Crystalyte trouble shooting help requested

Had a problem last night. While attempting an alternate route which includes several hills, my motor appeared to stop working 11 miles into the ride. When I twisted the throttle I had no or very little forward momentum. The motor sounded like it was turning, I could hear what I call a growling. This afternoon I put the bike up on the work bench to look into the problem. The wheel is turning but there is a difference than before. If I grasp the wheel and hold it while applying throttle I can hear the growling. A month ago if I did the same thing the motor/wheel would jump in my hand and click on and off. Turns out it was the low torque of the motor and I didn't realize that I could hold it still. There was no growling back then only a popping on and off. Now just growling and it doesn't stop until I release the throttle. Does this sound like a controller issue? I was under the impression that a single hall effect sensor gone bad would cause the wheel to spin erratically and multiple failed hall effect sensors then the motor run wouldn't turn with an instant start controller. I'm not sure of the symptoms of failed mosfets. Ideas appreciated.

Thanks,

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=764

Been there already. Read several times also. I've used Google and the search here, still my question isn't answered in that thread. What are the symptoms of a blown fet? Can the motor run with a bad mosfet in the controller? That thread is a great resource if I need to R&R the controller or mod it. But it's not helping me at the moment. Why does my motor growl today when it didn't do that yesterday? It sounds like the stator is spinning within the wheel but I know it isn't. Haven't ripped any wires loose yet. Thanks for trying to help.

steveo said:

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=764

Click to expand...

If you had some doubt about the Hall sensors in the motor,
then check the Hall signals to see if they are OK.

BTW. What is the version of the controller and the name of the motor?

I have no doubt about the hall effect sensors. If one were bad the motor would be jerky, if it were two then it wouldn't run. Wouldn't run if three or the power or negative were bad either.

What I don't know is the symtoms of a bad mosfet. I did the roll back test as fetcher describes, and it passed.

Do the "rollback" test. Turn on the controller (check that light is on) and try to push the bike (or scooter, whatever) backwards quickly, no throttle.
Turn off the power and repeat the test. If you get a strong resistance to rolling backward with the power on, then the hall sensors and low side FETs are working properly. There should be minimal resistance with the power off. If resistance is still present, check for shorted FETs.

Click to expand...

So the hall effect sensors and low side FETs appear to be good. Are there high side FETs that could be bad? The growling sound I'm trying to describe is similar to when I heard a three phase motor running single phase. I have some time on my hands today. I can swap my motor controller. I just have to do some rewiring for the main cable harness hall effect sensors as they are different on both controllers

The7 said:

If you had some doubt about the Hall sensors in the motor,
then check the Hall signals to see if they are OK.

BTW. What is the version of the controller and the name of the motor?

Click to expand...

It seems that your tests were done without using instrument.

Any chance that you could get hold of an DMM to check for voltage and resistance!?

You're correct. Yesterday I was short on time and did the test without connecting my multi-meter. Today all that is on my list to do is a computer service repair call. Portland is in the middle of a yearly bike celebration called Pedalpalooza. There are around 200 bike events in two weeks. I attended two of them last night and didn't get home till after 1:00am. I have a bit more time today as I don't think there are any events I'm interested in.

The7 said:

It seems that your tests were done without using instrument.

Any chance that you could get hold of an DMM to check for voltage and resistance!?

Click to expand...

Hi

The puma motors growl badly if you get the phases around the wrong way! I have also heard this with 1 x fet blown as well so it is possible that you have blown a fet, have you checked continuity between the phases and battery pos and neg as per fechters thread, the roll back test wont detect 1 x blown fet, you need to check all the phases from the controller with a meter.

Knoxie

Is this a new digital one or an analog controller?

Either way, it sounds like it's not firing on all phases. Usually when a FET shorts, it explodes, makes smoke, and blows your main battery fuse. Kinda hard to miss that.

Sometimes it's possible for one to blow the trace off the board and go open circuit.

