Need help with Lyen controller

[fiddlerpaul]

I purchased a used Lyen 24 FET 4110 MK II controller and a new geared hub motor (not sure of the make) 48V 500W. After setting it up and trying every possible (36) wiring configuration I get a repeating problem.
The motor runs for a few seconds very roughly and then stops. You then turn off the throttle and on again and it will run for a few more seconds.
I've made a video here https://youtu.be/_UGiU5NnuEU

As I say in the video, the guy who last had it had chopped off the three phase wires quite short and when I went to solder extensions onto them, I was forced to cut them back. On two
of the wires I had to cut back before a heavy metal connector that transitions the thick silvery fine stranded wire from the controller to a thinner copper wire. So I had to solder to the thicker wire.
This is the only reason I can think of for having problems but I don't know enough to say. I have checked all my connections with a meter so I have good current flow.
I programmed the Lyen successfully with the test profile for that controller.
Any ideas? I've sunk a lot into this and have nothing at this point.

 

[Ykick]

Welcome, please add your general location in user profile/avatar.

Stop giving it a lot of throttle. When testing for Hall sensor Phase power combinations you should only barely tickle it with throttle and not for very long. Although Lyen does a good job protecting those controllers you can still damage the expensive FETs.

Here’s some troubleshooting links: http://www.ebikes.ca/learn/troubleshooting.html

You will be interested in the Hall sensor motor testing. Since you have a geared motor you’ll need to turn the wheel backwards to engage the freewheel clutch and spin the motor while looking for change of state with the Hall sensor signals. I suspect there may be a problem with the Hall sensors. However, that ghetto wiring could be an issue too. But start with the Hall sensor and let us know what, if anything, Hall sensor testing reveals?

 

[greg176]

That's a whole lot of controller for that tiny motor. You'd be better off with a 6 fet controller.

 

[Ykick]

That's a whole lot of controller for that tiny motor. You'd be better off with a 6 fet controller.

True, 20-25A would be all the battery current I would throw at that motor. Will it even program that low?

 

[amberwolf]

Some geared hubs have such a high pole count and RPM that they exceed the eRPM limit of common ebike controllers, and require ones that have a high enough eRPM limit to work specifically with them.

Is the problem happening even at very tiny amounts of throttle, at low RPM, or only once it passes a certain speed / amount of throttle?

If the former, it's probably a wiring issue, either connectors or wires inside the insulation (either shorting or open--a broken hall wire is the first best bet in that case).

If the latter, then the eRPM thing becomes more likely.

Blown FETs in the controller is also possible.


Unrelated, but I agree that the controller is way more than the motor can handle, unless you can set the controller to limit the phase current quite a lot--you could easily burn that motor up with that controller.

(I almost destroyed a small 350w geared hubmotor on DayGlo Avenger with just a 6FET Lyen, when hauling one of the big dogs around (Nana, I think) in the trailer, years back, overheating it multiple times. I did eventually break the clutch with it.)

 

[Ykick]

Some geared hubs have such a high pole count and RPM that they exceed the eRPM limit of common ebike controllers, and require ones that have a high enough eRPM limit to work specifically with them.

Is the problem happening even at very tiny amounts of throttle, at low RPM, or only once it passes a certain speed / amount of throttle?

If the former, it's probably a wiring issue, either connectors or wires inside the insulation (either shorting or open--a broken hall wire is the first best bet in that case).

If the latter, then the eRPM thing becomes more likely.

Blown FETs in the controller is also possible.


Unrelated, but I agree that the controller is way more than the motor can handle, unless you can set the controller to limit the phase current quite a lot--you could easily burn that motor up with that controller.

(I almost destroyed a small 350w geared hubmotor on DayGlo Avenger with just a 6FET Lyen, when hauling one of the big dogs around (Nana, I think) in the trailer, years back, overheating it multiple times. I did eventually break the clutch with it.)

Edward assured me if using Hall sensors his controllers will drive practically any pole count motor. I confirmed this with 201 RPM Q128H. These high pole count motors were rumored unable to operate sensor less but they spin up nicely with Hall sensor signals sent back to the controller.

