When does oscilloscope ground not work the same as DMM?

[hurricanehenry]

I tried searching Yahoo Answers and this question is what I wanted answered, but the answers people gave do not actually address it:

When is the probe ground clip required to be connected when taking oscilloscope measurements?

I've often taken measurements without it connected. When first starting out with scopes I tried to connect it to an arbitrary point in the circuit to measure the voltage potential between two points (as one would do with a voltmeter) and found this did not work. Why's it there? When is it needed? When isn't it needed?

https://answers.yahoo.com/question/index;_ylt=AwrXgiOgSgFZwmwAPoF35O1_;_ylu=X3oDMTByb2lvbXVuBGNvbG8DZ3ExBHBvcwMxBHZ0aWQDBHNlYwNzcg--?qid=20070904190035AA5dSqn

I too want to know when do you need to connect the ground clip? We don't need to for calibrating the probe with the scope itself, and I assume it's because internally it already has a ground reference, so both the probe and the calibration post are at the same ground level.

But the strange bit from the Yahoo OP is that he tried to attach the ground clip like a DMM and it did not work. We don't know what he did or what he is probing, but is there a case where it would NOT work like a DMM? In other words, would voltage measured be different? It doesn't seem to make sense it would be different (say, you measured the potentials in between a series of voltage divider resistors).

[Brumby]

Your average DMM completely 'floating'.  Your average scope is not.  This makes a HUGE difference.

First thing I suggest is that you watch this:

[tggzzz]

Your average DMM completely 'floating'.  Your average scope is not.  This makes a HUGE difference.

First thing I suggest is that you watch this...

Just so.

Since this is the beginners' forum, the first thing NOT to do is "float the scope" by disconnecting the scope ground.  Or, slightly more accurately, if you want to float a scope you have to do something like:



In other words, don't do it.

Have a look at the references in https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/ for further information

[tautech]

To put it very simply think of the ground lead (as everyone calls it) as the reference from which the signal you wish to measure/display is referenced against. More correctly it need be called the reference lead but in most scopes it is also tied to mains Gnd hence why everyone calls it the ground lead. This brings the problem as clearly outlined in Dave's vid that it CANNOT be connected to just anywhere on a DUT but for signal fidelity it need be connected somewhere near the source of the signal BUT still at mains ground potential.

Understood ?

It matters little that the reference lead is used or not at Probe Cal amplitude and frequencies but it sure matters if you are to believe what you see at higher frequencies and/or much lower amplitudes.
Easy really. 

[Zero999]

Another option is a battery operated 'scope but since the common is adequately insulated from all the controls, it should not be used of hazardous voltages. Only use it with voltages under 60VDC or 25VAC.

[TimFox]

When dealing with junior members at work, before my retirement, I tried to teach them to treat "ground" as the "g-word", and not to use it freely.  For a given configuration, "protective earth", "chassis ground", "ground plane", "transmission line outer conductor", "circuit common", "power supply return", etc. are not interchangeable and it helps to use a specific term for the node in question.

[tautech]

When dealing with junior members at work, before my retirement, I tried to teach them to treat "ground" as the "g-word", and not to use it freely.  For a given configuration, "protective earth", "chassis ground", "ground plane", "transmission line outer conductor", "circuit common", "power supply return", etc. are not interchangeable and it helps to use a specific term for the node in question.

Quite so but any one (or all) of those could also be at mains ground potential for the purposes of connecting a scope reference lead to.
Good advice nonetheless.

[rstofer]

But the strange bit from the Yahoo OP is that he tried to attach the ground clip like a DMM and it did not work. We don't know what he did or what he is probing, but is there a case where it would NOT work like a DMM? In other words, would voltage measured be different? It doesn't seem to make sense it would be different (say, you measured the potentials in between a series of voltage divider resistors).

Sure, it's easy to come up with a scenario where the scope would cause a problem (or damage).

Consider a simple transistor common-emitter amplifier where the emitter is connected to logic ground and logic ground is tied to earth ground at the power supply.  The emitter is grounded - period!

Now, suppose I wanted to measure the current through the transistor by using a DMM and reading the voltage across the collector resistor.  That would work quite well.  But...  Suppose you wanted to do the same thing with a scope.  You can't connect the ground lead to the high side of the resistor (Vcc), that much is obvious, but, in this special case, you could connect it to the collector of the transistor.  Sure, you would short out the transistor but there would be no damage, the emitter is already at earth ground.  You could put your probe on the Vcc end of the resistor and get a display of a straight line at Vcc.  Any signal has been shorted out.  You're not looking at the signal, you're looking at some variation of Vcc.

