Controller suggestions??

[Archer321]

Because the red wire has power already and with the jumper installed it is sharing that power with the orange wire.
Adding a 48v source to an existing 48v source is pointless.
However, just to attempt to keep this moving forward, I have complied.

 

[Archer321]

I tried that too.
No difference.
Besides, as the red wire has battery voltage on it already, what would change?

Um, the MCU is powered by the orange wire. So basically nothing will happen if either the orange or red "ignition" wires are not both powered by your (switched) battery voltage. Why do you doubt us?

M

P.S. for those needing a little hand-holding, the wiring should look more like this:

redo PV wiring for IYI.jpg

Is this a close enough approximation?

(Still doesn't work)

 

[kiwifiat]

The "ignition" connector contains two small wires, one red, and one orange. The red has battery voltage on it and the orange has nothing. When I jump them together my CycleAnalyst powers up but nothing else happens.

Proving beyond doubt that the red wire on the ignition connector is directly tied to battery positive and that the real issue lies elsewhere. I am inclined to suspect that your controller has NOT been programmed for sensorless operation. I would like to see a closeup shot of the throttle connectors to confirm the wires are in the correct order on both sides.

 

[Archer321]

The "ignition" connector contains two small wires, one red, and one orange. The red has battery voltage on it and the orange has nothing. When I jump them together my CycleAnalyst powers up but nothing else happens.

Proving beyond doubt that the red wire on the ignition connector is directly tied to battery positive and that the real issue lies elsewhere. I am inclined to suspect that your controller has NOT been programmed for sensorless operation. I would like to see a closeup shot of the throttle connectors to confirm the wires are in the correct order on both sides.

Thank you for understanding that.
And good thought on the throttle wiring.
I checked it and noticed that while it is a direct plug-in, only the red wires lined up. So I depinned the connector and lined up black to black and white to green like this...


Now finally, something happens!
The motor jumps and twitches and draws a lot of power on the CycleAnalyst.
Like it's timed improperly.
Like it's looking for hall signals it's not getting.

At least it's progress...
What now?

 

[kiwifiat]

Being sensorless I would expect it to run with any phase wire combination but you could try this:

You know that Yellow --> Black, Blue --> Blue, Green --> Brown does not work. Try each of these in turn:
Yellow --> Black, Blue --> Brown, Green --> Blue
Yellow --> Brown, Blue --> Blue, Green --> Black
Yellow --> Blue, Blue --> Black, Green --> Brown

If none of those work I would recommend you buy the programming cable and confirm everything is properly configured. It will come in handy in any case for tuning.

 

[Talon]

This diagram should straighten you up 8) https://endless-sphere.com/forums/download/file.php?id=171606

 

[fechter]

In sensorless mode, any combination of phase wires will run. Half the combinations will be forward, half reverse.

Once the motor gets spinning, the BEMF should be enough for proper sensorless operation. It might have trouble from a dead stop.

 

[Archer321]

Yeah, no combination of phase wires made it run any better.
If this controller can't start this motor from a standstill then it's worthless to me.
I'll try spinning the motor up with a drill though just to check.
I'll let you know.

 

[Powervelocity.com]

Hey Steve and everyone on this thread. Thank you for being patient and helping.
I would have jumped in earlier if I was aware of the discussion.

Let's see, just to sum up a few things.
The controller was programmed and tested before shipping. It was programmed for 48v nominal. That means that the low voltage cutoff would be somewhere around 36v and high voltage cutoff at 54v.

It ran standard test fine on a 1kw hub motor that I use for testing. I must admit I tested with hall sensors only but the controller is dual mode - I have visually confirmed with another controller from the same run batch. Dual mode controllers have an additional comparator chip on the right side of the controller PCB. Sensored models have this area unpopulated.

No special programming is required for sensoreless operation. Disconnecting/not connecting halls would make a motor run sensorless. The other thing is that a throttle must be connected before you turn the controller on. If you connect the throttle after it was powered on, it will go into failsafe mode. So, it needs to see the right starting voltage from the throttle (normally 0.8-1v). That's why it may not work with POTs out of the box.

Anyhow, let me go back and double check the sensorless mode on a controller from this same batch. I'll report back soon.

 

[Archer321]

Thanks for the help PV.

 

[Archer321]

So I have resolution...
PV told me to open the case, pull the board, and cut a connection at a particular spot.


After that the motor runs fine.
I was not particularly excited to crack open a magic box I had just paid for, but he's going to send me a Bluetooth interface module for it as compensation for my trouble. That should make setting adjustments easier in the future.
Unfortunately, I lost too much time trying to get this up and running so I won't have it ready for my Electrathon race on Saturday. My old setup will have to do. I guess there's always next year for the new motor.

Thank you to everyone who tried to help me walk through this problem.