I think Paul might fiddling away somewhere? 'hope he makes it back to fill in some of the blanks?

 

[fiddlerpaul]

I've taken so long because my tester was crappy. I got my neighbours and just now tested it and it passed. Watched the reading shift from around 0 to 5 and back as I slowly turned the wheel.

I should explain my situation a bit here as it really has been bizarre. I started out ordering a ebike kit from a guy in Vancouver. Put it together, rode it for 3 weeks and fell in love with it. Then one day it stopped.
I had no warranty with this guy (stupid). I sent the wheel and controller back to him (I live on Vancouver Island in Victoria) and he tells me the controller is fried and sells me another. This was a cheap, cigarette pack size controller. I put that one in and nothing happens. I send it back to him and he tells me its fried. He sends me another....nothing. I give up on this guy and consult a guy I find on the online classified who says he builds bike batteries. He's an electrical engineer student who likes building big powerful ebikes. He talks me into buying this 24 FET controller from him used. Through various trial and errors we discover the battery BMS is also blown so he replaces that for me and after that doesn't bring success, we decide maybe the motor was shorting things somehow and order another hub motor wheel.
That's where I"m at now....so I've sunk $2500 into this and still don't have a bike.

I"m not super smart electrically like you guys. I'm a retired guy with some handyman/computer abilities and I'm a wicked fiddler. I'm hoping this fellow that has been helping me will have an answer because I'm definitely under water technically at this point.
Whatever help you guys have would be great too obviously.

 

[amberwolf]

The extra info is helpful, though more detail on exactly what testing was done at your end to determine each step would be useful.

We can help a lot better if the questions asked get answered in detail so we can eliminate various things and point you to the next step.

 

[fiddlerpaul]

The extra info is helpful, though more detail on exactly what testing was done at your end to determine each step would be useful.

We can help a lot better if the questions asked get answered in detail so we can eliminate various things and point you to the next step.

Yes detail....well, it will only keep running past a few seconds at super low throttle, about as slight a turn as you can do, but it will do it.

I really screwed up with cranking the throttle while testing...I had no idea this was a bad thing. It started doing all this very early in my testing though, that rough sound and the turning itself off.
It does it on several different wiring configurations.
I should also mention that on a few configurations the wheel would give a short lurch and then stop. In my ignorance I was cranking the throttle after it stopped there as well.

I am using some fairly cheap alligator clips to switch around the 3 alternate hall wires. The black and red are in a connector. There are soldered connections I did to extend the wiring both from the motor and from the controller.
Do you guys know anything about what I mentioned in the OP about having to cut back the phase wire to before the transition from one kind of wire to another? Do you have any ideas about that being an issue?
My friend also is mentioning polls...I have no idea what they are.
Is there anything else you want to know?

 

[Ykick]

You've spent a lot of money with little to show for it. Why not contact the good folks at Grin in Vancouver and seek their advice for testing your gear and perhaps they could recommend a reputable tech to hire instead of friends who merely cost you money?

No offense but I'm not convinced your skill set is up to sorting this out on your own without creating more problems? It would be simple matter for a knowledgable tech to qualify what you've got but the problem could be controller FETs, wiring, or even a defective motor? I think you said it's new motor but still?

What did the Hall sensor test reveal from the troubleshooting link I provided earlier?

 

[fiddlerpaul]

You've spent a lot of money with little to show for it. Why not contact the good folks at Grin in Vancouver and seek their advice for testing your gear and perhaps they could recommend a reputable tech to hire instead of friends who merely cost you money?

No offense but I'm not convinced your skill set is up to sorting this out on your own without creating more problems? It would be simple matter for a knowledgable tech to qualify what you've got but the problem could be controller FETs, wiring, or even a defective motor? I think you said it's new motor but still?

What did the Hall sensor test reveal from the troubleshooting link I provided earlier?

I agree I'm way over my head. I'm someone who needs something fairly plug n play to succeed, not all this complicated wiring process.
I'm not in Vancouver but maybe they know someone in Victoria.
The hall sensor registered everything as it said it should. The meter read from 0 to around 5 switching back and forth as the wheel was slowly turned. The overall voltage between black and red was about 4.8 V

 

[Ykick]

You've spent a lot of money with little to show for it. Why not contact the good folks at Grin in Vancouver and seek their advice for testing your gear and perhaps they could recommend a reputable tech to hire instead of friends who merely cost you money?

