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!

 

[runinlate82]

I found a stray blue wire in my canbus harness coming from the motor. The line doesn't go anywhere that I can tell. It's just poking out of insulation

 

[Victor Dupont]

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.

Actually this brings me to a question I wanted to ask you guys.

Are the phase circuits supposed to be identical?

This is the equivalent of my initial question in this thread, when I was asking whether the phase wire voltages are supposed to be asymmetric.

One reason why I am asking now, is because I am seeing an asymmetry through the resistance of the resistor between the bootstrap diode and the bootstrap capacitor. In other phases it measures 5.1 ohms, in the yellow phase around 50 ohms.

Could it be intentional? Or is it necessarily the sign of a problem?

Since the other phases' 5.1 ohm resistor is also marked 5.1 ohm, it's pretty clear that there is nothing in parallel that could influence this. Even if there was something in parallel to that resistor and that something went bad, it couldn't cause the resistance to go higher than 5.1 ohms. If this other something went short, resistance would go down. If this other something went open, resistance would stay at 5.1 ohms.

Thus it seems to me that either the resistor in the yellow phase circuit is supposed to be 50 ohms, or that resistor is blown.

What do you think?

 

[fechter]

Yes, all the phase circuits should be identical.
Either the resistor is blown or the wrong value was installed. I've seen several times where a wrong value part was installed on a board (bad quality control).
When testing with a meter, it's good to measure in both directions to pick up diode/transistor junctions.
If the resistor is blown, it may be due to something it feeds being shorted.

 

[Victor Dupont]

Yes, all the phase circuits should be identical.

That is REALLY helpful, thanks a lot!

I'm still finishing to recreate the schematic using my multimeter on the blue phase circuit, and comparing with the yellow phase circuit. I'll let you know as soon as it's done

Either the resistor is blown or the wrong value was installed. I've seen several times where a wrong value part was installed on a board (bad quality control).

Interesting

When testing with a meter, it's good to measure in both directions to pick up diode/transistor junctions.

Ah thank you. That's what I've been doing, but I didn't know whether I was being overly cautious.

If the resistor is blown, it may be due to something it feeds being shorted.

That would tend to confirm my hypothesis then, because that resistor is feeding precisely the part with the capacitor in parallel with the integrated circuit.

Here are some pictures of that particular resistor:










And in comparison, the same resistor in the blue phase circuit:

 

[fechter]

I found a datasheet for the gate driver chip. It shows a "typical circuit", which may or may not be the same as your board.
You can see the bootstrap capacitor and diode.

I also found you can buy 10 of them for $7.00 on Amazon. I've replaced chips like that before. Not super hard with the right tools.

 

[Victor Dupont]

I found a datasheet for the gate driver chip. It shows a "typical circuit", which may or may not be the same as your board.
You can see the bootstrap capacitor and diode.

I also found you can buy 10 of them for $7.00 on Amazon. I've replaced chips like that before. Not super hard with the right tools.

View attachment 379013

Yes exactly, that's the one.

From what I can tell so far, the circuit is almost the same.

I'll share the schematics as soon as I'm done, most probably in a few hours.

Thanks a huge lot again for the help!

 

[Victor Dupont]

Hi guys!

So I'm done recreating the schematics.

Here it is:



As a reminder, in place of "x", in the blue phase circuit, there is a 5.1 ohm resistor.

I also checked continuity between the top right pin of the integrated circuit and every single pad and component pin on and under the whole board: nothing.

Thus it seems to me that it's pretty certain that it's either C1C, the bootstrap capacitor, or U1C, the gate driver chip, that is faulty.

Other elements to hint one way or the other:

I also realised that I get a different resistance reading based on which orientation I place my probes in: 63 ohms in one direction, 74 in the other.

It thus seems to be more likely that it's the chip that went bad. Otherwise I would expect the capacitor to fail symmetrically.

Another measurement adds weight to the suspicion of the chip.

