Are these normal phase wire voltages?

[Victor Dupont]

Hi there!

As I am trying to troubleshoot a problem on my electric bike, and got stuck, I tried measuring the voltage on the phase wires, to know whether my controller is faulty.

I would like to know whether that looks normal or abnormal to you, since I have no idea what the correct values are, and I don’t know whether my methodology makes sense.

The way I measured is the following.​

In order to have something that might make sense, I disconnected the phase wires from the motor, so that it doesn’t move. I kept the hall sensors and speedometer wires plugged, so that the controller has the information coming from the motor. I disconnected the brake cutoffs and lamp, to avoid them interfering. I also opened the controller, to be able to measure voltage directly on the circuit board, to rule out potential wire issues.

Basically it’s a pretty normal setup, except for the phase wires that are disconnected and the controller that is open.

I turn the pedal when I want to measure, and then I use a basic multimeter to measure voltage on each phase wire’s solder point.

Again, note that the motor doesn’t move. The pedal movement just ensures that the controller gets the signal to kick off the motor.

In order to change the sensors value, I simply turn the back wheel backwards and measure the hall sensors wire solder points to know where it landed.

Let me know if you think this method of measurement is not usable.

If it's of any use, my controller is a Lishui EPAC Drive System, part number LSW06-90B1CF SB.

Here are my results:​

When idle, meaning pedal is not turning:



When turning pedal:

Some things to explain:​

- I made several experiments for each motor position, and noted the result of each measurement, separated by commas. I made sure to turn the wheel, and thus change the motor position, between each experiment.
- when I include an arrow, such as 0->9, it means that for a few seconds it was at 9, and after maybe 4 or 5 seconds it rose to 9
- twice I had results that were the same as when idle, and controller’s light was flashing differently, not regularly. After I turned the wheel it went back to « normal ». That’s what I indicated as « flashing ».

There are a few things I notice:​

1. when idle, meaning the pedal is not turning, the yellow phase wire has 13 V voltage when measured using DC mode on multimeter. I would have expected 0V.
2. my measurements when pedal is turning seem to hint that in a normal setting, each phase wire would be at either 0 or 35V, which is close to the battery’s voltage, and that the wire with the 35V would be a color that is at 3.3V in the hall sensor cables. This is just my guess, tell me if that’s how it’s supposed to be
3. It seems to me that the most abnormal measurements seem to be on the yellow phase wire, when it’s supposed to be at 35V, and is instead at 6 or 8V

But again, I am just guessing, I have no idea what the correct values are.

Do you know whether these values are normal or not?

I previously tested for shorts between the phase wires, ground and battery positive, to detect a blown transistor, but everything seemed normal.

Other measurement : with motor connected

I also tested voltage on phase wires when they are connected to the motor. That way everything is exactly as in real life, except for brake cutoffs and lamp which are disconnected.

I only tested when it's idle, meaning I am not turning the pedal.

Here are the results:



As you can see, there is voltage even though everything's idle.

If my understanding is correct, phase wires are connected through the windings in the motor. Thus I interpret this as controller sending 13V on yellow phase wire, and because others are connected to it through windings in the motor, they are all at 13V.

Do you know whether these values are normal or not?

For those who are interested in the whole history of this bike’s problem, here it is: Temporary power outages and motor noise

Thanks a lot in advance for the help!

 

[fechter]

That's a pretty unusual failure.
I'd open up the case (if possible) and do a visual inspection of the parts feeding the yellow phase wire. Possibly a bad solder joint, broken wire, or something burnt in the gate driver feeding the FETs.

 

[Victor Dupont]

That's a pretty unusual failure.
I'd open up the case (if possible) and do a visual inspection of the parts feeding the yellow phase wire. Possibly a bad solder joint, broken wire, or something burnt in the gate driver feeding the FETs.

Thanks a lot for the swift reply!

I'd like to react to your sentence: "that's a pretty unusual failure".

Are you referring to the asymetry in phase wire voltages? or the fact that the light bulb didn't light up?