Most likely you lost one hall signal. The motor can still sort of run with one out.

knoxie, fetcher thanks for the input. My controller is the old analog type it is a 36-72v 20a dual speed type. I don't recall anything out of the ordinary happening when the symptoms started the other night. The controller is up in the nose of the velomobile. I didn't hear nor smell anything. We were going up some slightly long upgrades. I don't doubt that things were getting hot, I was. The sound just started. Before when I hit stall speed on the motor it would give a kick on and off. That was one thing that confused me before when I'd stall the motor by hand. I understand the stall speed now but there is no more kick just a growling sound. The velo and I probably weigh upwards of 306lbs. I've set it up for the fairly flat lands of the east side of Portland and not the hills of the west side where I was attempting to go. Later on I'm looking into getting another controller from Justin, perhaps a 35a one. First I'm going to jump on the battery group purchase and get the lead out. I don't want to miss that opportunity. I've got the velo up on it's workbench and am getting ready to do some proper diagnosing. I'll post what I come across. Thanks again for the help and ideas.

It could just be a loose connector on the hall sensors or a smashed wire.
Sometimes the hall sensors just go bad for no apparent reason.

See if you can put a voltmeter on the hall signal wires while slowly turning the motor by hand (no throttle).

I also recommend not giving it much throttle when it's malfunctioning or you could blow up something else.

OK, here's the results for the test for the MOSFETS according to fetcher's post in another thread.

To check for shorted FETs, disconnect the phase wires from the controller to the motor. Use an ohmmeter on low range (or autoranging) and measure from each phase wire to each power wire (6 combinations) and look for a low resistance. Any measurement that looks like less than a few ohms indicates a shorted FET. Normally, all the resistances should be in the megohm range, and may vary with time.

Click to expand...

Ohm meter was set to 2M Ω

1.387
1.345
1.165
1.371
1.382
1.150

Number three and number six seem to be quite a bit different to me. Are these out of line? Perhaps I have a questionable FET. I'm going to install another motor controller I have. First I need to rewire the hall effect plug as they are different plugs. Looking forward to some interpretation of the test results. Will post follow up of different controller.

**EDITED 12:55am local time**

Here's the results of the replacement controller:
Ohm meter was set to 2M Ω

1.162
1.145
1.178
1.142
1.123
1.157

I haven't got it installed yet. Still working on the hall effect sensor plug but thought it might be wise and interesting to test it first.

**EDITED 1:37am local time**

Well I've finished up rewiring the hall effect sensor for the new controller. Things appear to be the same. When I hold the wheel at stall speed the motor growls and doesn't kick on and off like before. Tomorrow (Sunday) or later on this early am morning I'm going to test the hall effect sensors in the motor and see what I find.

Without having tested the hall effect sensors I'm beginning to wonder if I'm chasing my tail (once again). I don't understand why the sudden change in the motor noise. Looking forward to getting some of those LiFePO4 batteries and decreasing the weight of my velo by around 35lbs. It'll be nice. That and a 35a controller I'm hoping will hold me.

**EDITED 1:59am local time**
Well I figured it'd only take me a few minutes to check the hall sensors. They look good to me.
0-14.23 blue
0-14.25 green
0-14.24 yellow

I'm at a loss now. Why the loud growling at stall speed? Why doesn't it just bog down and stop? Ideas anyone? After I get some sleep I'm going to pull the rear tail fairing and have a look. I'm getting tire rubbing when I pedal under heavy pressure. Need to figure out why the wheel is rubbing.

Are the units in M Ω?
If so, still unable to draw any conclusion.

Ohm meter was set to 2M Ω on all test. Am going to edit the above post for a final time tonight.

The7 said:

Are the units in M Ω?
If so, still unable to draw any conclusion.

Click to expand...

Hi

OK the other thing is that you have mechanical problems with the wheel? you have wired up the new controller and it still does it, the halls seem to be firing ok? what are you using for a throttle? Magura? Hall effect? have you checked the lead on the throttle, I have seen weird things happen there esp when the signal wires short or get frayed, its worth a try? try a new throttle.

On the mechanical front it Might be worth popping the lid off it and checking the bearing in the hub motor plate as well, also have you changed the phase connectors on the motor? you have changed the controller ones of course, are your using Anderson connectors? they can go bad, change them even if they look ok, Anderson connectors can overtime go bad mechanically and or electrically, worth checking? have you any other motor at all to try the rig on? its not looking like the controller it is looking more like a connection or a lead, most of the faults in electrical systems of any kind come from connections!

Oh and good news on the battery front, where your bike is one of the vehicles that can happily run lead acid because of the aerodynamic efficiency you will see a huge gain in adding in a pack of those lifebatts, really pleased for you! I am also getting 30 of them to make up bunches of cells for all my rides.

