Can't dial in my new 48V regen system

[Bikemad]

I can swap hall wires, go through the phase detection procedure again, and try it again and it is still the yellow wire that slows things down.

Try swapping the phase and motor wires and repeat the test to determine whether the fault is in the motor windings or the controller.
(Yellow motor wire or Yellow controller wire)

You beat me to it!

If with a different combination of phase wires the yellow Controller wire still slows the motor, it looks like the controller is at fault.
If however it's the yellow motor wire that slows the motor, then it sounds like a winding fault.

If the motor is wound in a delta configuration you may have a fault in the winding which goes directly between the Blue and Green wires.
(Most likely a poor soldered connection at the phase wire)
You will still get a resistance reading across the Blue and Green cables, but via the two other windings in series.

See Attachment below.

Alan
 

[Electrobent]

I have triple checked it and it is the yellow wheel wire that slows things down.

[Electrobent]

How could a delta run at all with two wires disconnected? One broken and one disconnected?

[Bikemad]

How could a delta run at all with two wires disconnected? One broken and one disconnected?

It can run because it is a triangle, which allows the current to take the long route when the short route is broken.
If it were a Star configuration this would not be possible.

With a break in the windings as shown it would run on two phases. (As yours appears to be doing)

If you disconnect the yellow lead it becomes a single phase motor, but will run out of phase because it is using two windings in series instead of the correct broken winding.



Look again at the diagram and hopefully it will make sense eventually.

Alan
 

[Electrobent]

OK--I get it I was conflating disconnecting the lead and a break in the coil.

So now what?  OPen in back up and see what is going on under those woven tubes  that you can see in my photo?

[Bikemad]

So now what?  OPen in back up and see what is going on under those woven tubes  that you can see in my photo?

That would be my next step.

I do hope it's just a poor soldered joint, and not a break in the windings!

Alan
 

[Electrobent]

Ok--I took the wheel apart again and disconnected each phase wire from the 8 small enamel wires.  I am hopeful that I am changing something as the solder seems really chalky and crumbles off with heat rather than melting.  Because the solder is so bad, I go ahead and redo all three: unsolder, separate, sand, tin, re-twist, and resolder.  I carefully vacuum out the hub, reassemble, and test. 

I get the same damn thing: pull the yellow phase wire and it slows by half, pull the blue or green and no change.

There were 8 enamel wires on each connection so that means each coil is 4 wires--I find it hard to believe that they could all break but I guess they can. 

At this point I am wishing I had taken the time to split the 8 into the right two groups of 4 so that I could have put an ohm meter across each coil.

So I guess this means that I am hosed: Golden Motor sent me a bad wheel.

How do I get them to send me another one without paying them another $295?

[Electrobent]

I sent the email and am waiting for a reply.

its $295 to get another one here so I hope they take care of me,

[TRK]

Those scope-shots you made earlier, are they from what's coming out of the controller?  If so, could you also make scope-shots of what is coming out of the wheel (unconnected) when you give it a spin?  I'm asking mostly out of curiosity but maybe something out of the ordinary shows up, which would be expected if there is a problem with one or more of the windings I reckon.

[Electrobent]

I had the scope leads in parallel with the phase wires so yes they would be of pulses coming out of the controller.

You are suggesting that we see what things look like when we run the wheel as a generator.

That should work if I can spin the wheel fast enough.

I need to find a source of consistent wheel speed.  Maybe I can rig something up with a drill motor.

I mean need 8 hands for all of this but it sounds like it will be worth doing.

Thanks!

[Electrobent]

OK, I hooked the scope leads up to the 3 possible pairs of power wires coming out of the hub and drove the wheel in forward motion by holding my drill's chuck against the tire.  I was surprised to get three very similar sine waves.  Sorry for the bad photos--I could not find my tripod and held the drill in one hand and the camera in the other.

How can this be reconciled with the slowing being associated with the yellow MOTOR wire only and regardless of which controller wire it is connected to?

Could the sine wave be going around the long way without noticeable loss of amplitude?
Or is it a matter of current and I am getting the wave through some of the four strands but not enough to do the work under load?

Can we post videos anywhere?  I could make a video of the disconnect one phase wire at a time test.

I will be institutionalized before I ride this bike!

[Electrobent]

So no one has any ideas?

Have I already lost my mind?

[TRK]

The first picture is to dark, but if it's a very similar sine wave then I don't see anything out of the ordinary though as that's what I would normally expect to see. 

Anyway, thanks for posting the pictures, sorry I can't be of more help.

[Electrobent]

OK--I started from scratch with freshly charged batts.

All wires color matched--both phase and hall.

connectors individually seated.

I went through the phase mapping sequence outlined in a previous post and found the best forward candidate. I then recorded full speed on my phone.  I disconnected the blue pair and started it up again--no change, I disconnected the blue pair and again no change; but when I disconnect the yellow pair, the motor slows to what sounds like about half speed.

I reconnected the wires so that the green wires on the wheel went to the blue wires on the controller, the yellow wheel wires to the green controller wires, and the blue wheel wires to the yellow controller wires--for both phase and hall.  Again, I carefully seated each contact and went through the phase mapping routine.  This time I first unplugged the green wheel wire from the blue controller wire to notice no change, then the yellow wheel wire from the green controller wire which slowed the wheel down again by about half, finally, disconnecting the blue wheel fire from the yellow controller wire made no difference in the operation of the motor.

[Bikemad]

Here are my thoughts regarding your findings:

It's obvious to me that the Yellow phase wire on the motor is the only one out of the three which is working properly, which is why the motor slows when this wire is disconnected.

As the Blue and Green Phase wires are not affecting the speed, it would appear that the problem is common to both of these wires.

As the most obvious thing these two wires have in common is the windings which go directly between them, it would be reasonable to expect that the fault is with these windings.

We already know that the windings are not shorted to ground.

It is highly unlikely that two pairs of windings have broken on a new motor.


I still suspect that the windings between the Blue and Green phase wires are incorrectly connected:

Option 1
One of the two windings has been connected in reverse, each cancelling out the magnetic field of the other.
(As indicated in diagram B above)
This is the option I would place my bet on.

Option 2
Both of the windings have been connected in a reverse direction, giving a reversed magnetic field.
(Similar to the lower winding shown in diagram B above)

Option 3
Each of the windings has been connected to just one phase wire, and would therefore produce no magnetism at all.
(As indicated in diagram C above)
If you had obtained two low and one high resistance reading (i.e. 0.3, 0.3 and 0.6 ohms) I would have chosen this scenario.

I could be barking up the wrong tree, but I feel sure it's option 1, but I'm not going to rule out option 2 just yet.

Does anyone else have any opinions on these findings?

Alan