There are many other examples where attaching the ground lead is disastrous.  What you need to think about it whether the point where you are going to attach the lead is already at earth ground potential or whether there is no earth ground connected to any part of the circuit (battery powered).

One thing is certain:  When you attach that probe ground lead, you WILL be connecting that point to earth ground through the scope.

Just THINK before you attach that ground lead!

[hurricanehenry]

Thanks for all the great replies.

So, I have an isolation transformer, which I usually use for the DUT, since it is more likely a DUT would have an energized chassis fault, than my scope.

Should I instead use it for my scope?

[tautech]

Thanks for all the great replies.

So, I have an isolation transformer, which I usually use for the DUT, since it is more likely a DUT would have an energized chassis fault, than my scope.

Should I instead use it for my scope?

NO, DUT only.

Think about where you might clip the scope reference lead, if to a DUT that is powered via an isolation transformer the reference leads connection point has zero hazardous voltage in relation to mains GND.
Reverse the situation and all metal parts of the scope have potentially fatal voltages applied.

No 1 consideration: Stay safe.

[Electro Detective]

If you want to stay healthy and in one piece, and the same goes for your equipment, you better watch a LOT of Youtubes and visit  electronics/electrical forums and websites to understand how you may be blinded, killed or fried alive because you did not understand how earthing/grounding/shielding/metal chassis/bare concrete floor and electrics in general work, from the power station to your test gear terminals at home or work.

It's simple to do right, and yet even simpler to get it wrong, and DIE assuming you got it right. 

The real problem is that it's never explained properly, because a LOT of people including mis-educated or lazy apathetic 'industry professionals' GET IT WRONG (or don't gas) and pass half baked BS to others who blindly believe their words.

For starters, there are MANY variations of 'isolation transformers' and if you don't know them, it's RISKY BUSINESS to trust them.
Even if you have a true isolation transformer, it may have issues or metal shavings or loose screws rattling about inside making contact between ground/earth and neutral (Line 2)   

or some coffeeholic d***head office twat prankster (with small private parts) playing machine gun with the aloof secretary's stapler has fired staples that have found their way into the transformer's vents and on to the terminals inside, eventually causing a bridge to ground/earth and neutral (Line 2) on any given day =  goodbye  'isolation'   

Most battery powered multimeters are isolated/floating devices, but you can blast your meter and yourself into OBLIVION under the correct WRONG conditions (do yourself and meter a favour, to find out how and why NOW, not after it happens!)   


There are better things to do in life than be the centre of attention... at your own FUNERAL   

[Luminax]

This reminds me of something about 1 or 2 months back...

So our company have an automatic Remote Control PCB electrical checker which checks, among other thing, the burst current flowing through an IR LED.
The Tester was designed by HQ and utilizes a floating DMM (34901A, if anyone recognizes these sequence of numbers and letter), and some weird calculation formula to derive the peak-to-peak voltage, divide it by the limitting resistor, and gets the peak current from an RMS reading of the DMM.

TL;DR, a problem occur when I was off work, and my boss, in his infinite 'wisdom' did the thing that I shall depict in the attachment below...

Take a guess at what happened?





(if your guess involves words like 'smoke' and 'magic', you just won something!)

[Electro Detective]

Oscilloscope probe ground clips have their place on the bench,
usually in labelled plastic bags in a drawer, where all the alligators and crocs party on till needed.

DIFFERENTIAL OR DEAD is where it's at for me, 

if the measurement or troubleshoot does not need a solid earth/ground/terra reference, it's ADD-INVERT Ch2  time! 

Millions may disagree 

[ebastler]

Oscilloscope probe ground clips have their place on the bench,
usually in labelled plastic bags in a drawer, where all the alligators and crocs party on till needed.

DIFFERENTIAL OR DEAD is where it's at for me, 

if the measurement or troubleshoot does not need a solid earth/ground/terra reference, it's ADD-INVERT Ch2  time! 

Now I understand why, in that other thread, you were so vocal about needing four channels plus separate trigger input all the time! 

If you are really serious about the above advice, and don't just mean to provoke, you are limiting yourself and the use of your scope quite severely. So, have you ever tried a "real" three or four channel measurement? 

[tautech]

Oscilloscope probe ground clips have their place on the bench,
usually in labelled plastic bags in a drawer, where all the alligators and crocs party on till needed.

DIFFERENTIAL OR DEAD is where it's at for me, 

if the measurement or troubleshoot does not need a solid earth/ground/terra reference, it's ADD-INVERT Ch2  time! 