Have a go at what I mentioned above, I had a very similar problem on my Kollmorgen motor once, took me ages to find the problem, it was a connection of course!! I was using 5 pin XLR connector for the hall wires, when I pushed the plug and socket together one of the pins was loose and was getting pushed out of the plug body causing intermittent cut-outs!! when you pulled them apart it would pull the pin back out of the body of the plug and it looked ok from a visual inspection!! I only found the problem when I plugged it in with the back of the plug exposed and saw the pin sticking out :lol: .

Good luck and report back with how you are getting on.

Knoxie

The problem remains the same with the new controller. So the fault seems to be relating to the motor windings and/or its connection and possibly the nylon gears.

Would the "growling" noise relate to the nylon gears?
There is one STALL PROTECTION feature in the old C-controller:
If the rotor is not rotating for a couple of seconds after the throttle is ON, the controller will cut-off the power to the motor. It will only be RESET when the throttle is returned to OFF positon.
Thus this show the rotor was actually rotating during the "growling" because it did not activate the STALL PROTECTION.
The rotor was actually rotating but the planetary nylon gears shed and were unable to drive the ring gear when you stopped the hub by hand. Thus making the "growling" noise !?

Have you measured the resistance across any two phases of the motor windings?
If not, you should do it.
Disconnect all the phase wires between the motor and controller.
Use the low ohm range to measure the resistance between
a) G and B;
b) B and Y;
c) Y and G.

Edited: STALL PROTECTION in old C-controller

The7 said:

The problem remains the same with the new controller. So the fault seems to be relating to the motor windings and/or its connection.
Would the problem relate to the gear?

Click to expand...

No, not really. The noise is experienced is at stall speed or when under a load while ridding.

Have you measured the resistance across any two phases of the motor windings?
If not, you should do it.

Disconnect all the phase wires between the motor and controller.
Use the low ohm range to measure the resistance between
a) G and B;
b) B and Y;
c) Y and G.

Click to expand...

Meter set to 200M Ω With the meter set below 200M Ω the readings were zero
a) G and B; 0.9
b) B and Y; 0.9
c) Y and G; 0.9

I just got up from my sleep and have read Knoxie's and your post. Haven't done a thing with the rear wheel yet. When Knoxie wrote about a bad connector and my velo being aerodynamic, it made me think. One thing not mentioned is that since my controller is in the nose area and the motor is in the rear wheel, I've had to extend the cables a good four feet or so. The only way I could get the cables to the motor with out drilling a hole in the rear bulkhead was to run them down the chain tube along side the chain. At this moment I'm not questioning the cable harness yet as it's well wrapped and secured away from the chain be zip ties. What I'm wondering is my choice in wire gauge for the hall effect sensors. I couldn't find a five wire 22g cable so I used a five or six wire phone cord. The gauge of the wires were 22g.

I'm off to another bike event called "Sunday Parkways" or also "Ciclovia" The city is shutting down six miles of road to auto traffic and making a giant circular park for the people to bike, walk skate in the street with out worries of auto traffic (except at the controlled cross streets). Wont be till later on that I can get some more time to look into the rear wheel problem.

Thanks all for the help.

Please read again my edited post concerning "growling" and "STALL PROTECTION".

Would the "growling" noise relate to the nylon gears?
There is one STALL PROTECTION feature in the old C-controller:
If the rotor is not rotating for a couple of seconds after the throttle is ON, the controller will cut-off the power to the motor. It will only be RESET when the throttle is returned to OFF positon.
Thus this show the rotor was actually rotating during the "growling" because it did not activate the STALL PROTECTION.
The rotor was actually rotating but the planetary nylon gears shed and were unable to drive the ring gear when you stopped the hub by hand. Thus making the "growling" noise !?

Not sure if your dual speed C-controller have such STALL PROTECTION!?

mi7d1 said:

OK, here's the results for the test for the MOSFETS according to fetcher's post in another thread.

Ohm meter was set to 2M Ω

1.387
1.345
1.165
1.371
1.382
1.150

Number three and number six seem to be quite a bit different to me. Are these out of line?

**EDITED 1:59am local time**
Well I figured it'd only take me a few minutes to check the hall sensors. They look good to me.
0-14.23 blue
0-14.25 green
0-14.24 yellow

I'm at a loss now. Why the loud growling at stall speed? Why doesn't it just bog down and stop? Ideas anyone? After I get some sleep I'm going to pull the rear tail fairing and have a look. I'm getting tire rubbing when I pedal under heavy pressure. Need to figure out why the wheel is rubbing.

Click to expand...

