Shielded Hall Wires for large DD hub motor

[hias9]

Usually for DD hub motor outrunners, the hall wires go parallel with the phase wires. I sometimes have some vibrations in certain speed ranges at very high phase amps in hall based mode, so I would like to try shielded wire for the hall wires.
What would you connect the shield to?

 

[amberwolf]

Controller main ground (battery negative).

Don't connect the shield to the motor end of things.

 

[hias9]

Thank you. How would you connect the copper shield to a wire? Just leave some excess shield and twist it so it will look like a wire and solder it to a wire which is connected to B-?
I did not find suitable shield solder sleeves for this size.

 

[fechter]

How would you connect the copper shield to a wire? Just leave some excess shield and twist it so it will look like a wire and solder it to a wire which is connected to B-?

Yes, that's how I would do it. Be careful not to let the heat from soldering go back down the braid to the wires and melt the insulation. I use a hemostat clamped to the wire as a heat sink to prevent this during soldering. Even a big alligator clip or needlenose pliers with a rubber band around the handles can do this.

 

[Two4Seven SuprDan]

cat5 or 6 twisted pairs work good for noise elimination but i think shielded twisted pairs might work twice as good in this case.....

 

[larsb]

Since the hall wires are routed close to or even on the coils inside the motor - without shielding - i think that nothing is gained by using shielded wire. It could also be that the problem is stray flux from the coils disturb the hall sensor itself.

But if you try this and it works then please report it here, it’s interesting if it turns out that shielding solves anything.

 

[fechter]

Good point. If the noise is picking up from the phase wires, it would be from magnetic coupling. Shielding won't stop a magnetic field (unless it's magnetic shielding). But it may be worth testing anyway. Sometimes things don't work the way we think they do.

 

[hias9]

The vibrations only occur only if phase current at that moment is above approximately 350A and they only occur within a certain rpm range.
The strange thing is that they don't always occur. Sometimes even at higher phase amps there are no vibrations.

The hall wires travel about half a meter parallel and next to the phase wires, so what I am worried about is inductive coupling coming from the phase wires at high phase current.
Would using twisted pairs for the hall wires also help or would this not have an effect because phase wires are not twisted of course?

At the moment I use AWG 26 wires for the hall sensor wires. Do you think using bigger or smaller wires would have any effect on this?

 

[fechter]

The wire size of the hall wires shouldn't make any difference. Twisting might, as this tends to cancel out induced noise. The physical spacing between the hall and phase wires may be more important, especially if they are close together now. If there is any way to put some distance between the hall and phase wires, even just temporarily for a test, it might do something. 1cm of spacing will greatly reduce induced noise compared to 1mm.

 

[E-HP]

I sometimes have some vibrations in certain speed ranges at very high phase amps in hall based mode,

What are the vibrations like? Something you feel through the drivetrain; and are they accompanied by noises/sounds? Also, I think you implied it, but just confirming that it doesn't occur when running sensorless under the same conditions, correct?

 

[hias9]

I also have vibrations around the same rpm range in sensorless mode, but only with field weakening enabled.
And sensorless mode does not work good at very high phase amps (when stator comes into saturation), so I am using hall based mode at the moment.
So what I wrote in this thread is only about the behaviour in hall based mode.

 

[E-HP]

I also have vibrations around the same rpm range in sensorless mode, but only with field weakening enabled.
And sensorless mode does not work good at very high phase amps (when stator comes into saturation), so I am using hall based mode at the moment.
So what I wrote in this thread is only about the behaviour in hall based mode.

If you have vibrations in sensorless mode as well, then it may not be related to noise on the hall wires. If it occurs based on RPM and phase current (not sure how you are measuring that), and phase current is high when loads are high, so maybe it's a mechanical issue happens based on RPM and load on the motor.

 

[hias9]

To answer the question asked earlier: The vibrations feel like commutation errors.

No, the vibrations/jitter in sensorless occur probably because senserless control just does not work good when the stator is already in saturation. It is different to what occurs in hall based mode. In sensorless mode the vibrations/jitter are reproduceable and when above about 450 phase amps in this case it is better to just forget about sensorless because the jitter gets extreme.
Please just ignore it for this thread, because it is about the vibrations occuring in hall based mode.
I don't have the vibrations in hall based mode on every ride. On some rides I have the vibrations, on some I don't.
But at very high phase amps like 600-650A phase current, it never feels completely smooth.