Why Can Not Using a Differential Probe Blow Up Scope

[bostonman]

I'm sure this has been asked many times, but I'm confused about probe grounding and how a scope can get damaged.

The ground lead on a scope is Earth ground. If I measure a battery or a circuit referenced to a battery, the probe ground lead would go to circuit ground. This would tie battery ground to Earth ground.

Now if I measure a circuit all referenced to the secondary of a transformer, maybe a capacitor on a full wave bridge rectifier, normally I just stick probe ground onto any bare metal I can find.

If the secondary is all referenced to the transformer, why does this not cause the measurements to be incorrect?

Also, how does a scope blow up just because someone takes the probe ground, uses it on the secondary, and measures the filter capacitor?

I'm thinking of getting a differential probe, but I'm uncertain when to use it, why I need to use it, and whether using it will be as accurate as using an expensive scope probe.

[james_s]

You should watch Dave's video on exactly this topic.

In a nutshell, probing a battery power circuit or something on the secondary side of a transformer does NOT require a differential probe. Consider though something like a SMPS where the transformer is in the middle of the circuit rather than right on the input. In this case "ground" on the primary side can be floating at mains voltage above earth ground so if you connect the scope ground (which is tied directly to earth) to this point it should be obvious why you're going to get a flash & bang.

[Siwastaja]

There is no magic, it's really simple. The "ground clips" of the channels just are internally connected together, and further to the protective earth of the mains plug.

Whenever you would connect the ground clips (or the mains protective earth) to different circuit nodes so that there exists a path for current, you will create a short circuit through the scope. That's the problem.

[bostonman]

You should watch Dave's video on exactly this topic.

Good suggestion, I just watched the video.

The visual was a big help as I somewhat understood the concept, but seeing was better. Floating grounds, placing DC voltage on a ground to change the reference point, etc.. have always been confusing. I like to think ground is 0V, and everything is referenced to just plain old "ground".

Anyway, two questions. First is Dave mentioned that providing the device doesn't have Earth ground on the IEC connector, then the scope ground lead can be connected anywhere. Just to confirm, I think that is assuming it's connected "anywhere" on the secondary side of the transformer. I'm not going to try this, but thought if I touch HOT on my AC receptacle to Earth ground, then I'll pop the breaker. I believe this is true because Earth ground is isolated at my house (it's just a rod that goes several feet into the ground), but, neutral is tied in with Earth ground back at the power station. Obviously correct me if I'm wrong.

The other question is regarding somewhat my original question. How does the scope physically blow up by touching the Earth ground lead to a high voltage/current source? I understand the physical path current takes (more now after watching the video), but I imagine any current is just pulled to Earth ground bypassing all the electronics the scope until (hopefully) a circuit breaker or fuse pops or the scope probe blows up in your hand because the thin ground wire couldn't handle the high current thus immediately heating until it melts off.

In other words, I touch the scope ground lead to a mains, and the probe can blow up in my hand and/or the ground wire on the probe can melt until it's an open circuit. Simple enough to understand and certainly not a good position to be in because of injury risks. Assuming the ground lead and/or probe doesn't blow up in my hand to open the circuit (and ignoring whatever damage can happen to the unit I'm measuring. i.e. opening a trace, popping the fuse, etc...), I'd assume the only "damage" to the oscilloscope would be a ground wire opening inside the scope which would (hopefully) be a simple fix (open the scope, solder a new wire from the BNC to Earth ground, and remove the old burnt wire).

The only situation I can imagine is that the spark arcs to the electronics and blows up the scope, but, as quickly and easily this can happen, I'm questioning a theoretical situation where high current simply flows to Earth ground until a breaker/fuse opens.

[TimNJ]

I haven't watched the video in a long time, but regarding this...

First is Dave mentioned that providing the device doesn't have Earth ground on the IEC connector, then the scope ground lead can be connected anywhere.

It doesn't matter if a device physically has an earth pin or not. If you are trying to measure a mains circuit (i.e. on the primary-side of a transformer, or a circuit with no transformer at all), you cannot place the ground clip at any point which is not actually the building's earth. That's the only rule.

This is because earth and neutral are galvanically tied together somewhere in the building.

I think what he was saying was: If you are supplying your DUT from a power supply (i.e. wall wart) with only two AC input blades, the output (secondary-side of the transformer) is guaranteed to be galvanically isolated from mains. In this particular case, if your DUT is powered by this isolated supply, you can put your ground clip wherever you want in your DUT's circuit.

[Faringdon]

Just noting there is a serious electrical warning in this post.......in italics (i dont want  anyone to come to grief)

If you remove the earth from your scope's mains plug, then you can connect the scope ground anywhere (well, anywhere on a typical mains circuit with voltages below ~450v from earth potential) and you wont get blow up.
However, that gives you a "floating scope".
I have worked in places where they do just this.
However, if you do it, then remember that your scope ground connects to earth "normally".........and that earth connection , in your scope's power supply, conencts to the y capacitors which also connect , at their other ends, to mains line and mains neutral.......so in other words, if you cut off the earth from the scope.........then do not touch the scope's ground clip while the scope is plugged in....because you will get a capactive shock through the y caps.......do not touch anything metal on the scope.......if you want to change the scope probe position, you have to unplug the scope, and move the probe, then plug the scope back in..