I tried testing for resistance across these two nodes, and staying there. After 20-30 seconds, the resistance increases a little bit, going from 63.0 ohms to 63.1. After maybe 30 more seconds or a minute, it goes to 63.2.

Thus it seems to me to look like a capacitor charging. Initially it is providing a very small path for the current, thus decreasing very slightly the resistance across the probes. As it charges, its resistance increases, the small path decreases, and thus the resistance across the probes increases slightly.

Let me know if you think this reasoning is wrong.

If it is correct, it seems to me that it hints towards the chip being bad.

I know that I'll only be sure after desoldering one component or the other.

I am just trying to take the best guess, so as to avoid damaging a good component when trying to desolder it. If I can start desoldering the component that's bad, I can avoid more problems.

Cheers!

 

[fechter]

I would guess bad driver chip. Only way to tell for sure is to disconnect something.

Last time I replaced one of those chips, I used a tiny wire cutter to snip each leg off the chip close to the body, then remove each leg from the board with a solder vac. If you have a hot air station, you can probably get the chip off in one piece. You have to be real careful not to lift the traces off the board, which is real easy to do when heated to solder melting temp. Idea is to avoid pulling up on anything unless it's fully melted. Soldering the new one on is challenging, but not impossible. You need to avoid bridging the pins with solder. Solder wick is handy if you get a bridge. I used my smallest soldering iron.

Once the driver chip is removed, you can measure the cap again. I bet it's good.

 

[Victor Dupont]

Thanks a lot for all these pieces of advice. They are extremely helpful.

I would guess bad driver chip. Only way to tell for sure is to disconnect something.

Last time I replaced one of those chips, I used a tiny wire cutter to snip each leg off the chip close to the body, then remove each leg from the board with a solder vac. If you have a hot air station, you can probably get the chip off in one piece.

I have neither a wire cutter, nor a solder vacuum, not a hot air station.

However, I found a technique with a beer can. I am learning it, practicing on a scrap board. So far it seems to work amazingly well.

You have to be real careful not to lift the traces off the board, which is real easy to do when heated to solder melting temp. Idea is to avoid pulling up on anything unless it's fully melted.

Thanks a lot for the advice, it might well have just saved me from ruining my board.

Soldering the new one on is challenging, but not impossible. You need to avoid bridging the pins with solder. Solder wick is handy if you get a bridge. I used my smallest soldering iron.

Thanks for this one too, I'll keep it in mind!

I need to practice a bit more with the beer can technique, and then I'll try desoldering at least two legs from the chip, to see whether it was the one causing the short.

Thanks a huge lot for all the help!

 

[fechter]

One other caution:
Discharge the main caps by shorting the main battery wires together before working on the board. If it's been disconnected a long time, it should be discharged, but if it was connected to a battery anytime recently, there could be enough juice to zap a part.

 

[Victor Dupont]

One other caution:
Discharge the main caps by shorting the main battery wires together before working on the board. If it's been disconnected a long time, it should be discharged, but if it was connected to a battery anytime recently, there could be enough juice to zap a part.

Ok thanks a lot, I'll do that!

 

[Victor Dupont]

Hi guys!

So after a lot of practicing, I am very happy to announce that I succeeded in desoldering two pins of that chip, without damaging anything.

Aaaaaaaand the faulty component was...


......


[...Drums rolling...]


......










...

THE BOOTSTRAP CAPACITOR!

In case pictures weren't clear: now when I measure in diode mode across the chip's two relevant pins, I get 604 in one direction, open line in the other. It's exactly the same reading on the blue phase wire's chip.

Conversely, the reading across the capacitor hasn't changed: about 70 ohms in one direction, about 60 in the other.

While I was surprised, it seems clear it's the faulty one.

I am so glad to have reached this step.

Now the question is: how do I know which capacitor to replace it with?

Since there is nothing written on it, and the datasheet is not available online, how can I choose a replacement one?

Thanks again a huge lot for all your help guys!