I started this thread because I didn't know whether the measurements above made it possible to conclude that the controller is faulty,.

I have been yearning to open the controller to investigate more, but I was weary to do this as long as I didn't know WHETHER it was bad.

I didn't know whether these measurements were normal, whether my methodology made any sense or not, and thus because of that, I didn't know whether I could conclude that my controller was bad.

Without knowing that it is bad, I was weary to open it and risk damaging it if it is fine.

Thus are you saying that these measurements are a clear indication that the controller is faulty?

Thanks a lot again for the help!

 

[TommyCat]

Well done on your testing!

So if I understand the aim of the test correctly, I am guessing that the leak that is causing these unexpected 13V is a very small one.

Yes!

If I follow your results correctly, with the light connected, you didn't ever get it to light, or any voltage reading on the YELLOW phase output...? This was in the static or no call for motor operation.

The next step would be to check the voltage output of the YELLOW phase output with the light connected as above, but calling for power. Check for voltage output at all of the 6 step control positions. Motor phase wires still disconnected.
I'm speculating that you will never get a voltage reading.

If you didn't do this, go ahead and do so. If your statement below holds true, my response to it holds true.

Anyway there remains the fact that the yellow wire never gets 35V, but at most 9V, whereas the others are getting 35V at some point.

If this is true, I would conclude that the controller's YELLOW phase, battery positive mosfet is unable to properly close the electrical contact when energized.

You could verify what voltage the mosfet's gate is getting and when, if desired.

But this should leave no doubt that the controller needs repair or replacement.

 

[Victor Dupont]

Thanks a huge lot again for the reply, the help, and the hope it brings me!

Well done on your testing!

Thank you

Yes!

If I follow your results correctly, with the light connected, you didn't ever get it to light, or any voltage reading on the YELLOW phase output...? This was in the static or no call for motor operation.

Correct. Just one precision: I had all phase wires connected to the motor, thus connected to each other. I plugged the light between the yellow phase and battery negative, and same for the multimeter, but in essence it doesn't change anything which phase I chose because all phase wires were connected through the motor.

I think the test is the same as with phase wires disconnected, since I also read 13V when they are connected to the motor, and no pedal turning.

The next step would be to check the voltage output of the YELLOW phase output with the light connected as above, but calling for power. Check for voltage output at all of the 6 step control positions. Motor phase wires still disconnected.
I'm speculating that you will never get a voltage reading.

If you didn't do this, go ahead and do so.

Got it.

I just would like to confirm my understanding of the rest of your message before:

If your statement below holds true, my response to it holds true.


Victor Dupont said:
Anyway there remains the fact that the yellow wire never gets 35V, but at most 9V, whereas the others are getting 35V at some point.


If this is true, I would conclude that the controller's YELLOW phase, battery positive mosfet is unable to properly close the electrical contact when energized.

You could verify what voltage the mosfet's gate is getting and when, if desired.

But this should leave no doubt that the controller needs repair or replacement.

I would like to avoid any misunderstanding, one direction or the other.

This statement of mine is not new, it's just my rephrasing of the measurements I made initially and shared in the first message of this thread.

Was there a misunderstanding about it?

Second misunderstanding avoidance: when you say "But this should leave no doubt that the controller needs repair or replacement.", are you saying that in the current state of things, even without doing more tests, I can conclude that the controller is faulty?

Again, thanks a huge lot for all your help, I really appreciate it.

 

[Victor Dupont]

Ah, maybe I am understanding what you meant.

You meant that if I STILL only get 6 or 9 V at most on the yellow phase wire with the light bulb connected, in the same setting as my initial test, THEN there is no doubt that the controller is faulty?

 

[TommyCat]

You meant that if I STILL only get 6 or 9 V at most on the yellow phase wire with the light bulb connected, in the same setting as my initial test, THEN there is no doubt that the controller is faulty?

Please write testing setup hardware configuration before results to eliminate confusion and make a summery of all testing easier…

 

[Victor Dupont]

Please write testing setup hardware configuration before results to eliminate confusion and make a summery of all testing easier…

Ok thanks a lot!