No offense but I'm not convinced your skill set is up to sorting this out on your own without creating more problems? It would be simple matter for a knowledgable tech to qualify what you've got but the problem could be controller FETs, wiring, or even a defective motor? I think you said it's new motor but still?

What did the Hall sensor test reveal from the troubleshooting link I provided earlier?

I agree I'm way over my head. I'm someone who needs something fairly plug n play to succeed, not all this complicated wiring process.
I'm not in Vancouver but maybe they know someone in Victoria.
The hall sensor registered everything as it said it should. The meter read from 0 to around 5 switching back and forth as the wheel was slowly turned. The overall voltage between black and red was about 4.8 V

If all 3 Hall signals switched as described, confidence in your motor is high. Being a new motor I would expect that to be the case. But it's a simple test and tells us the sensors themselves to be switching.

Next thing would be to test the controller FETs as described in the troubleshooting guide. I'm beginning to suspect the problem might be blown FETs? Or, ghetto wiring and alligator clips preventing a clean test of the proper Hall/Phase combinations?

Some combinations won't turn at all some will but it's rough as your video captures. Although it might be too late, inserting a 5A fuse between the battery and controller would prevent damage from wrong Phase/Hall combinations.

 

[greg176]

I did a quick search and came up with these guys in Victoria:
http://gotimeelectric.ca/parts--service.html
Other option is to get us to walk you through sorting this out. I'd start with identifying that motor and getting an appropriate controller for it.

 

[fiddlerpaul]

If all 3 Hall signals switched as described, confidence in your motor is high. Being a new motor I would expect that to be the case. But it's a simple test and tells us the sensors themselves to be switching.

Next thing would be to test the controller FETs as described in the troubleshooting guide. I'm beginning to suspect the problem might be blown FETs? Or, ghetto wiring and alligator clips preventing a clean test of the proper Hall/Phase combinations?

Some combinations won't turn at all some will but it's rough as your video captures. Although it might be too late, inserting a 5A fuse between the battery and controller would prevent damage from wrong Phase/Hall combinations.

I'll do that test. I like the idea of the 5A fuse..will look into doing that first.

 

[fiddlerpaul]

I did a quick search and came up with these guys in Victoria:
http://gotimeelectric.ca/parts--service.html
Other option is to get us to walk you through sorting this out. I'd start with identifying that motor and getting an appropriate controller for it.

Thanks. I'll persevere with Daniel who has been loosely guiding me through this. He may have and answer and I'm also conversing by email with Edward.
I also have a friend looking into getting me some expert help through someone he knows. I'm loathe to spend much more money at this point, so would rather not buy
another controller, unless this one could be turned over for a decent amount.

 

[fiddlerpaul]

I just found out that a friend of a friend who is a genius at everything mechanical and electrical has said he will take it and fix it for me for free, just give him a few weeks.
I have good friends, thank God.
I am going to do the FET test now and will let you know how it goes, but as of Friday I will be handing it over to this guy.

 

[fiddlerpaul]

Here's what happened when I tried testing the mosFETs.

I have a power connector from the controller that has a deep female hole in it so I did my best to stick the probe and make contact along the side of the hole.
At both pos and neg it starts out reading .8 before I touch the phase wires. In all cases, pos and neg, when I touch the phase wires it then reads .5

Judging from what the test article says, this is bad...?
I"m not sure about the spinning the wheel backward for resistance test but if I give it a quick flick with my finger it will spin slightly over half way before stopping. When I unplug the controller from the motor, it is very difficult to spin. This is backwards to what the article says, so does that mean it's ok?

 

[Ykick]

Motor phase wires probably touching each other when disconnected from controller.

Entirely possible you may not have found the correct combination?

Did you use this chart? https://endless-sphere.com/w/index.php/Determining_the_Wiring_for_a_Brushless_Motor

 

[fiddlerpaul]

I realized I hadn't disconnected the controller wires from the motor.
But when I did that all that changed was it now goes from .8 to .7 on all tests.
No wires touching each other anywhere.