 

[fechter]

According to AI, the bootstrap capacitor should be at least 10x the gate capacitance. Typical bootstrap capacitor values vary from 0.47µF to 10µF. I would lean toward the larger size.

If you use a MLCC, you don't have to worry about polarity. In really old controllers, I've seen aluminum electrolytics.
25v rating should be enough.

I would try desoldering the cap and test again once removed from the board. That's a pretty rare failure.

 

[Victor Dupont]

Thanks a lot for the swift reply, and thanks again a lot for the help!

According to AI, the bootstrap capacitor should be at least 10x the gate capacitance. Typical bootstrap capacitor values vary from 0.47µF to 10µF.

Interesting. In the schematics of a different controller, the "KU63" or something, here, taken from this guy's page who reengineered it, his bootstrap capacitor is 47 uF. So I guess 10 uF could be too low.

I would lean toward the larger size.

Is there any risk in taking more? Is bigger better for a bootstrap capacitor?

If you use a MLCC, you don't have to worry about polarity.

Ok thanks

In really old controllers, I've seen aluminum electrolytics.

I am not sure I understand what you mean as consequences of that. Does it mean I should check whether that's what I have on the board?

25v rating should be enough.

Ok thanks.

Apart from voltage rating and capacitance, is there anything else that matters? Does the fact that it will be discharged and recharged at a high frequency matter?

I would try desoldering the cap and test again once removed from the board. That's a pretty rare failure.

Ok I'll do that. Anyway the next step is to desolder it.

Again a huge thank you for the help!

 

[Victor Dupont]

Update: I have desoldered the capacitor

Result:

• taken off the board, the capacitor still measures almost like a resistance: 55 ohms in one direction, 59 in the other. The numbers are different from when it was on the board, not sure why.
• conversely, on the board, its pads now measure open line. The pads are still good, I have continuity between each of them and the chip's corresponding pads. If you are wondering why I don't get a diode measurement in one direction, it's because I still have one of the chip's pins unsoldered, so not connected to its pad
• the chip itself is still showing 604 when measured in diode mode across these two pins in one direction, and open line in the other direction, exactly like the blue phase's chip

Thus it seems to me that it's confirmed that it was this capacitor.

Now I just need to find what to replace it with.

Again a huge huge thank you for the help!

 

[Victor Dupont]

Hi guys!

A little update.

I have measured the capacitor using the blue circuit, it seems to be around 7 microFarads.

The way I did it is to place my 1 MOhm voltmeter in series with it, feed everything a 10 V supply, and measure the time it takes to charge to 63.2% etc.

I know that everywhere tells you that you should take the capacitor off the board for measuring, but I thought that I can avoid this risk of damage because:

1. I know the circuit, so I know that the only thing in parallel to this capacitor is the chip
2. I did the same measurement on the chip from the yellow circuit. Since it's now alone, that gives me its capacitance, which I can deduct from the capacitance measured in the blue circuit

In the end I didn't have to deduct anything, because the measurement of the chip alone gave something completely negligible compared to the capacitor. Even after multiplying the resistance by 3 by adding more resistors in series, the voltage instantly showed 0, whereas I had had to wait a minute or so for charging to occur with the capacitor.

Let me know if I'm missing something, but it seems to me that I'm all good knowing the capacitor's required capacity.

Other capacitor characteristics

I read in this note that

It is generally recommended to use low ESR and ESL surface mount multi-layer ceramic capacitors (MLCC) with good voltage ratings (2xVDD), temperature coefficients and capacitance tolerances.

I was wondering what qualified as "low ESR and ESL". After some readings, I am under the impression that any multi-layer ceramic capacitor qualifies as low ESR and low ESL.

Thus it seems to me that I can simply select a 7 microFarad multi-layer ceramic capacitor, with a voltage rating of 25V, and check that its capacitance is indeed the right one at 14 V, which is the voltage that this capacitor is going to experience.

Does this reasoning seem right to you?

Cheers!