I'll do that right now.

Just so that I understand the goal of this test: the idea to add the light bulb and do the same measurements is to simulate a load, and thus be closer to real life?

 

[Victor Dupont]

Sunday morning's test report has arrived!

Here's the data:

The light bulb never lit up. The measurements are the following:



The way I would summarize it: it's the same results as before, EXCEPT for yellow phase voltage in the two "weird" positions, where it used to be 6 or 9V, and is now 0.

I also measured gate voltages for the two yellow phase MOSFETs. Again the only change compared to before is that the high MOSFET's gate used to be at 6V or 13V in the two "weird" positions, and is now at 0V.

If my understanding is correct, it seems to me that your idea to add a light bulb was a really good one! It looks like it made the test results clearer, and thus the methodology more usable and robust.

Again if my understanding is correct, I feel like it would be interesting to also add a light bulb to each phase wire.

However, if we can conclude without doubt that the controller is faulty right now, I would rather unglue the wires and start troubleshooting on the board, because these tests on the bike take a lot of time.

The next step would be to check the voltage output of the YELLOW phase output with the light connected as above, but calling for power. Check for voltage output at all of the 6 step control positions. Motor phase wires still disconnected.
I'm speculating that you will never get a voltage reading.

If you didn't do this, go ahead and do so. If your statement below holds true, my response to it holds true.


Victor Dupont said:
Anyway there remains the fact that the yellow wire never gets 35V, but at most 9V, whereas the others are getting 35V at some point.


If this is true, I would conclude that the controller's YELLOW phase, battery positive mosfet is unable to properly close the electrical contact when energized.

You could verify what voltage the mosfet's gate is getting and when, if desired.

But this should leave no doubt that the controller needs repair or replacement.

With these results, do you confirm that there is no doubt anymore that the controller needs either repair or replacement?

 

[fechter]

For sure looks like a controller problem.
Not sure if you can get good gate drive readings on the high side FETs due to the bootstrap circuit, but there should be some output at some position.
I would pull the controller and try to inspect the board. If you're lucky, you might find a bad connection or other obvious problem. If not, you don't have much to risk since the other option is to replace the controller.

 

[Victor Dupont]

For sure looks like a controller problem.
Not sure if you can get good gate drive readings on the high side FETs due to the bootstrap circuit, but there should be some output at some position.
I would pull the controller and try to inspect the board. If you're lucky, you might find a bad connection or other obvious problem. If not, you don't have much to risk since the other option is to replace the controller.

Amazing, thanks a lot!

When I started doing these measurements, I had not read anywhere of such a way to test a controller.

Is this methodology something that you have done yourself? Or do you simply find it to make sense?

Thanks again a lot for the help!

 

[fechter]

I've done plenty of troubleshooting on dead controllers. Back in the analog days, things weren't so reliable, especially when pushing them 2x their design rating.
I guess it would be nice to write up a guide similar to the Grin document and put it in the knowledgebase.

So, if controller is bad, no risk in taking it apart. If the soldering on the yellow phase wire to board looks OK, then I'd suspect a gate drive problem. Depending on the design, it may be impractical to repair the gate driver parts. New controllers are not that expensive these days.

 

[TommyCat]

With these results, do you confirm that there is no doubt anymore that the controller needs either repair or replacement?

With no power going to the YELLOW phase at all, and the addition of no power coming to its mosfet from its gate driver.
The above statement is certainly confirmed.

Very nice Sunday testing results, well not so much for the controller, but well done. Especially keeping the steps in order, NICE.

If my understanding is correct, it seems to me that your idea to add a light bulb was a really good one! It looks like it made the test results clearer, and thus the methodology more usable and robust.

Again if my understanding is correct, I feel like it would be interesting to also add a light bulb to each phase wire.

Yes, adding a bit of a load seems to have cleared it up nicely.

It would be very interesting to add additional light bulbs to the other phases. But as mentioned before these would have to be able to handle the voltage provided by the good working mosfets. I think that your comments have indicated you're using a 36volt nominal battery. So, 40-volt D.C. bulbs would be the best fit. But for fun you could just use household 120volt A.C. bulbs to see what you could get...

