CycleAnalyst causes motor pulsing / power cutting in and out

[RVD]

I recently put back together my TidalForce frame bike. It has a Crystalyte HS3540, Lyen 12 fet controller, and is running 12s lipo at 15ah. After a big move 3 years ago + 2 more kids, I finally got around to getting this bike back together again and functional.

This morning was my inaugural ride on this bike and it was performing well for the first 5 miles. I wasn't pushing it too hard mostly riding at around 15-20 mph and according to the CycleAnalyst the motor was pulling 10-30 amps most of the time.

I have a basket secured to the rear rack (kind of like milk basket with less height) and my lipo is in some hard cases that i made with floor tile. I took bumps slow and easy because I don't want a big bump to cause a battery pack to come out flying from the basket.

I have 2 set of 3x6s 5ah lipos in parallel and then series.

About 5 miles into the ride, I hit a bump and then immediately noticed that power was cutting in and out (sort of the feeling if regen were to turn on and off every 1/2 second or so). When pulling the throttle, the bike would move but sort of jerk a little bit on/off. I let off on the throttle and checked all of my connections and found that one of the parallel packs came loose. I plugged it back in and tried to ride but still pulsed.

I pull all of the wires apart from the battery to controller and re-plugged everything in and rode and same thing.

I visually inspected all of the connections to the controller (halls, phase, regen, etc.) and they appear fine. I have an electric brake and that also seemed fine. Voltage reading on the CA was in the 46.5-47v range.

I was in limp mode for about a mile or so pedaling and periodically trying out the motor again and about a mile later, the motor started to be stable again and I was able to make it to work (another 2 miles).

Any thoughts on what happened? Could something have overheated and then cooled down? Like I said, I don't think I was pushing the motor or controller by pushing a lot of watts to it (maybe 1500 at peak and 600-900 sustained)?

My ride is a little hilly but nothing major. I recently installed hydraulic disc brakes (front and back) and it sounds like the back scrapes a little bit but not much.

 

[teklektik]

Start by unplugging the CA and see if the problem disappears.

• If not, then you can plug it back in and look elsewhere for problems...
• If so, then the CA may be thrashing in and out of limiting (e.g. current/LVC) due to an intermittent connection.
  Unplug and re-plug the CA-DP connector a few times to clean the contacts (do not try to 'clean' them with anything abrasive!).
  If that doesn't help, inspect the CA cable and connector from all the way back to the PCB checking for broken/intermittent connections.

 

[RVD]

ok thanks! i'll try messing with the CA. Hmm, I used to run the bike at 14s and 48v LifePo4 so maybe I set the LVC to something higher back then and running it with 12s is triggering something.

 

[teklektik]

maybe I set the LVC to something higher back then and running it with 12s is triggering something.

 

[RVD]

maybe I set the LVC to something higher back then and running it with 12s is triggering something.

lol, i don't think i did but it's been 3 years so almost anything is possible. and i don't know if it explains why things seemed to work again after not working for about a mile.

i have a hill going home later today and i surely hope there aren't any problems then. i'll really be in limp mode trying to ride this heavy bike up the hill.

 

[RVD]

thanks! you helped saved my commute back home today. i had to leave work early and everything was fine right up until i got to my hill on the last part of my way home. this started happening again. i disconnected the CA and it seemed fine. i went through settings and my CA lvc was set to 44v. i was at 45.5 volts and even under load it wasn't going under 44 (i wasn't accelerating quickly) but i changed the lvc to 40v anyway.

problem is that it still didn't work...either throttle did nothing or it was cutting in and out again. i ended up disabling the CA and riding home and it worked well.

any thoughts on why the CA would do this? any other settings other than lvc?

 

[RVD]

i ran the bike today at 14s lipo just in case there's a phantom setting somewhere that's limiting current but no go.

i tried plugging in the connector in/out a few times to hopefully clean it out but that didn't work. i suppose i can try the age old trick of blowing into it to remove any possible debris.

at this point, if the CA is plugged in, the bike doesn't move even though the CA turns on. sometimes the bike will move for a few feet and then die. i'll inspect the CA cable some more but otherwise i may just need to get a new one or plug in a watt meter for now.

my CA version is a little old (2.2.3).

i just rode blind this morning. i should be fine as my commute is pretty short and i know my bike/range pretty well. i'm running 14s2p (52v 10ah) and my total commute round trip is around 16 miles (usually use around 6-7ah max) and i don't really push too hard (around 18-25 mph).

 

[teklektik]

i ended up disabling the CA and riding home and it worked well.

any thoughts on why the CA would do this? any other settings other than lvc?

Okay - this is actually good. We know it's the CA - and probably the settings. The v2.23 is a good reliable device so there should be no problems once it's tweaked in.

