIRC from my continuity checking at the time, there's nothing in the path from the speedo input to the MCU itself that would stop a battery-voltage short (like a phase wire) from reaching the MCU pin, which probably cannot handle that level of voltage. So if the V2.x CA speedo path is like the V3's, the MCU input on yours may be damaged, preventing any speed signal from being read.

To find the path on mine, I set the DMM to continuity check, and then started probing from the speedo input to various places, eventually finding an MCU pin. (I don't recall which one, and I apparently never posted that to my CA repair thread here
https://endless-sphere.com/forums/viewt ... =2&t=19540
and don't see it in the CrazyBike2 thread either).

User avatar
Samson
1 W
1 W
Posts: 62
Joined: Nov 03, 2011 4:14 pm
Location: Ontario, Canada

Re: Cycle Analyst V3 preview and first beta release

Post by Samson » Nov 13, 2017 11:42 am

I have a CA3 ver 3.03 firmware, 14S battery and em3ev 6 Fet sensorless controller on a small gear motor.
I'm having a problem with the motor slowing momentarily during throttle up under normal road loads. When I open the throttle the motor accelerates then backs off after about .5 - 1 second momentarily before resuming acceleration. The CA3 shows that the throttle voltage is dipping momentarily and the A flag flashes indicating that the CA3 is current limiting. I tried first increasing the Max Amps to 25 from the 12 Amps I normally run with no change. I then tried the AGain setting + - 30%, 50% and 120% from the default value of 150 with no change. Throttle Uprate is 1 V\s, Fastrate is 1.5V\s though I tried setting these at other values without effect. I set the block time on the controller down to .1 seconds from 1 second with no change. Wiring is solid and the bike runs well on the road. Any idea why this would occur?

tanstaafl
10 W
10 W
Posts: 91
Joined: Apr 21, 2017 2:42 pm

Re: Cycle Analyst V3 preview and first beta release

Post by tanstaafl » Nov 13, 2017 1:46 pm

I'm having a strange problem with my CAv3. It's running the latest beta firmware. I recently increased my battery voltage to 60v nominal (from 36v) and I now notice a much more pronounced "pulsing" from both my throttle control and from my torque-PAS (TDCM). I've decreased both AGain and WGain a bunch, and the pulsing has decreased, but it's still there. As I decreased the gains, my throttle has started behaving "funny". I don't get any motor response until I "floor" the throttle, but after it starts up, I have reasonable control over the whole range. The throttle itself is working -- I can see the indicator change on the main CA display -- but I get no acceleration until I've moved it all the way. (and the pulsing is still there) (BTW, didn't mention this at first, but I've have the throttle set to control Amp -- but it had the same problem in pass-through mode -- maybe because of my watt limits?).

What are reasonable values fro AGain and WGain? The starting values seemed strange to me -- AGain was something like 150. I've reduced it an order of magnitude and will probably take it lower still. UPDATE: still a problem at 10. WGain was much lower and I've reduced it to 7. I'm still noticing the "pulsing" however from both PAS and manual throttle control.

BTW, everything was better with the 36v battery. I probably needed to adjust the settings a little bit, but they weren't bad (not nearly as bad as I'm noticing now, even after lowering the gains a TON).

Can anybody suggest some other parameter that I should tweak?

2nd Update: I just lowered the UpRate to 1.0 and the FastRate to 1.5. Still feel some surging when I'm using the torque-PAS. I'm able to hold a steady power level with the throttle now. However, I have "the worst of both worlds" instead of the best -- maxing out the throttle control doesn't seem to be enough to get things started (at least, not without a _long_ delay). If I peddle and PAS kicks in the motor power, however, the throttle then becomes instantly responsive and I can pretty much set whatever level I like and hold it there. It would be nice if I could just get a nice, smooth response while I peddle and only have to use the throttle when I'm climbing the big (15% slope) hill. Also, one other tidbit -- the upper watt limit that I set, 1000w, seems to be held pretty consistently. If I've maxed out the throttle, I notice the watt output stays really close to 1000 (maybe 980 or 990, then up to 1010 or 1020 most of the time). So I don't think the WGain is the reason I have the surging problem from the Torque-PAS. If there's any other information I can provide, or any settings I should try, please let me know.

User avatar
amberwolf
100 GW
100 GW
Posts: 23948
Joined: Aug 17, 2009 6:43 am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by amberwolf » Nov 14, 2017 12:36 am

Samson wrote: the A flag flashes indicating that the CA3 is current limiting. I tried first increasing the Max Amps to 25 from the 12 Amps I normally run with no change.
What is the actual current it's reaching?

Post Reply