This is not a new idea of using lights to troubleshoot as there are several models of inexpensive E-bike testers that essentially do the same thing.
Probably using LEDs but giving some indication of a controller's phase output.

Examples of E-Bike Testers...

I have not used one, but if you decide to try it, I think it would make a second good write-up for you of the results achieved.

My goal in this thread was to help discover good testing procedures using a meter.

To that end a quick question for fechter.

Measuring the voltage is prone to those "ghost" readings that are meaningless.

Are you saying in your troubleshooting experience, that often erroneous "ghost" or other voltage readings are obtained from a controller's phase output wires during such testing as described below? Even from known good controllers.

Using the phase wires disconnected from a hall sensored motor, with the hall sensors connected, and calling for operation, moving the motor thru the 6-step commutations as testing procedure requirements.

Do you have any other good procedure(s) testing tips using a meter to identify a bad controller besides the mosfet checks?


Regards,
T.C.

 

[fechter]

Even a little dirt or moisture is enough to throw off readings to a voltmeter. A light bulb works, but I would generally use something like a 1k resistor across the meter probes. Mosfets have some inherent leakage, which is usually in the microamp range, but that's enough to give a meter reading.

When testing controllers, I first see if it runs the motor and what the no-load battery current is.
Detecting an open circuit mosfet bank is hard with just a meter. It's also a pretty rare failure mode. Usually the mosfets short, which you can test with an ohmmeter. I have an oscilloscope that makes testing easier, but many folks don't have one of those.

 

[Victor Dupont]

Hi guys!

This message is not a request for help. I would just like to share with you two things I find really funny.

On Sunday, after I had your confirmation, I unglued the phase wires and pushed them a bit.

The first thing I tested for continuity was a capacitor near the yellow phase wire, in diode mode. Instantly: beeeeep!

How incredible is that?

The very first spots I placed the probes were the nodes surrounding the problem.

Now I know that there might be reasons for a short, but the thing is, this capacitor is labeled C1C. I placed my probes around the capacitor labeled C1A near the blue wire, and C1B near the green wire, and there was no short. Thus I am pretty convinced that's an indication of the problem.

Again, I know this might not be this very capacitor that is faulty, but it could be something else in parallel. So I know I have to investigate more. So far i seems to me that the only thing connected to these two points is the gate driver integrated circuit, so it seems to me it's either that capacitor or that chip. But I'll have to dig more. Especially since I don't see how it could yield the results I measured.

I don't want to turn this thread into the troubleshooting of my controller. I just wanted to share with you how fun it is that the very first place I tested is the place with a problem.

Second thing is that I am incredibly lucky! These two components are the ones closest to the edge of the board, and least surrounded by other components. Thus they are the two most convenient components to replace on that side of the board.

Incredible, isn't it?

 

[fechter]

Can you post a picture of the controller board? I'm just curious.
No reason this can't be a troubleshooting thread.

 

[Victor Dupont]

Can you post a picture of the controller board? I'm just curious.

Well there are still wires above it. I can move them when I want to access one area, but I can't see more than one specific area at a time. Would you like a picture of a specific area? The part close to the yellow wire I presume? Or several pictures for the whole board?

No reason this can't be a troubleshooting thread.

I was thinking of keeping this thread about "what are normal phase wire voltages", and a potential methodology to detect a faulty controller with just a multimeter, so that other people wondering the same thing in the future can have the answer rapidly.

But if you have thoughts on the controller of course they are welcome!

 

[fechter]

Start with the area around the yellow phase wire. Maybe an overall pic so I can see what kind of controller it is.

 

[Victor Dupont]

OK!

Here they are:

 

[fechter]

Interesting layout. It has current measuring shunts for each phase but not on the battery.
You’re right, it’s hard to see with all the stuff in the way.
Even if it was just a bad driver transistor it would be hard to replace because the parts are so small. The failure could possibly be in the main microprocessor as well.