The CA v2 can limit voltage (LVC), current, and speed. All have 'gain' settings that control the speed and magnitude of the CA response. It sounds like one or more of your gain adjustments are out of whack and you are getting overshoot in the corrections - this gives the familiar surging effect.

Examine the manual and restore the current and speed limits and gain settings to defaults. This 'should' work as nothing will be limited.

Then set your max current and see what happens. You need to read the manual very carefully - it can be a little difficult to grasp what's going on.

Post back if you don't have success or aren't happy with the result. We can expand on the manual a bit...

 

[RVD]

LVC is set to 40v. Max speed is 99 mph. Max Amps is 53 amps. The only one of these that I changed was the LVC from 44v->40v a few days ago.

I have never messed with gain settings so they should be at the default levels but I'll take a closer look when I get home today. I did change the rshunt back when I switched from a 9C motor to a Crystalyte motor.

I also tried cleaning out the contacts that connect the CA to the controller. I just blew into it to dislodge any dust/dirt if any and also tapped at it to dislodge anything. After plugging it in, I tried the throttle and it worked. However, I am not confident that it will continue to work reliably and I don't believe it's as simple as dust/dirt getting into the contacts.

 

[RVD]

I've been racking my brain on this all morning looking at settings, etc.

I am running the bike at 18s2p lipo (59v 10ah). This morning everything worked fine for about a mile or so and then it started happening again. I eventually disconnected the CA from the controller and rode back home.

I have looked at settings and can't figure this out. Any chance I simply have a faulty CycleAnalyst? It is about 3 years old at this point so it's old and it hasn't been used in about 2 years so maybe in that time off something went bad?

Here's a quick YouTube video of my settings. For awhile the motor will have power cutting in and out to move slowly step by step but eventually it just fails and pulling the throttle does nothing (no movement).

https://www.youtube.com/watch?v=baP1wtiX4t4&feature=youtu.be

[youtube]baP1wtiX4t4[/youtube]

 

[teklektik]

Okay - that's an old small screen CA.

I went back and read the OP again....
So I'm a little confused - was this working fine before and stopped working on the impact or was is wonky after you did the battery change?
Am I correct that you have a speed pickup on you wheel?

There's no point in looking at your settings if this is a problem to a previously working unit as a result of physical damage (differing from some settings issue as a result of a hardware configuration change).

At this point I suspect your shunt or Gnd connection in the CA-DP connector/cable are not good. This might even be an issue with the wiring where it attaches to either the controller or CA PCBs. The transient failure that you report seems to indicate a connector/connection issue. I would inspect the wiring from the CA PCB out to and including the CA-DP connector. Check the continuity with a DMM. If you don't find anything, do the same on the controller side.

 

[RVD]

It could just be a coincidence but I haven't ridden this bike in over 2 years. I was riding the bike for the first somewhat long ride (more than just around the block), went about 4 miles, went on a few bumps, and lost power. The motor was cutting in and out. I stopped and examined all of the connections and found one of the parallel battery pack connections loose (12s 10ah lipo so one of the 6s 10ah packs were loose). I fixed the connection and started riding again but the motor was still jumpy. This setup was working 2 years ago when I used to commute 32 miles/day on it. This never worked after putting things back together again since I just put this bike back together again last week or so. The odd thing is that the bike sometimes (not always) seems to work for a short time if you let it rest for a bit (5-10 minutes or so). Unhooking the CA always works.

Maybe the loose connection shorted something out. Or maybe the loose connection has nothing to do with this.

I suspect physical damage at this point. I have a speed pickup (CA-DPS) on the front wheel and the CA plugs in directly to the Lyen 12 fet controller via CA-DP. I inspected the connector going from the CA to the controller and it visually looks fine (tugged at the wires to make sure that none are loose, made sure that they are not gunked up with debris, etc. The speed thing probably isn't needed anyway since the wheel has hall sensors but it's still there from a short time that I had a wheel without hall sensors.

I'll see what I can find with a DMM. I'm not necessarily new to putting together electric bikes but I'm not as well versed as most of you in the electrical details on how everything actually works under the covers.

 

[RVD]

After some awesome help from Justin (yeah, that Justin), I don't know exactly what's wrong but I disabled the last wire (blue in my case) and plugged in the CA and it seems like it works now. Of course the disabled wire controls the throttle so CA features such as LVC, amp limiter, etc. will not work but at least I'm no longer riding blind. This is a good enough solution for me since I don't need to go out and buy another CA or controller. I don't know if it's an issue in the CA or if it's something in the controller connection but whatever, I don't think I'll troubleshoot for now.

 

[teklektik]

Good you got it on the road - hope you get back to it to figure this out...

The mod you made to disable throttle control is explained in the Guide in section "3.3 Monitor-Only Mode".