Potentiometer causes motor drive erratic behavior, but only when mounted/grounded

[Jeff.64]

It seem that grounding the pot case is capacitively coupling the resistor to gnd. The leakage current will be insignificant through the DC path if your 20 meg measurements from pot case to wiper and resistor ends is accurate.

It would be interesting to squeeze one lead of a 0.01 µF or 0.10 µF (not electrolytic) capacitor under the pot nut and washer. Then connect the other end to the housing with the pot floating in space and see if the behavior repeats.

Tripping the CFGI is troubling. Something is causing current to flow through the supply gnd wire. Check the resistance of the armature connections to actual gnd. Do the same with the field.

Just for S&G's, try swapping the L1 and L2 connections on the drive board and retest.

Try connecting a gnd wire to the pot as described in the capacitor test, but make a home run connection to gnd and retest.

Look between the heatsink and the bottom of the drive PCB for metal chips or anything that could be providing a path to gnd.

Verify that the neutral and gnd are connected properly to plug that goes to the wall outlet.

 

[FliesLikeABrick]

It seem that grounding the pot case is capacitively coupling the resistor to gnd. The leakage current will be insignificant through the DC path if your 20 meg measurements from pot case to wiper and resistor ends is accurate.

yeah that would seem to be the case, I wish I knew a bit more about the theory of operation of the drive. My comfort with inductive and capacitive stuff is not great, but capacitive coupling generally only occurs with AC, right? Somewhere around here I have a megger, I wonder if I could dig that out and test this at 100-200 volts to see if perhaps it doesn't leak at the 3 or 5 volts my DMM uses, but does leak at higher voltages...

It would be interesting to squeeze one lead of a 0.01 µF or 0.10 µF (not electrolytic) capacitor under the pot nut and washer. Then connect the other end to the housing with the pot floating in space and see if the behavior repeats.

thanks, I will need to see if I can scrounge up any ceramic capacitors around somewhere and try this. It is an interesting idea to test

Tripping the CFGI is troubling. Something is causing current to flow through the supply gnd wire. Check the resistance of the armature connections to actual gnd. Do the same with the field.

With my DMM, no continuity found from the field or armature connections (measured at the terminals on the drive board to test as much of the system as possible)

Just for S&G's, try swapping the L1 and L2 connections on the drive board and retest.

Tried that, no change in behavior

Try connecting a gnd wire to the pot as described in the capacitor test, but make a home run connection to gnd and retest.

Look between the heatsink and the bottom of the drive PCB for metal chips or anything that could be providing a path to gnd.

Will test these in the next batch of tests

Verify that the neutral and gnd are connected properly to plug that goes to the wall outlet.

100% this is done correctly, but always good to check the simple stuff

 

[markba633csi]

I just looked at the hookup diagram again and I think it could be a transient from the relays coupling into the
pot circuit which is exacerbated somehow by grounding the pot, probably by creating an unwanted parasitic capacitance.

The bottom/lower contact of each relay is used to short the wiper of the pot, probably as part of the slow-start function.
It's like an antenna hanging on the pot wiper.
If motor current transients get into the pot wire bundle then that could cause weird stuff. At >10K it's a semi-high
impedance circuit, and sensitive to transients from the motor circuit.

I would try separating the pot leads and the leads from those bottom relay contacts from the other wires
You definitely don't want to have them bundled all tightly together
A small cap like 0.01 or 0.1 uF or a 1.0 uF (non-polarized) might be good also, across the pot outside terminals
In fact, a cap might cure it altogether, without having to do anything else.

 

[FliesLikeABrick]

Well this isn't a particularly satisfying outcome -- but putting in a different potentiometer appears to have resolved this symptom. If I do some more testing, I'll share back some findings. I need to break a bunch of this down and clean it, and may do some testing to see if a resistor or capacitor on the S2 input can help filter the noise as I put this back together.

 

[WobblyHand]

Seems that you had a leaky pot. That's when the grease or lubricant gets contaminated with some conductive material and it finds its way either to the handle or the case. Under some conditions of rotation, temperature, phase of the moon or something else it leaks (conducts). Glad you found it. The fact that you fixed the problem should be quite satisfying, I'd be happy.

 

[FliesLikeABrick]

Seems that you had a leaky pot. That's when the grease or lubricant gets contaminated with some conductive material and it finds its way either to the handle or the case. Under some conditions of rotation, temperature, phase of the moon or something else it leaks (conducts). Glad you found it. The fact that you fixed the problem should be quite satisfying, I'd be happy.

yeah this certainly seems to be the case... but what really threw me off that trail is that I put 2 brand new pots in here (albeit same make/model/source) and had the same problem. Weird.

The motor doesn't run 100% smoothly but I think that's because of how much oil has leaked down into the motor commutator. I expect once I get that all cleaned out this should be a solid setup!

 

[WobblyHand]

yeah this certainly seems to be the case... but what really threw me off that trail is that I put 2 brand new pots in here (albeit same make/model/source) and had the same problem. Weird.

The motor doesn't run 100% smoothly but I think that's because of how much oil has leaked down into the motor commutator. I expect once I get that all cleaned out this should be a solid setup!

Wondering if those pots were from the same production lot, then they might share the same defect. It happens.

Oil on the commutator doesn't help. Let's hope it runs smoothly after that. Clean the brushes and the pockets for the brushes, they're likely full of oil too.

 

[markba633csi]

Very strange problem- let's hope it stays away
Remember the cap trick/re-routing the wires if the symptoms return (post #23)
It seems the circuit is right on the edge of stability