 

[fechter]

Yes. I agree with everything. That capacitor will be a lot easier to replace than the chip. I've never seen one go bad before, possibly a manufacturing defect. The exact value of the bootstrap cap is not really critical. It just has to be big enough. It should only be seeing around 12v, so a 25v rated part is good.

 

[Victor Dupont]

Thanks a lot for the swift reply!

I'm glad that things are clear, I'll be able to move forward and replace it.

The exact value of the bootstrap cap is not really critical. It just has to be big enough.

That's also what I thought, but I was confused by these sentences from that same document:

It is important to note that values below the minimum required bootstrap capacitor value could lead to activation of the driver's UVLO therefore prematurely turning off the high-side FET. On the flip side, higher values of the bootstrap capacitor lead to lower ripple voltage and longer reverse recovery time in some conditions (when initially charging the bootstrap cap or with a narrow bootstrap charging period) as well as higher peak current through the bootstrap diode

I had trouble interpreting it, but it seemed to say that there are downsides to the capacitor being too high.

Do you understand what they mean?

Thanks a huge lot again for all the help!

 

[fechter]

It depends on the switching frequency, which I don't know, but would guess is around 20khz.
On a couple of old controllers I have with through hole components, the bootstrap cap is 10uF, 25v. I would go with that.

 

[Victor Dupont]

Hi guys!

Small update.

I retrieved the datasheet for the MOSFETs, and applied the recommendation about sizing the capacitor, taken from the same document:

As a general rule of thumb, this bootstrap capacitor should be sized to have enough energy to drive the gate of the high-side MOSFET without being depleted by more than 10%. This bootstrap cap should be at least 10 times greater than the gate capacitance of the high-side FET.

So the total gate charge of the MOSFET is only 100 nC.

Divided by 10 V, that yields 10 nF. Multiplied by 10 for room ends up being 100 nF.

Thus 10 uF shouldn't be a problem at all, but conversely it seems weird, it seems completely oversized, 100 times bigger than what would be necessary.

 

[fechter]

From that I would assume anything between 100nF and 10uF would work. I wouldn't worry about it too much.
In the past, I've used through hole components soldered to the pads where a SMD part belong (because that's what I had in the junk box).

 

[Victor Dupont]

From that I would assume anything between 100nF and 10uF would work. I wouldn't worry about it too much.
In the past, I've used through hole components soldered to the pads where a SMD part belong (because that's what I had in the junk box).

Thanks a lot, that's really telling.

Speaking of junk, I might be able to get that capacitor directly on a scrap motherboard.

However it's not the case for the bootstrap resistor. It's supposed to be 5.1 ohms, and I couldn't find any of such value on that board, or in my parts.

I do have 10 ohm ones. So I could put two in parallel, and reach 5 ohms.

Do you think that would be ok?

I read that the purpose of this resistor is twofold. One is to protect the bootstrap diode. And second is to avoid negative voltage on the phase wire pin of the chip. I read that in another note here: https://www.onsemi.com/pub/collateral/an-6076.pdf:

In the general case I would have tended to think that 0.1 ohms, being just 2% of the value, wouldn't change much. However it seems that for this particular resistor, the exact value matters.

What do you think?

 

[fechter]

I’m sure 5 ohms would work fine. Even 10 ohms would probably be ok. If the high side fets fail to turn on for a few pwm cycles I don’t think you would notice it. It would be like less than a millisecond delay.

 

[Victor Dupont]

Ah that would be great news! I was fearing the risk would be greater, of burning some other components etc.

So that wouldn't even damage the motor progressively because of small hiccups?

I actually just started searching for a resistor to buy, and I got confused by the wattage requirement.

In the first moments of the capacitor charging, the current flowing through that resistor will be Vdd / R, which would be around 3 Amps, thus a power of around 45 W. Yet none of the surface mount resistors that appear on component sites with 5.1 ohms seem to be able to withstand that power.

 

[fechter]

The bootstrap cap charges up very quickly so the average current in the resistor is low. I would just try to duplicate the size of the original.