Isolation procedure for oscilloscope?

[potcore]

Hi,

I am getting back into drive testing and am hoping to clear up a safety issue first, due to a previous accident...

Myself and a colleague were doing tests on electric drives. The drives had a three phase 415V (rms) input which was fed to a 6 pulse diode rectifier. This then produced the Drives 600V DC Bus which was switched by the inverter to supply the servo motor.

Some of our tests necessitated connection of our scope leads across this DC bus. -Or sometimes just the scope "ground" clip to the "negative" rail of this DC bus.

I was using a Fluke scope which was plugged into the mains (240V RMS). Only two wires came from this plug into the scope (so obviously there was no earth wire coming into the scope).
My colleague was using a Tektronix TDS210 scope which had a 3 pin plug for its mains power connection. My colleague was told to cut off the earth connection in the 3-pin plug, and also to plug this plug into an isolation transformer, which in turn was plugged into the 240V RMS mains electricity socket.
(The isolation transformer was inside a plastic housing some 30cm cubed -it was specifically an "isolation" transformer for 240V mains electricity).

I could not see why it was necessary to use the isolation transformer as well as cutting off the earth connection in the plug. -After all, i was not using an isolation transformer and was having no problem.

Anyway, another worker (accidentlly) swapped my colleagues supply lead (the one with the earth connection cut off) for a lead with the earth pin connected. My colleague then applied the 415V rms mains to his drive with his scope connected across the drive's DC Bus and the drive promptly exploded in my face. (no lasting damage to me but i can still see the white flash in my "mind's eye"). The thing is, he had been using the isolation transformer even though his earth lead was (unbeknown to him) connected in.

I cannot understand why it exploded since he was using the isolation transformer. Also i was not using an isolation transformer and was getting away with it. (maybe there was an isolation transformer in the "wall-wart" plug of my scope -but i doubt it because it was only small (~palm size).

Does anyone know why the drive exploded and why The isolation transformer did not prevent the explosion?

 

[LionelHutz]

The isolation transformer was probably only isolating the power and not the ground connection.

The Fluke scope is likely built with isolated inputs to allow that type of connection. There are some portable TEK scopes that also have isolated inputs and can have their ground leads connected to a hot circuit without a problem.

Use this as a warning and get smart. Buy isolated probes for the scopes, at least rated >1000VAC and CAT II. You do know that using that isolation transformer allows the whole scope case to be at the same voltage as the probe ground wire? It will also still blow up when you decide to pull out the second lead and connect it's ground to a different potential.

Personally, I'd drop kick that "isolation" transformer into the nearest dumpster and never consider using it again. That's a "stuck in a corner, must make the measurement right now and it's my only option" type of solution.

 

[BobbyNewmark]

Ah, so Test Your Ability To Avoid Four Letter Words Day falls on April 9th this year.

"My colleague was told to cut off the earth connection in the 3-pin plug, ..."

Classic last words.

Unless he was also told to lock the scope inside a transparent insulating box, the person doing the telling should be sacked. The TDS210's enclosure is not designed to provide safety isolation from its chassis. Someone will inevitably be electrocuted by touching the scope when the DC bus is energized.

"(The isolation transformer was inside a plastic housing some 30cm cubed -it was specifically an "isolation" transformer for 240V mains electricity)."

And is this transformer meant to provide perpetual life-critical isolation at 600 VDC? Or is it just meant to reduce annoying "hum" in audio applications? If the latter, it is all too likely to fail in service, turning the 240 VAC mains into 600 VDC, spreading death and destruction throughout the building.

"I cannot understand why it exploded since he was using the isolation transformer."

It apparently passes through the ground connection, and the TDS210 is not designed to have 600 volts flying around inside it.

You should to know that the first batches of TDS210's had a safety recall. The connection between the internal ground and the earth pin was made with a too-small conductor. It could burn out if you accidentally touched the ground clip to a power supply, allowing the chassis to float to high voltages. Tektronix recalled and repaired a huge number of scopes at great cost, TO PREVENT EXACTLY WHAT YOU ARE DOING ON PURPOSE.

Buy a proper safety-rated isolated probe. The cost is nothing compared to what a weeping widow would be awarded by a court.

 

[Skogsgurra]

Yes, never use an isolation transformer.

For two reasons:

1 The reason you just witnessed.

2 Because the scope ground will still be connected via transformer capacitance to ground. Not DC connected, Not 50/60 Hz connected, but kHz - MHz connected.

Any careless connection to the live inverter (not only high-side gate and drain, but also low side, if low side is not tied to ground - which it is not) is likely to cause unwanted effects. Like blowing it up again.

Always use an isolation amplifier or differential probe with low capacitance and 1000+ V withstand voltage. Higher if you work on 690 V inverters. Using two channels and subtract function usually causes problems. The two channels and probes have to be very exactly matched to get any meaningful results.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Madcow]

The usual path of destruction when I fry a scope probe
is that most scope probe returns are tied to case grnd
in the scope. Your 415 vac is probably a Y connection
so you have about 200 vac line to neutral. The neutral
gets tied to grnd and when you connect the grounded scope
probe to the line, you complete the circuit. I think
the isolation transformer was ment to power the unit under
test.
Like lionelHutz said I believe the Fluke scope is an
isolated input check the manual for the safety rating.
ie. 1000 V CAT II and 600 V CAT III safety certified .
I use the Tek P5205 isolated probes now.
NEVER CUT THE GROUND!