Cause of RS motherboard failure: the no-luck wheel.

[I2x2]

@DWOTR I hope you get your parts in and get riding again soon. how many km do you have on your RS & where did you buy your bearings? 
I have a friend how just hit 1000km and he needs his bearings replaced already. I just have 200km on mine but I’ll be putting marine grease in sometime this week hoping to extend the life of the bearings. But it would be nice to have spares to save on the down time. 

[DWOTR]

2 hours ago, I2x2 said:

@DWOTR I hope you get your parts in and get riding again soon. how many km do you have on your RS & where did you buy your bearings? 
I have a friend how just hit 1000km and he needs his bearings replaced already. I just have 200km on mine but I’ll be putting marine grease in sometime this week hoping to extend the life of the bearings. But it would be nice to have spares to save on the down time. 

@I2x2 I think mine was somewhere around 750 or 800 KM. I avoided water quite well (and the bearings were clean)- but I carve quite a lot and would take it just about anywhere. They were covered under warranty- but if you need a set, I would check with gowheels.ca . I assume he either had them in stock or had quick access to them as I think they arrived in just a few days.

Edited May 28, 2021 by DWOTR

[DWOTR]

4 hours ago, Jia Liu said:

For your reference. It's for the new black board, but the motor wire order should be the same.

I'm hoping that it arrives as well. It is now a race between the board and the Sherman to see which is going to get me mobile first. That is a great diagram of the Begode RS motherboard, and I'm positive will help a number of people if they use the search function in the future. It's nice not having to backtrace everything if you want to know what plug goes to what component.

Edited May 28, 2021 by DWOTR

[DWOTR]

So, the new board is in. And it does not fit. I'm hoping that somebody has an idea. The new board is bigger, and there is an obstruction cast in plastic. I don't know what the point of it was, so I don't really want to just cut it out. But I don't see any other way of getting the new board to seat and keep the water out...

 

 

[DWOTR]

And I'm thrilled to find that my head light wires (at a minimum- possibly others) are going to be too short. I'm getting more than just a little frustrated with this whole thing.

Edited July 8, 2021 by DWOTR

[Arbolest]

Would it be possible to swap the aluminum heatsinks? If the mounting holes on the PCBs themselves match up, that should let you install the new board without issue. The only downside is the smaller heatsink on the original board wouldn't be quite as nice for cooling.

Just make sure that if you do swap them you take great care with the thermal pads beneath the MOSFETs!

Edited July 8, 2021 by Arbolest

[DWOTR]

7 minutes ago, Arbolest said:

Would it be possible to swap the aluminum heatsinks? If the mounting holes on the PCBs themselves match up, that should let you install the new board without issue. The only downside is the smaller heatsink on the original board wouldn't be quite as nice for cooling.

Just make sure that if you do swap them you take great care with the thermal pads beneath the MOSFETs!

Thanks. I had considered that- but the daughter boards are not the same size. I could drill the old board and make it fit. But being as the MOSFET's were the original point of failure, I'm really hoping to avoid this. My best hope is that that little circle that appears to just be filled with foam at the bottom of the case be be ground out to make room. I just don't know what it's purpose is...

 

But I appreciate the suggestion. It may come down to that.

[Flying W]

@DWOTR it's my understanding that the msp had it's beeper mounted in that circle. I'm pretty sure our RS's use the same inner shell but the RS has the beeper moved. Gotway just left the old mounting location. There shouldn't be any issue cutting it off and covering that part with a piece of plastic and epoxy. 

I would like a black board but I'm not going through the hastle until my green board dies. 

[yuweng]

 

[DWOTR]

Thank you both. I'm going to pull out the foam tomorrow. If there is nothing that hole- I will clip / grind the bloody thing down and mount the new board complete. I so don't want to tempt fate right now. It looks like the four main screw holes will match up (leaving two bottom ones that have not corresponding holes.) I will post how it works.

[Arbolest]

Good luck! Also, I believe that circular protrusion actually used to be the mounting point for the old buzzers/beepers that Gotway used to use. That's what it was for in my old MSuper V3S+. You should be able to remove it with no worry.

No idea why I didn't just say that to begin with, sorry... 

Edited July 9, 2021 by Arbolest

[DWOTR]

Well, this is interesting. Plugged it in and turned it on. The thing started whinging and you could feel it trying to go back and forth. I got it powered off in time that nothing blew- but I'm pretty certain that it is going to kill the new board if I leave it powered on... I'm at a real loss now. Maybe I'm just being paranoid? I mean they have to balance themselves- they have to try moving just a little... But that whine really has me concerned. 

 

The hall sensor checked out as described for testing. The motor seems to have checked out at the motor shop. I'm probably just paranoid? The only option may be to turn it on and see what comes of it. I don't really know what other options I have at this point... Probably do that and then post the results. for now, I will say that the round part of the housing was easy to grind down. However, the foam plug is in it to prevent water (it is actually a hole through the housing.) So I will have to seal it up. All of the wires are nearly too short (with the headlights being outright too short.) I am going to see if I can tease just a little more length from the taught wires...)

