how to safely take measurements with oscilloscope??

[kasumyku]

Hi everyone,

Already looked thru posts and watched a lot of vids trying to get as much info as possible regarding oscilloscopes but there is not much about SAFETY!!!
English 2nd language so I hope you'll understand my question,also im 1/2 way into electronics technician course so I'm still a newbie.
Just bought my first oscilloscope and I'm still afraid of short circuits!!!
I understand the way grounding works(thanks to Dave's video and Google) and if attaching ground clip to positive you'll get a nasty shock or even wors, and that's because neutral and ground are connected together so you form a very nice loop
SO my question is : HOW TO SAFELY PROBE YOUR CIRCUIT?
Maybe it sounds stupid (once again still a newbie) but beforeprobing should multimeter be used to check for polarity, or what advices can I get from more experience guys like you.

Thanks!!!

[c4757p]

To put it simply, look for return paths. For current to flow in a short circuit, it must have somewhere to flow back. If there are no return paths (the device under test is powered by a battery, or an isolated lab supply), you can stick your probe anywhere, because there's nowhere for current to flow.

Keep in mind that once you stick that probe somewhere, it becomes a return path for the second probe, so you will then need to exercise care when connecting channel 2.

If there is a return path, you must not connect the ground lead to a voltage that is different from the oscilloscope's ground, because a voltage difference across the wire will cause current to flow.

Don't probe mains.

[WBB]

Checking the test points with a multimeter before attaching a scope is an excellent practice. Even more so if the DUT is plugged into mains. Checking the voltage between the bnc jack on your scope and the point on your circuit you are about to hook the scope ground to can prevent a nasty surprise.

[kasumyku]

Thanks guys for such a quick reply! So after all is good to test polarity with DMM before probing, at least until I become more knowledged about scopes and electronics in general...

[c4757p]

Well, remember what I said about return paths - if there is no return path (which will happen most of the time!), your DMM may not read zero and yet it will still be safe to connect. You need a DMM with a "low impedance" mode to really check this. Otherwise, manually put it into a high range (at least one above the 2V range), and hold the probes there for a minute and see if the number slowly drops. (The range is because some  meters have an extra high impedance on the low ranges.) If the number jumps around as you move the probes, you're probably OK. This is because there can be capacitance between the device and ground, which is charged up, and the meter will read this and make you think there's a voltage between them. Technically there is, but it will go away as soon as you make your connection and discharge it.

The best way is to know the circuit and think about whether there is a path for current to be conducted. Obviously the ability to do this comes with experience, but you should start trying to develop that.

[IanB]

Simply put:

1. The circuit you are testing must be isolated. It should be powered from an isolated lab power supply or batteries.
2. If using two or more probes, never connect the ground clips to different parts of the circuit.
3. As others have said, check voltages with a multimeter before probing them.

[edavid]

1. The circuit you are testing must be isolated. It should be powered from an isolated lab power supply or batteries.

This is misleading.  It is normal for the circuit being tested and the scope to have a common ground.  You can then probe any part of the circuit safely, except the primary side of a line operated power supply.  What you can't do is connect the scope probe ground clip to anywhere except another ground point.  If you need to be able to do that, then yes, you need an isolated power supply.

[IanB]

This is misleading.  It is normal for the circuit being tested and the scope to have a common ground.  You can then probe any part of the circuit safely, except the primary side of a line operated power supply.  What you can't do is connect the scope probe ground clip to anywhere except another ground point.  If you need to be able to do that, then yes, you need an isolated power supply.

I accept your point, except that for a beginner this might be an unwise working practice. It is very easy to absent-mindedly clip the ground clip to a part of the circuit you want to measure, forgetting that it is not an isolated DMM probe. If you clip it to the wrong place you could let out some magic smoke or worse. I have done this myself and was only reminded I was being dumb when sparks flew from the ground clip.

[eKretz]

I recommend that beginners check with a multimeter first if there is any question in their mind. I even still do this from time to time to be safe if I'm unfamiliar with a circuit/device.  What the others are trying to tell you is that you can probe your device safely, you just need to watch your grounds. Sometimes ground isn't ground, you see? LOL. There's mains earth and there's device ground, and they are not the same. Many circuits have both, and they can not be safely interchanged when probing. You really ought to consider using an isolation transformer for your devices under test also.

[edavid]

This is misleading.  It is normal for the circuit being tested and the scope to have a common ground.  You can then probe any part of the circuit safely, except the primary side of a line operated power supply.  What you can't do is connect the scope probe ground clip to anywhere except another ground point.  If you need to be able to do that, then yes, you need an isolated power supply.