Time for a new controller.

Thanks for posting the pictures.

 

[Victor Dupont]

Thanks again for your thoughts!

Interesting layout. It has current measuring shunts for each phase but not on the battery.

I am AMAZED that you are able to see this from such a small amount of information!!!

You’re right, it’s hard to see with all the stuff in the way.
Even if it was just a bad driver transistor it would be hard to replace because the parts are so small. The failure could possibly be in the main microprocessor as well.

I am surprised by this possibility.

I felt pretty confident that the short I had measured was a failure, and it stands on the opposite side of the gate driver integrated circuit compared to the microprocessor.

Are you saying there could ALSO be a failure in the microprocessor?

Time for a new controller.

Actually I would like to try and repair it. If it's just changing the capacitor that is close to the board, it seems doable to me. You don't think so?

Thanks for posting the pictures.

You're very welcome. Thanks again for the help!

 

[fechter]

It's rare to see a capacitor fail by shorting, but not impossible. More likely something else it's connected to. Disconnect either side of the capacitor (whichever is easiest), then measure the cap again.
I couldn't really make out the gate driver parts from the picture, but you can visually trace the high side yellow mosfet gate to see what's driving it. High side will be the one that goes to the B+.
Some designs use discrete transistors and some use IC driver chips. The high side will also have a "bootstrap" circuit, which may be the capacitor you're seeing shorted.

 

[Victor Dupont]

Thanks a lot again for the reply and for the help!

It's rare to see a capacitor fail by shorting, but not impossible.

From my recent readings my understanding was that big round capacitors fail by going open, but small ones, which are usually multi layer ceramic capacitors, fail by going short.

But then I don't have any experience, it's just what I learned recently.

More likely something else it's connected to. Disconnect either side of the capacitor (whichever is easiest), then measure the cap again.

Ok thanks.

I was thinking of doing that if I didn't find another way to check.

Because I am afraid to damage other things while doing so, I am first building a schematic of the board using the other phases' circuits, then checking everything I can on the yellow part with it.

I couldn't really make out the gate driver parts from the picture, but you can visually trace the high side yellow mosfet gate to see what's driving it. High side will be the one that goes to the B+.
Some designs use discrete transistors and some use IC driver chips. The high side will also have a "bootstrap" circuit, which may be the capacitor you're seeing shorted.

It's the part that you see on the last picture. That smaller chip on the right is a gate driver integrated circuit. I know it because I searched the web for what is written on it, and got its datasheet.

From that chip's datasheet, it seems that the capacitor I mentioned, which is the one you see right above it, is indeed the bootstrap capacitor.

The diode on the left of the chip is the bootstrap diode.

And then in between there is a component, which I cannot read. In other phases circuits it's a 5.1 Ohm resistor, between the bootstrap diode and the bootstrap capacitor. But in the yellow phase circuit, I can't read what's written on it. I don't know whether it's because I scratched it unintentionally with my meter's probes, or because it blew. Testing its resistance I get 50 ohms I think.

Resistance measured across the bootstrap capacitor is 70 ohms.

Thus one hypothesis I had was that the bootstrap capacitor went short, the smaller resistor thus blew up, and now bootstrapping is not happening.

I am not 100% sure yet, I still have more checks to do on the board.

 

[fechter]

You can compare readings with the corresponding parts on the other phases.
If something shorted you’ll have to start disconnecting things to find the shorted part.

 

[runinlate82]

I've never used one of these sites before but I am desperate for help. I have a bafang m625 mid drive kit and it won't turn on.

 

[n00b]

Pull up youtube; grin posted a really good tutorial on how to test and replace MOSFETs on a controller since that's typically a primary failure point on them.

Replacing MOSFETS? I surely do not know many people that can do that.
Maybe 2. In the world. Yeah, I'd just get another controller.
I doubt ShamWow can replace mosfets. He can design where they're supposed to go, and what they need to be, but probably not take one off and put one back on.
You got to show me this mosfet replacing stuff. I gotta few things that could use that.
It's gotta require a hot air station, I bet.