FYI on scope / probe usage

[BEP]

I continue to see troublesome usage of scope probes and the incorrect or misleading results from this.

Rarely do I see such problems here but even when the technician has decades of experience some problems are rooted in equipment miss use.

This post is spurred by my emergency repair of company equipment (NI PXI system) because a newbie college grad thought a scope worked like a DMM, destroyed the DUT and the customer's equipment.

I pasted this on his computer monitor:
(JIC: These are not my opinions. The same information can be found in most scope user manuals and tech-docs.)

[Farmhand]

Thanks for this tip WaveWatcher, Is there any chance we could get a list of "Definitely do not do" things with dual probe scopes ?

Ie. If I want to measure the AC output of a transformer and the DC input from battery, what is the right way to do that, and what to avoid ?
Also when I measure the AC output of a DC switched transformer with a scope powered from the grid what to avoid ?

There must be other things.

Can we feel free to ask a specific measurement question as to probe usage as we need to, if we provide a drawing of the proposed device and test circuit ?

Is it a general rule that the "ground clips" on the probes must be all connected to the same potential ? And also with reference to the scope supply ground ?

When in doubt I just use only one probe, but I think problems can still be encountered even with just one probe.

Thanks for the tips already, much appreciated. This thread is well named for the issue, it would be good to see more reference info here.

I don't mind to delete this post to clear the thread for the important info.

Cheers

[BEP]

.99 and others have provided a great deal of very good scope usage information.

I'm happy to answer questions but my time for the forums is becoming even more limited.

As far as the probe ground clip is concerned - just remember that that clip is directly connected to the scope chassis and further to the ground prong on the power plug. Connecting it to anything except another grounded point is always bad practice.

Anything you connect that clip to is then shorted to ground.

One exception.... when the scope probe and/or the scope is designed for connecting the clip to any point. Then, the clip is electrically isolated from Earth ground, unlike almost all scopes/probes.

I've heard many say the correct action is to remove the ground prong from the power plug. That is fine if you wish to kill someone or yourself.

Even when voltages aren't lethal, such an action adds the scope chassis to the circuit under test.

The same fools that chop the ground prong off are the ones that complain a scope is useless because the display is always distorted or doesn't show what is really there  

Idiots  

[Farmhand]

Thank you very much, that all makes complete sense. I think I did the shorting thing with my old scope. Now I undertand

Cheers

[exnihiloest]

...
I've heard many say the correct action is to remove the ground prong from the power plug. That is fine if you wish to kill someone or yourself.

Even when voltages aren't lethal, such an action adds the scope chassis to the circuit under test.

The same fools that chop the ground prong off are the ones that complain a scope is useless because the display is always distorted or doesn't show what is really there  

Idiots  

I agree

There is a way to bypass the problem: using two probes in differential mode. The scope sees only A minus B and so, the signal from the ground is canceled. The limit is obtained when A minus B is a weak signal while the signals at A and B are strong relative to the ground. In this case the scope inputs can be saturated and the A minus B signal completely distorted.

[EMdevices]

Thank you very much, that all makes complete sense. I think I did the shorting thing with my old scope. Now I undertand

Cheers

Don't feel bad, I once fried a SMPS copper trace pretending to "fix" it, and I already had the training class.  I just touched the probe ground clip to the rectified ground reference past the diode bridge where the (-) terminal of the 200 + V electrolytic cap was hooked, and sparks flew everywhere.  For a second I thought I shorted two traces with the clip, so I touched the ground clip to the center island and again big sparks everywhere and a burnt hole in the copper island.  I was very puzzled to say the least but then I recalled what they said in class,  you need an isolation transformer on the SMPS, because the scope ground shorts out the diode bridge.  Oops!  lol

That early lesson was valuable as later I corrected others working and troubleshooting million dollar equipment and not just hooking up one channel but four, into a harness breakout box.  

Here's a pitfall:  some think that their scope channels are "isolated" therefore they hook up one probe without thinking about the other, with the real danger being that the ground clips are connected to different locations that can have a potential difference, nevermind the fact that the clips are also tied to electrical ground, which can be "remedied" by using a 3 prong to 2 prong ground isolator, but be ware, now the instrument chasy is floating and if you touch a probes ground clip to 220 v that will be the potential, but don't do that, just get yourself a differential probe, or use the scopes A-B or B-A differential settings.

Anyway if you have a portable scope you can ignore pretty much everything I said.  LoL

Happy troubleshooting!

EM

[BEP]

which can be "remedied" by using a 3 prong to 2 prong ground isolator

In my business that action is a safety violation. It is more dangerous than working on a live circuit with no lock-out-tagout. Either can get you fired on the spot.

I had Tektronix provide a 'basic scope use class' just three weeks ago to supposed 'seasoned' bench technicians.

You may ask, why so severe?

277VAC does not feel very good when you touch the scope chassis when someone well trained is scoping a 3-phase 480VAC signal. When the boss gets the sh#t knocked out of him because his techs don't know how to use a scope - expect hell to break loose. (The 'boss' knows more about test equipment usage than all of his technicians combined.)

Isolation transformers are also unacceptable for such measurements. Connecting the scope chassis to anything except a point already grounded, at a minimum, affects the signal displayed. ISO xFmrs should only be used to prevent external noise or transients from interfering with measurement or prevent DUT generated transients from leaking to the mains or others systems.

When scoping a floating potential, the only acceptable options are the differential method or the use of a differential probe. Even the portable scopes affect the signal unless each channel is designed as a differential probe. We have those, too.

Anyone only having the test equipment training provided in physics or engineering classes should attend some current real-world equipment safety and usage training.

Am I being an ass? Yes, you can become that way after mopping up the leftovers and attending a couple of funerals.
---------

On the lighter side: I am serious when I say the scope chassis will become part of the circuit under test when you connect the ground clip to a floating point. I've seen too many false positives/negatives in diagnosis and testing caused by this.

Folks should just read the manuals supplied with the equipment.

[EMdevices]

I placed "remedy" in quotes because it is a fallacy and very dangerous with high voltage.  We're on the same page.  BTW, the isolation transformer is for the SMPS not for the scope.

EM