Question #1: Just 3 days ago one of our technicians came to me seeking advice for a problem - when he hooked his scope to a DUT it would trip a circuit breaker. He wanted to know if I had an isolating adapter that would disconnect the scope earth ground connection.
I did have one, stolen and carefully hidden away so that NO ONE could ever use it.
I informed the technician in the presence of management that he needed to requisition an isolation transformer for the DUT, and that one must NEVER float the scope unless contemplating suicide. Waiting to see what happens now...
zeeglen can you explain why it is so unsafe so that everyone can learn from it? Especially this case: You have the DUT floating and the scope not. Now you connect the ground of the scope to the DUT, as a consequence the DUT is NOT floating anymore. Why is this still safe?
One side of the mains electricity is always connected to earth, usually via the neutral leg at the switchboard.
If you have a Device Under Test (DUT) which is directly connected to the mains for its power, some or all of its circuitry will be at mains potential. Consider a switch-mode power supply (which most wall wart and laptop power supplies are these days). They have a lot of electronics on the mains side, So if you are trying to troubleshoot such a circuit, when you connect your Oscilloscope, which may have its earth at mains earth potential, you may create a direct short from a live mains point to earth through the scope. This may cause a circuit breaker to trip (as in Zeeglen's case above) or a large spark. And if your finger happened to touch that mains point, you would get a nasty shock or even kill yourself.
If you run the DUT through an isolation transformer it removes the connection to earth and if you touch either of the power connections you would not feel any shock (or at the most a mild tingle) and it certainly would not kill you. You still have to take care, because there are still high voltages present.
As an aside, if you are working on this sort of equipment it is also wise to use an RCD (Residual Current Device) sometimes called an ELCB (Earth Leakage Circuit Breaker). This is a refinement of the standard "trip switch" circuit breaker, which cuts the power off if more than 15-20 mA flows through the earth - such as would happen if you touched a live mains point with any part of your body. They are sometimes a pain in the butt as some equipment seems to set them off, but in a workshop environment they are an excellent safety precaution.
@David I totally agree that you should have your DUT floating. Either using an isolation transformer or an AC source with good isolated floating outputs. Also the use of a RCD / ELCB, or called GFCI, is a very good practice. And your explanation is clear, thanks.
Further it would be good to use an isolation transformer with low primary to secondary capacitance as this would lower any shocks when you actually touch the DUT. There are huge differences in parasitic capacitance between different isolation transformer models.
Most (if not all) scopes have the ground of the inputs connected to the ground of the mains. If your DUT is now floating and you connect the ground of the scope to the DUT, the DUT will not be floating anymore and your test setup is unsafe again. Therefor you should put your scope behind a (different) isolation transformer as well.
My ideal setup would be:
Use an AC power source with isolated floating outputs.
Put your measurement equipment behind an isolation transformer.
Use a (High Voltage) differential probe when ever posible to measure on the mains side of the DUT. Especially when you want to do measurements on the secondary side as well, to prevent tying the ground of both sides together.
I would love to hear from a safety specialist if I am overlooking anything.
Hi again Paul. Sorry, you obviously know a bit more than I assumed when I replied.
> If your DUT is now floating and you connect the ground of the scope to the DUT, the DUT will not be floating anymore and your test setup is unsafe again. Therefor you should put your scope behind a (different) isolation transformer as well.
That is partly true. 230V is 230V and if you put it across your body it will give you a shock, true.
But a live DUT relative to earth is VERY dangerous. If when you are connecting your scope ground clip you happen to contact the DUT live point yourself, you'll get a shock.
But this will not happen if your DUT is behind a (good) isolation transformer. Yes, your DUT is again earthed, but it won't trip your circuit breaker and it will not give you a shock. And holding your plastic scope probe won't either, even if you contact the earth clip wire. Sure, if you go poking your finger onto a place where there is a high voltage relative to ground, you'll get a zap, but this shouldn't happen if you are careful.
One ideal as you say is to have your scope isolated as well, but if you poke your fingers into the right places you can STILL get a zap. And you also run the risk of picking up a lot of hum that you otherwise might not.
I'd also love to hear from anyone else who is more qualified than I to chime in on this?
PS sorry, I'm a product of the British oriented world. Where I put earth, read ground :-)
@phsdv Paul - if your DUT is now floating and you connect the ground of the scope to the DUT, the DUT will not be floating anymore and your test setup is unsafe again. Therefor you should put your scope behind a (different) isolation transformer as well.
Reasonable argument, but I still would not do this. The scope could still be unknowingly grounded through an RS232 or USB cable connected to a grounded PC, meaning that one might think the scope is floating but unaware that it actually is not.
THe exposed metal points of a scope are all AC ground, including any exposed metal of the probes, all the BNC connectors, the calibration terminals. If the scope is floated AND the DUT too, the next user of the scope (say 2nd shift) may not be aware that a day shift tech has ungrounded the scope and attempt to use it on a non-isolated DUT. Then that hazard exists all over again.
Then too floating scopes tend to have more probelms with AC main stray pickup in their displays.
I once heard a story (urban legend?) about a tech leaned over his bench to reach someting while holding a hot wire, possibly a ungrounded scope probe ground lead. His bench was made of metal; his trousers zipper was made of metal. Need I say more?
I'm not really sure about these isolation transformer techniques. I know people use them. Early in my career there was an isolation transformer on the scope cart, but I don't think I used it. My take is that isolation tranformers work but they are not safe. I've used the differential probe method. Either a good differential probe or the pseudo differential probe. A psuedo differential probe uses two channels with one channel inverted and the ADD mode to give you Channel A minus Channel B. The grounds of the probes are tied to each other but not to the circuit board ground. These are all explained in this Tektronix Application Brief.
NASA's Orion Flight Software Production Systems Manager Darrel G. Raines joins Planet Analog Editor Steve Taranovich and Embedded.com Editor Max Maxfield to talk about embedded flight software used in Orion Spacecraft, part of NASA's Mars mission. Live radio show and live chat. Get your questions ready.
Brought to you by