Now I understand why, in that other thread, you were so vocal about needing four channels plus separate trigger input all the time! 

If you are really serious about the above advice, and don't just mean to provoke, you are limiting yourself and the use of your scope quite severely. So, have you ever tried a "real" three or four channel measurement? 

Each of us have differing use cases and some specs we see as mission critical with equipment selection. There are zero hard and fast rules, it's purely personal preference.
Much comes down to knowing your instruments capabilities, strengths and weaknesses and how to work with them for best results.
The perfect instrument to suit our needs is still a dream.
Few get this right first time as the knowledge required is way beyond beginner level just like the topic of this thread and how it waits to trip up the unwary.
That some more experienced scope users on the forum keep guiding those that do not know better is great. 

[Electro Detective]

Oscilloscope probe ground clips have their place on the bench,
usually in labelled plastic bags in a drawer, where all the alligators and crocs party on till needed.

DIFFERENTIAL OR DEAD is where it's at for me, 

if the measurement or troubleshoot does not need a solid earth/ground/terra reference, it's ADD-INVERT Ch2  time! 

Now I understand why, in that other thread, you were so vocal about needing four channels plus separate trigger input all the time! 

If you are really serious about the above advice, and don't just mean to provoke, you are limiting yourself and the use of your scope quite severely. So, have you ever tried a "real" three or four channel measurement? 

Ah yes, the ~DSO Defenders~, greetings y'all    

My comments above are related to SAFETY i.e. "Re: When does oscilloscope ground not work the same as DMM?"

aimed at misfortunate people who try/attempt to use their oscilloscope probe and ground/earth croc lead like a floating 'Red/Black lead' multimeter in a typical 'new player' and 'old player' momentary brainfart BANG! situation.

I prefer to go in to an -unknown fault scenario-    be it suspect voltages or plain curiosity waveform sniffing, with poor man's differential mode first, before setting up croc grounding and probe prodding.
Recommended if one is clueless or apathetic about DUT isolation, and types of isolation transformers (some which may or may not truly isolate)

You can save money too with an initial differential sniff, big money
(and no baby crocodile or alligator clip needs to die before their time, or have their tail fried) 

My diff work gets done with an analogue scope first, before I consider hooking up any DSO

And yes, I was safely performing 8 channel single ended? measurements with Tek cros fitted with dual 4 ch plugins back in the day and beyond,
before 'entry level' half@$$ed firmwareD 4 ch DSO snails came on the scene to spoil my expectations (and pocket)...
UPDATE: hoping that will change asap and I can retract some negatives. 

Again, millions may disagree,
and free to exercise their choice to happy go lucky clip n prod away to their heart's delight            

[tautech]

My diff work gets done with an analogue scope first, before I consider hooking up any DSO

Why ?

In my experience many CRO's don't have the input ratings of more modern equipment that is made for the global marketplace where once upon a time the big brands focussed on their counties and ratings that were suitable for them.

[Muxr]

DMM is your friend. Before you bring out the big guns (the oscilloscope) measure everything there is to measure with the DMM. Try to learn all the rails, analyze the schematic.. you can even check via continuity if the DUT is grounded and which are the ground points. Once you're sure about it, then you can proceed to clip your oscilloscope's ground to the DUT's ground. If your probe is grounded properly, you should be relatively safe other than around high voltage circuits, but you should know this before you bring out the oscilloscope from the above point.

Approach oscilloscope measurements analytically, don't just measure with the scope willy nilly.

[Electro Detective]

My diff work gets done with an analogue scope first, before I consider hooking up any DSO

Why ?

In my experience many CRO's don't have the input ratings of more modern equipment that is made for the global marketplace where once upon a time the big brands focussed on their counties and ratings that were suitable for them.

Economics: I won't cry too much at the loss of a -cheap to replace- analogue 'spotter dog' oscilloscope if it gets taken out,
and with 10x probes the old vs new scope input rating differences are meh 

[xrunner]

Oh cool - a good old fashioned analog vs. digital shootout.

I love these. 

[Electro Detective]

Oh cool - a good old fashioned analog vs. digital shootout.

I love these. 

Hang on to something solid and enjoy the ride   

[Luminax]

With regards to differential... I(or my company actually) don't really have the budget(read: Cheap-ass boss) to buy differential probes sadly.
That being said, the above pic I posted is just a matter of simply using 2 probes, measure V1 and V2, and use subtract math function for the peak-to-peak value of resistor voltage drop (in other word, current).

...granted I am lucky enough that I have the DSOX2014A with the legitimate (*ahem*) MATH function