 

Well, as an update, it was not my paranoia. I now have video of it "grinding" and vibrating. Had it turned off before it blew the board again. I assume it has to be the motor as there are not many other options. I will try to find a way to get video off of my phone and post it here. I'm really getting disappointed now. I doubt I will get it repaired before the fall when it is time to put them away.

 

*Yes, the motor wires are off again in this picture...

 

 

Edited July 12, 2021 by DWOTR

[DWOTR]

Well, here it is uploaded to youtube. Any ideas short of a sledge hammer or a jerry can of gasoline? 

 

So, I bought the board rather than wait for any chance of warranty. However, I think I probably need to reach out to the seller and ask what the remedy to this is. As much as I didn't want to spend the money on a board, I recognize that a board is usually "user error." But this cannot be "user error." The bloody thing wants to blow itself to kingdom come... and the board is new. For anybody asking- yes, the wires are hooked up in the correct order.

 

 

Edited July 12, 2021 by DWOTR

[Arbolest]

I know you said that you tested the hall sensors, but is there perhaps a short in the hall sensor wires somewhere?

[DWOTR]

33 minutes ago, Arbolest said:

I know you said that you tested the hall sensors, but is there perhaps a short in the hall sensor wires somewhere?

It is not beyond the realm of possibility. I took the video and walked away in disgust. Tomorrow (the wheel is at work) I will have to take it apart again and pay special attention to that for testing continuity and possibly a short between two wires. I guess it would be nice if it were that simple.

 

It does kind of make sense. When I first turned on the wheel and it was almost perfectly upright the first time there was not much noise. It was only when I "leaned" the wheel that it got upset and threw a fit.

 

*Also that hard noise it made the first time... It almost was like it was slamming forward and backward before the mosfet blew. Almost a cracking or "banging" sound (at least the most that an electric motor can "bang.")

 

It all fits with your theory.

Edited July 13, 2021 by DWOTR

[mrelwood]

You could connect everything else but the main motor leads and do some measurements to make sure the hal sensors and their wires are working correctly. Slowly rotating the tire should show as alternating between 5V and 0V on the colored wires (hal sensor outputs).

You can also measure the board outputs, unless the board goes to failure mode even without the main motor cables.

[DWOTR]

Spoiler

 

 

On 7/13/2021 at 8:05 AM, mrelwood said:

You could connect everything else but the main motor leads and do some measurements to make sure the hal sensors and their wires are working correctly. Slowly rotating the tire should show as alternating between 5V and 0V on the colored wires (hal sensor outputs).

You can also measure the board outputs, unless the board goes to failure mode even without the main motor cables.

I decided to follow through with that plan of slow rotation of the wheel. It is funny- the wheel did rotate very well when I first hooked it up. There was no excessive rotational resistance. Now I will attach a couple of videos. It went back to doing what it did before the new bearing was replaced ("grabbing and releasing" as seen in the first video of this post) and now it is close to locked up (as seen in the second video.)

 

Now, I would say it was bearings... But the bearings were just swapped out. I know somebody will say "you must have hit the race with a hammer or installed them incorrectly." But I actually had the motor shop do it while it was there getting checked out. This is a shop that does the motors for the Toyota forklift dealer and many oil and gas companies. I'm positive they would not have wrecked the bearing (and it is doing the same thing that the first one was which really raises rad flags.)

 

I'm now thinking that the Mosfet blew trying to overpower this extreme resistance to turning the wheel (I am not a small guy and can barely move it by hand.) I think it is all going to come down to a wheel that will not rotate. Assuming it is not the bearing, then it almost has to come down to the geometry of the bearings in relation to one another (the bearing holders / motor covers?) I really can not think of anything else that would cause the new bearings to react to rotation the same way as the old bearings...

 

Again, this wheel is becoming a massive headache for something that was not overly cheap and probably has under 500 Km on it. But if anybody has any ideas, I'm open to them.

[DWOTR]

 

 

 

[Arbolest]

Are you making sure none of the phase wire plugs are touching while you rotate it? If they are it's basically going to turn the motor into a brake where the induced current from the rotation will fight the movement, which looks like what you're dealing with here. If you haven't yet, physically separate the ends of the phase wires (or wrap them in tape temporarily for testing) and try again.

[DWOTR]

Haha. I was sure they were not. But my cobbled together spacer with a zip tie came out. (I picked a stupid time to have apparently ran out of tape...) Yes, two were, and now I don't feel so smart... I actually came back to work to check as I was positive they were not...

 

Good reason to have a whole forum of eyes. Thanks. 