I accept your point, except that for a beginner this might be an unwise working practice. It is very easy to absent-mindedly clip the ground clip to a part of the circuit you want to measure, forgetting that it is not an isolated DMM probe. If you clip it to the wrong place you could let out some magic smoke or worse. I have done this myself and was only reminded I was being dumb when sparks flew from the ground clip.

On the other hand, if the circuit is isolated, the beginner might forget to connect the scope ground at all, and then be confused by seeing AC line leakage everywhere 

[c4757p]

Sure, but "remember to connect the ground lead" is right up there with "remember to set the timebase". It doesn't really belong in a discussion of safety, it's equally as basic as "turn on your oscilloscope".

[kasumyku]

This is misleading.  It is normal for the circuit being tested and the scope to have a common ground.  You can then probe any part of the circuit safely, except the primary side of a line operated power supply.  What you can't do is connect the scope probe ground clip to anywhere except another ground point.  If you need to be able to do that, then yes, you need an isolated power supply.

The beginner got confused now 

[c4757p]

Easy to do, don't feel bad. What exactly is confusing you?

[kasumyku]

Let me know if I understood wrong what you explained...

If both,device under testing and scope share same ground then HOW can you probe any part of the circuit safely?
My example was: what if ground clip touches by mistake a voltage difference , then ....booom

Once again, sorry but maybe I understood wrong your explanation...

[c4757p]

The two must share a common ground or else you won't get a proper signal - but if they don't already share one, the ground clip will establish one. If they already share a common ground through some other means, then attaching the ground clip to something in the circuit that's not ground will short that point to ground. You can still connect the probe, you just have to connect it to a ground point.

Even if they don't share a common ground, there can still be a voltage between them. This voltage is stored in the capacitance between the device and the oscilloscope - you can see it with the voltmeter, which is why the DMM isn't always the best way to check, but once you connect the ground lead, the voltage on that capacitance will go away.

Sorry if it's still confusing. There is a lot of information to take in at once here.

[vk6zgo]

Let me know if I understood wrong what you explained...

If both,device under testing and scope share same ground then HOW can you probe any part of the circuit safely?
My example was: what if ground clip touches by mistake a voltage difference , then ....booom

Once again, sorry but maybe I understood wrong your explanation...

If it is just that you touched a 5v supply rail or similar in the equipment,you now have given the unit a new fault

If the DUT does have a designated "ground",it is probably connected to any metal frame of the device,or on a PCB,a large area of copper.

If the internal "ground" is connected to the Mains supply Earth,you can actually check various points for the presence of voltage with respect to ground using the probe with the ground lead/clip disconnected,(removed),as the Oscilloscope power cable provides the necessary return path to Earth.

If a voltage is present,you know you can't put the ground clip on that point!

If you need to critically check waveforms at various points,obviously having several tens of metres of wire between the two Earth points will not give you the best results,so once you have determined the "ground"of your DUT,clip the "ground" clip to that & take all your measurements with respect to that point.

 NOTE:I have used the terms "Ground" AND"Earth",as in the DUT,"ground" may simply be a "common" rail with no connection to the Mains Supply "Earth" (or any other part of the Mains).
When the "ground" & "Earth" are connected they become the same thing.
Sometimes it would be nice to see the voltage across a component where both ends are above ground,but this can be calculated if you know the voltage at both ends,or you can measure both ends at the same time w.r.t ground using two channels & use the A-B function on your 'scope.

If,for any reason,you need to probe the Mains supply,use the "no ground clip" method described above.

WARNING: Not all probes are rated to look at Mains supply voltages,so you will need to be sure of this.
It is probably better at this  stage if you leave this till you are more experienced.

[kasumyku]

MAybe it sounds stupid but what about this situation:

Ok, so what if you do the same mistake but the other way around...
Like, both DUT and SCOPE have a common ground and by mistake you put your SCOPE probe ( not ground clip) to  the DUT ground, would it happen almost the same thing, like wouldn't short the SCOPE? because it also creates a path the other way around, it's like attaching the ground clip to the probe...

Does anyone know a good book about oscilloscopes and how to take proper measurements so not to bother you guys anymore.

[c4757p]

You're not going to get a short through the probe tip because the oscilloscope input has a very high impedance (1 megohm). There is a current path, technically, but it's only going to conduct a few microamps. Obviously you do have to respect the maximum voltage of the scope input (typically around 250-300V).

[kasumyku]

Thanks for clearing things up, obviously I need to do a lot of research before I do anything stupid!
As requested before, DO YOU GUYS KNOW A GOOD BOOK ABOUT OSCILLOSCOPES AND HOW TO TAKE PROPER MEASUREMENTS?

Thanks!

[c4757p]

DO YOU GUYS KNOW A GOOD BOOK ABOUT OSCILLOSCOPES AND HOW TO TAKE PROPER MEASUREMENTS?

NO I DO NOT!