I tried 36 combinations IOW every one possible as far as I could tell unless there are more I missed and charted the results. Several gave what I've described.
I like the chart...makes it look a lot simpler.

 

[Ykick]

So, which way is it? Harder to turn the motor with the controller connected as compared to when it’s disconnected and no wires touching each other? Or, is it harder to turn when they're disconnected? If the former, that’s almost certainly shorted FETs.

I’m not sure what readings your talking about .8, ,7, etc? It’s either continuity or open (infinity). What reading does your Ohm meter return when you hold the probes together?

 

[fiddlerpaul]

So, which way is it? Harder to turn the motor with the controller connected as compared to when it’s disconnected and no wires touching each other? Or, is it harder to turn when they're disconnected? If the former, that’s almost certainly shorted FETs.

I’m not sure what readings your talking about .8, ,7, etc? It’s either continuity or open (infinity). What reading does your Ohm meter return when you hold the probes together?

It's much harder to turn when the controller is completely disconnected from the motor. With it connected it will spin about a half turn when you give it a light flick.
I put the ohm setting to 200 and it reads .8. When you hold them together it goes to zero. (it's digital)

 

[Ykick]

I don't think I'm gonna be able to interpret some "bizarre things" here well enough to help you? Something ain't translating for me so maybe somebody else can help. Sorry, good luck...

 

[amberwolf]

It started doing all this very early in my testing though, that rough sound and the turning itself off.

Early, but not immediately? Meaning, it worked at first and now it does not?

If that is the case, try to revert to the conditions it was at when it worked (including being sure you have the battery fully charged), and retest.

If you don't know what the conditions were when it worked vs when it did not, then it's tough to say what might be wrong, beyond the possibility of blown FETs in the controller (or some other controller-related issue).

Do you guys know anything about what I mentioned in the OP about having to cut back the phase wire to before the transition from one kind of wire to another?

Wire is wire, so there's nothing wrong with that as long as everything is fully insulated and cannot possibly touch any other wire's conductor or alligator clip, etc.

If anything from one phase touches another phase while power is applied to the controller, especially when throttle is being used, it is possible to damage or destroy FETs and other parts in the controller (gate drivers, etc).


It is also possible, though less likely, that an alligator clip, if not fully clipped onto a wire, could have arcing between it and the wire, causing RF from the sparks (which you might not see or hear), which could also cause damage to FETs/etc., when throttle is used.

It's much harder to turn when the controller is completely disconnected from the motor.

This is with all of the motor's wire ends completely separated from each other, so they cannot touch? If they are touching, they short and make it much harder to turn.

If this is not the case, then your results don't make any sense, as I can think of no case in which a motor would be harder to turn when not connected to a controller with no power to it.

 

[fiddlerpaul]

Ok sorry for the confusion with my metering but it seems I get different results but now I'm getting the results that say my FETs are ok. My digital meter acts weird doing the ohm stuff but works better for volt tests. Went back to my needle meter and if I put the red into the controller power and the black to the phase I get the ohm results that say their ok. IOW it doesn't go to zero from the power it goes to 1 and from the ground it goes to about 9.

It never did go continuously and keep going from the beginning.

Oh and I did have two sensor wires touching each other which did change the wheel turning. Now with nothing touching it turns almost the same with controller or no controller, maybe slightly easier with the controller, but nothing that noticeable.

 

[Ykick]

I felt bad bailing on you brother Paul but it seemed as if I would only create more confusion if we were to continue going back and forth without some common understandings for testing procedures used to qualify controller MOSFETs?

So, I made a 3 min video that I hope might “show” you, and perhaps others, how I test a controller. Is it 100% correct? I really don’t know? I’m a NYC/Broadway guitar player who commutes 6 or more days per week on an eBike and understands just enough Ohm’s law to be dangerous.

This is how I learned to go about it…

[youtube]SFrJ9Uh2GM0[/youtube]

https://www.youtube.com/watch?v=SFrJ9Uh2GM0&feature=youtu.be