Edited July 15, 2021 by DWOTR

[DWOTR]

On 5/18/2021 at 4:53 AM, mrelwood said:

The hall sensors sense the position of the wheel, and any ongoing issues with them or their cabling tend to just make the wheel try to change the horizontal position very abruptly. Usually the fw recognizes this as an error and will turn off the motor with beeping and an error code.

 But the windings really are bad. Two pinched coil wires are enough to create a short. I would definitely put the rogue windings back in their place, with coil varnish or epoxy if need be.

 What the first photos made me think though was an actual physical short. I have seen rogue solder blobs on some GWs, and if there was one on the heat sink when you assembled the board, that could’ve caused the board to fry.

 

 The coils themselves can be measured by the resistance between the motor main wires as you turn the wheel, and the hall sensors can be measured from their connector when it’s plugged in and powered on. The five cables are common +, common -, and an output for each of the three sensors. The outputs switch between 0 and 5V in relation to their common ground as you turn the wheel. The motor main cables need to be unplugged as you measure.

 As you are measuring the hall sensor outputs, I’d also measure the voltages on the board at the unplugged motor outputs, which I understand you did before as well.

 

I had measured the board with the motor outputs unplugged and it was steady depending on the angle that I tilted the wheel to. I don't really know exactly what readings I'm looking for- but there was nothing "jumping" that might indicate a reason for the vibration. I have some jumper wires for the hall plug so that I can measure the output. From what I can tell, the first pin closest to the hub (in this case, the brown wire) provides 5 volts and is the common +, and the one furthest from the hub (the black wire) is the common -. From my understanding, I should be able to measure the voltage between each of the other three wires (yellow, green and blue) and the common ground looking for it to change the output from 0 - 5V as the tire is slowly rotated?

 

I just want to be sure that is correct before I go experimenting. Oh... And the only jumpers necessary to hook up between the board and the plug would be the common + and common - ? Then turn on the power and measure for voltage?

 

Edited July 16, 2021 by DWOTR

[mrelwood]

 

4 hours ago, DWOTR said:

From my understanding, I should be able to measure the voltage between each of the other three wires (yellow, green and blue) and the common ground looking for it to change the output from 0 - 5V as the tire is slowly rotated?

Exactly!

4 hours ago, DWOTR said:

Oh... And the only jumpers necessary to hook up between the board and the plug would be the common + and common - ? Then turn on the power and measure for voltage?

That will probably result in the mainboard going to a failure mode, since it doesn’t receive sensible data from the hall sensors. But it will probably still give out the 5V, just cutting off the main motor outputs.

 But usually you can measure the hall sensors directly from the connector at the place where the wires enter the connector. Unless it’s covered with glue you can’t get through.

[Unipilot69]

The first rotation video sounds like the tire knobs rubbing somewhere in the shell.

[DWOTR]

On 7/16/2021 at 12:30 AM, mrelwood said:

 

Exactly!

That will probably result in the mainboard going to a failure mode, since it doesn’t receive sensible data from the hall sensors. But it will probably still give out the 5V, just cutting off the main motor outputs.

 But usually you can measure the hall sensors directly from the connector at the place where the wires enter the connector. Unless it’s covered with glue you can’t get through.

Just as an update- you were correct. It hit failure mode. So I put in all 5 jumper cables to make measuring the voltage easier. It was a consistant0 - 3.3 volts for each hall sensor as the wheel rotated. There were very fast changes (almost direct from 0 - 3.3 volts with almost nothing in between.) Supplied voltage to the sensors was 4.4 volts.

 

I'm now at the point where I am going to plug in the motor again, lift it, and hit cut out speed for the heck of it. I want to see what it does, and I'm frustrated. Maybe the marine grease that I packed the bearings with is too thick and it is causing slight resistance forward to back making a vibration as it does not really know what to do? I'm honestly running out of patience with this thing... If that were it, I may cry. (or laugh.)

 

Still would not explain the initial mosfet blowing itself to kingdom come. But sometimes you just have to live with mysteries... Wish me luck.

[DWOTR]

Well, that was fun. I'll link the youtube video below. there is a pretty major wobble in the wheel. So I used a improvised runout gauge consisting of pliers holding a pointer (so not really a gauge.)  The wheel seemed OK, so I went to the rim. There is a pretty good bang in it with a deformity.

 

I have to assume this is from hitting a 90 degree curb at around 40. The wheel worked just fine after that- but maybe when it was apart some stress relieved itself and knocked things off center or the bearing holders / covers warped?

 

With that said, it cannot be completely the ding in the rim. That would not cause the grinding sound at idle when it is not moving. That would almost have to be the motor, itself. Something electrical as the bearings are not turning and could not be making noise until a higher RPM.

 

I will have to post a picture of the bent rim later. Right now, the phone won't send it. For now, you can listen to the noise it makes running itself up to speed. The good news is that it did not appear to blow up the new board during my experiment.

 

 

Edited July 27, 2021 by DWOTR