The FETs look fine. A bad one will be like zero.

Assuming your hall signal readings toggle back and forth from zero to 14v, that would indicate they are working properly.

At this point, I agree with Knoxie that it could be something mechanical. If a bearing wore down enough to allow contact between the magnets and the stator, it could make quite a noise. If they were not actually touching, but the gap was reduced to near zero, they could possibly only make contact when you powered it. Seems like you may be able to detect free play in the wheel if this was the case. Put the wheel up and wiggle the rim sideways to see if there is any slop.

The7 said:

Please read again my edited post concerning "growling" and "STALL PROTECTION".

Would the "growling" noise relate to the nylon gears?

Click to expand...

No gears. The Crystalyte is a direct drive motor.

There is one STALL PROTECTION feature in the old C-controller:
If the rotor is not rotating for a couple of seconds after the throttle is ON, the controller will cut-off the power to the motor. It will only be RESET when the throttle is returned to OFF positon.

Click to expand...

That's what I'm talking about. Before it would cut off the power and then back on as the load was removed. Now it no longer cuts the power only starts to growl.

Thus this show the rotor was actually rotating during the "growling" because it did not activate the STALL PROTECTION.
The rotor was actually rotating but the planetary nylon gears shed and were unable to drive the ring gear when you stopped the hub by hand. Thus making the "growling" noise !?

Not sure if your dual speed C-controller have such STALL PROTECTION!?

Click to expand...

Not my stator turning on the shaft because the wires would be ripped out of the shaft by the spinning stator on the axle shaft.

fechter said:

Assuming your hall signal readings toggle back and forth from zero to 14v, that would indicate they are working properly.

Click to expand...

The reading was like a .08 to 14v

At this point, I agree with Knoxie that it could be something mechanical. If a bearing wore down enough to allow contact between the magnets and the stator, it could make quite a noise. If they were not actually touching, but the gap was reduced to near zero, they could possibly only make contact when you powered it. Seems like you may be able to detect free play in the wheel if this was the case. Put the wheel up and wiggle the rim sideways to see if there is any slop.

Click to expand...

Interesting, I hadn't thought of the stator contacting the magnets. I checked quickly the other day for lateral freeplay in the wheel, didn't feel any but perhaps the magnets could be holding it straight and only the working under load (such as my manly legs) allows the motor to shift. This is the motor that I replaced the axle on. Perhaps the bearing may have been damaged. I'll have to defiantly take a look now.

Take a look inside. Another possibility in your case is the stator may have developed a little looseness between the spider and the axle. This could allow the stator to vibrate, but not enough to spin around the axle. Is there a spline there?

If the controller 'sees' the hall signals changing, the locked rotor protection will not actvate.

Very interesting about the loose stator. When we did the axle swap it did have splines on the axle. I'm going to pull the tail fairing off now and have a look. Will follow up afterwards.

fechter said:

Take a look inside. Another possibility in your case is the stator may have developed a little looseness between the spider and the axle. This could allow the stator to vibrate, but not enough to spin around the axle. Is there a spline there?

If the controller 'sees' the hall signals changing, the locked rotor protection will not actvate.

Click to expand...

Well I took of the tail fairing and what I found wasn't pretty. I have one broken spoke and the axle is stripped. There is what I consider a major gap between the dropout and where the motor attaches. A larger photo can be seen here:http://flickr.com/photos/wmbates/2602462008/sizes/l/ Besides replacing the broken spoke, it broke at the hub end, there doesn't seem much I can do with this motor unless I can get another axle. The machine shop at work made me a custom long front axle for my other motor. I'm going to see if they'd be willing to make me another one to replace this stripped one. Is there any other way to fix this besides removing the axle and turning it on a lathe? If Justin had rear axles it wouldn't take long to replace this one. Last I heard he only had the front axles. I'm kind of depressed now. I can't afford another motor for at least a month.

Hi

Well at least you found the problem what a shame, contact Justin see if he can help you with an axle, I am struggling to see how this could happen though, has the bearing pulled through? do you think its the side loads that may have caused this, its an odd one, was there much space between the drops and the motor plate before? When you say the axle is stripped you mean the threads? how tight did you clamp the nut? looking at the picture it looks like there must have been a lot of space inbetween the drops and the hub, did you fit spacers at all? or did you tighten the nut against the drops forcing the drops to the wheel? if so this is prob what has stripped the threads.

Good luck with the fix

Knoxie