DC motor test board strange fault

[bhm1712]

I had a strange fault at work today which resulted in a belt and wanted your guys opinions as to what is wrong.

The test board im using is professionally made. It has 2 DC motor controllers built in. One is 80 VDC and another is variable from 0-300VDC.

i was testing a 180vdc motor from a capping machine which production engineering had asked me to verify if it was serviceable. First off i did a visual check and found no apparent faults with either the cable, connections or brush assemblies etc. I then meggerd the motor to check to make sure the windings were not down to earth, and did a continuity check on the windings.No faults detected.

I connected the motor to the variable DC supply and set it to 180 volts. i switched on the supply to the motor and it began to turn as expected. I touched the motor with my right hand in order to turn it around, and then touched the potentiometer to adjust the speed. As soon as i did this, i got a nasty belt. i was able to let go, so right away after hitting the ESTOP i assumed AC volts present somewhere. So i barrier the area off and get a colleage to spot me. i put on a pair of 1000v gloved and test for voltage. Between the motor casing and earth i was getting 200vAC! and between negative and earth, 280vDC. i verified this in disbelief using a voltage indicator.

to cut a long story short, after further examination i found that the earth to the safeblock connector had become detached. Upon fixing this, the voltage was no longer present on the negative or on the casing.

disconnecting the earth replicated the fault.

That is as far as i got,and so i left the panel locked off. My questions are, why did the RCD on the AC supply side not trip off, and why is the lack of earth causing those voltages to be present?

i know that the negative is usually tied down to earth, and it appears that the lack of earth caused the negative to float up, but that doesnt explain how the motor casing became live when there are no faults on the motor, i checked another motor and the fault occurred again, both motors check out fine! still cannot work out why the rcd didnt trip, and yet it did when i used a fluke multifunction on RCD test

 

[bhm1712]

just to confirm, at the time of testing the motor, no live terminals or cables were exposed, due to the nature od the connection blocks ("safeblocks") so it wasnt just a silly mistake of me touching a live terminal..

http://www.thelightworks.net/images/EGLY003_l.jpeg

 

[ColJack]

280VDC shock and you could let go? my guess would be an induced voltage on the casing.. you completed the path back to earth....

 

[ericmark]

I got some nasty belts from a starter motor and alternator test rig which ended up being generated from slipping belts.

I will guess isolation transformers will be stopping the RCD tripping but sometimes one can't work it out.

Sizewell I was called to a crane because of 1 inch long sparks from hook to items being lifted. The whole crane was steel. Turned out it was wind generating static only cure was nylon strops.

I would want plastic knobs fitting to start with.

 

[Spark123]

Is it servo controlled? I have had a nasty belt from a DC motor which was hovering from a servo drive unit, it is basically a forward/backward motion causing the motor to stand still but replicates a change in current (similar concept to ac) required to induce a voltage to the housing via magnetism.

 

[bhm1712]

no its not servo controlled,

the 280vdc was not between casing and earth which i touched, it was 200vac. i found the 280vdc betwen negative and earth when testing everything afterwards.

 

[NotHimAgain]

Are there any capacitors in the motor connected between line(s) and earth? These could be forming a low pass filter.

 

[bhm1712]

no, there are no capacitors on the motor, the live and negative go straight onto the brush assembies.

i have isolated the 80vdc controller as this was giving me the 280vdc on the negative to earth. However, the 0-300vdc side still has 80 volts on the negative. this comes straight from a variable AC power supply which is then rectified to DC. The power supply is a "cressall torovolt"

i have checked all the earths right through to the distribution board and the switchroom. all fine.

i just cannot get my head round the fact that when you connect an earth to the motor the voltage on the casing disappears, it only becomes live when the earth is disconnected. However as previously stated , the motor windings or any live wire is not down to earth! so where is this voltage coming from?

 

[bernardgreen]

i just cannot get my head round the fact that when you connect an earth to the motor the voltage on the casing disappears, it only becomes live when the earth is disconnected

The voltage on the motor may be induced by capacitive coupling from the windings to the un-earthed frame of the motor. The voltage is there but via a high impedance so unable to source any significant current. The voltmeter if it is a digital one will take almost no current so the voltage remains. Use a moving coil meter and you may find the meaurement show a much lower voltage as the coil takes current to move the meter's needle. Or use the digital meter and connect a resistor ( > 100Kohms ) or lamp between the motor frame and earth and see if the voltagse drops.

 

[bhm1712]

i have used a martindale voltage indicator and the voltage is there on that,

it was enough to give me a nasty belt! made my arm ache for 3 hours.

i always trust the martindale.

i will try the meter and lamp to see what happens there

 

[bernardgreen]

I then meggerd the motor to check to make sure the windings were not down to earth, and did a continuity check on the windings.No faults detected.

Did you rotate the motor slowly by hand while doing these tests ? Some faults may only occur when the armature is in a certain position.

 

[bhm1712]

i have tried with several different motors and the fault occurs with all.

this afternoon i put an instrument ( temperature controller) which requires a 0.9 amp, 100-250v ac supply between motor casing and earth. The voltage on the motor casing powered up the controller. so it is able to draw current and so not a phantom voltage!

what i found later on may explain something, There was a Live wire from the dc supply exposed in the trunking contacting an earthed din rail.

 

[Spark123]

The fact you are measuring 200v AC says there is an AC component to the supply voltage, could just be from an unsmoothed supply.
As I was getting at above, in order to induce a voltage across via magnetism there has to be a change in current- a DC current will not induce a current across - same reason as a transformer doesn't work with DC.