But I do know how to solve that problem - I think your shift key is stuck. Check it for gunk underneath; clean with alcohol. 

[dr.diesel]

Check it for gunk underneath; clean with alcohol. 

Will this work: 

[c4757p]

The best part is that I almost called it IPA for isopropyl alcohol, in which case you still could have used that picture 

Sorry, kasumyku, for being a prick on your thread, I'm in that kind of mood... Most of us didn't learn this from a book. Just play around. Seriously, as long as you respect the few basic rules we've established here, and you don't go waltzing about near mains voltage, you're not going to blow up your scope.

[kasumyku]

  you are not being a prick, but that was a good one,btw just fixed it, look actually its even better now!
thanks guys for all your advices!!!

[vk6zgo]

MAybe it sounds stupid but what about this situation:

Ok, so what if you do the same mistake but the other way around...
Like, both DUT and SCOPE have a common ground and by mistake you put your SCOPE probe ( not ground clip) to  the DUT ground, would it happen almost the same thing, like wouldn't short the SCOPE? because it also creates a path the other way around, it's like attaching the ground clip to the probe...

Does anyone know a good book about oscilloscopes and how to take proper measurements so not to bother you guys anymore.

If both the DUT & the 'scope have the same ground,there is no potential difference between the grounds of the two devices.

If you place the probe on the DUT ground,it is the same as touching the ground of the 'scope,so you will be looking at zero volts,& as the probe is not a source of current,there is no "path the other way round".

[dr.diesel]

btw just fixed it

BTW, which IPA did you use?

[c4757p]

By the way, even if you did put your ground clip on a dead short, you're unlikely to blow your scope. The ground connection from the front-panel connectors goes straight through the massive bulk of the chassis (or the slightly thinner shielding metal of a newer digital scope) to the earth pin on the mains connector. You'd need a hell of a power supply to blow that up. You're more likely to damage your probe, or even more likely the circuit you're testing, which is less of a loss than an oscilloscope.

[kasumyku]

@ c4757p    that was pretty much my main concern, just bought my first brand new scope so I didn't want from my first few testings to blow it up, that will be a drawback...but as every experienced electronics guy says " I hope you will fail the first time..." so we can all learn from mistakes 

[ddavidebor]

Checking the test points with a multimeter before attaching a scope is an excellent practice. Even more so if the DUT is plugged into mains. Checking the voltage between the bnc jack on your scope and the point on your circuit you are about to hook the scope ground to can prevent a nasty surprise.

Well, a multimeter is not necessary.

Just try to measure with your scope without the ground clip.

Since the scope is ground referenced, you'll find voltages between ground and a ground referenced path.

[IanB]

Checking the test points with a multimeter before attaching a scope is an excellent practice. Even more so if the DUT is plugged into mains. Checking the voltage between the bnc jack on your scope and the point on your circuit you are about to hook the scope ground to can prevent a nasty surprise.

Well, a multimeter is not necessary.

Just try to measure with your scope without the ground clip.

Since the scope is ground referenced, you'll find voltages between ground and a ground referenced path.

That's not the point. You do not want to find any high voltages with your scope probe since the scope input cannot necessarily handle them. You want to find the high voltages with your multimeter so you can avoid probing them with your scope and protect your expensive scope from harm.

[ddavidebor]

well, you MUST know that order of magnitude is the voltage in your board.

it's 3v battery? 30v power supply from transformer? 300v bridge rectifier? 3000v hight tensione crt? or a 15000 ones?

next, just probe with 10x and live happy

[vk6zgo]

@ c4757p    that was pretty much my main concern, just bought my first brand new scope so I didn't want from my first few testings to blow it up, that will be a drawback...but as every experienced electronics guy says " I hope you will fail the first time..." so we can all learn from mistakes 

I've never subscribed to that idea,you don't learn Electronics theory by just diving in like a "Bull in a China shop".
What you will learn from things that don't work is to pay more attention to theory & think things through before leaping in.

Learn from other people's mistakes & listen to what experienced people say.
Some err on the side af too much caution,others are too "laid back",& most are somewhere in the middle.

[G7PSK]

I have just come across this book today, ordered a copy to have a look as it appears interesting the seller still has another copy. Might be cheaper elsewhere but I was not about to bother looking.

http://www.ebay.co.uk/itm/150955438475?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

[alm]

Reading the Tektronix appnotes "XYZs of Oscilloscopes" and "ABCs of probes" would also be informative and is free. Note that there are several versions, the old ones focussed on analog scopes and the more recent versions (eg. the one you can find on their website) are about modern digital scopes.

[fpliuzzi]

In addition to the two excellent appnotes alm mentioned, here's another Tektronix appnote that may be of interest. The attached PDF file is "Fundamentals of Floating Measurements and Isolated Input Oscilloscopes".

Regards,
Frank