If you are like me, you have a micsig dp10007 diff probe, which is good, but cannot be used to well  probe signals , as the "dangley" probes pickup loads of common mode noise......so if i need to see a signal  nicely, then i sideline the micsig....and i  feed the scope from an  100W mains isolation transformer....and cut off the earth from the scope plug.....and then use a home-brew coaxial probe to probe the signal....and i make heck sure that i dont touch the coax probe whilst the scope is plugged in

[james_s]

"Ground" is an arbitrary concept, it means a point in the circuit from which all voltage measurements are referenced, this can actually be any point in the circuit but usually the node that connects to the negative terminal of the battery or bridge rectifier is used. It may or may not be tied to earth ground, in fact "ground" in a circuit can be the live wire coming out of the wall, it just depends on the circuit. Ground of the scope is earth ground, if you try to connect this to a ground that is at some other potential then current is going to flow.

The scope is not going to literally blow up as in explode, but it can be damaged. The reason should be obvious, if you create a short circuit through the ground clip of your probe a large current is going to flow until the weakest link fails. This could mean a trace on the scope PCB is vaporized, or the probe cable burns up or any number of other things depending on the path the current takes.

Until you understand this concept intuitively like the back of your hand, don't connect the scope to anything that plugs into the wall unless it uses an external power brick or wall wart.

[Ice-Tea]

1) I personally feel floating of a scope should not even be mentioned, especially not to someone who is slowly learning about "earth" "zero volt" and the like. The idea should not even be enterained untill those concepts are clear as day.
2)

However, if you do it, then remember that your scope ground connects to earth "normally".........and that earth connection , in your scope's power supply, conencts to the y capacitors which also connect , at their other ends, to mains line and mains neutral......

This is an underestimation of the risk. One would float a scope to be able to probe mains, mostly, so all exposed metal (probe clip, BNC shell, screw on thousing, whatever) will be live. Connecting to line and neutral through Y caps is the least of your worries as those connections would be, by design, quite safe.

3) In addition, it is my personal belief scopes should not be floated period, not even with an isolation transfo. That gear is designed to work correctly (ie conform to safety and EMC standards) with earth connected.

[rsjsouza]

1) I personally feel floating of a scope should not even be mentioned,

I personally always disagreed with this. Anyone in this field with two neurons can think of isolating the earth to measure something. The subject must be mentioned at all times and the proper disclaimers must be emphasized.

Not to mention there are test equipment that are earth floated by design - especially the older models of oscilloscopes, function generators, etc. Knowing the pitfalls is very wise. 

2)

However, if you do it, then remember that your scope ground connects to earth "normally".........and that earth connection , in your scope's power supply, conencts to the y capacitors which also connect , at their other ends, to mains line and mains neutral......

This is an underestimation of the risk. One would float a scope to be able to probe mains, mostly, so all exposed metal (probe clip, BNC shell, screw on thousing, whatever) will be live. Connecting to line and neutral through Y caps is the least of your worries as those connections would be, by design, quite safe.

Indeed. The oscillloscope or any other test gear will be fully isolated if powered by an isolated power supply, evem with the neglectful leakage of the Y capacitors in case a switching supply is used (power transformer designs have even more neglectful currents). The real problem is you (the operator) becoming the wire that  closes the circuit between the equipment being tested and ground.

[james_s]

I personally always disagreed with this. Anyone in this field with two neurons can think of isolating the earth to measure something. The subject must be mentioned at all times and the proper disclaimers must be emphasized.

Not to mention there are test equipment that are earth floated by design - especially the older models of oscilloscopes, function generators, etc. Knowing the pitfalls is very wise. 

You're entitled to an opinion, but just remember that competent, experienced professional engineers have been killed by floating a scope to take a measurement. We're dealing with a beginner who doesn't fully understand the concept of ground here, it would be negligent to even suggest floating a scope to them. Yes there is equipment that is designed to be floating, that isn't what we're discussing here.

[Faringdon]

Also, how does a scope blow up just because someone takes the probe ground, uses it on the secondary, and measures the filter capacitor?

...Aaaah, this may be something different....
I once unplugged a mains circuit, and there was residual voltage left on  a capacitor, without me realising. I had the scope ground clip connected to the capacitor ground.
I then took my soldering iron (which obviously has an earthed tip)....and  then soldered the positive terminal of the capacitor......so basically i shorted the capacitor out.
Maybe thats what you are referring to?

Anyway, another method for you to probe, without having to cut off the earth from the scope mains plug...would be to supply your DUT via a mains isolation transformer, (MIT) and be sure that this M.I.T   doesnt connect earth to  either of the secondary outputs. Then you can put your probe ground anywhere on that DUT,  and it doesnt have to be a diff probe in that case. Obviously still dont touch the high voltage isolated outputs.

[bdunham7]

You're entitled to an opinion, but just remember that competent, experienced professional engineers have been killed by floating a scope to take a measurement. We're dealing with a beginner who doesn't fully understand the concept of ground here, it would be negligent to even suggest floating a scope to them. Yes there is equipment that is designed to be floating, that isn't what we're discussing here.

I won't disagree with the overall notion that this is not an intro-level subject, but do keep in mind that the one known example (AFAIK) of an experienced, professional engineer getting electrocuted this way involved the apparently common but obviously dangerous method of simply breaking off or disconnecting the ground pin using regular mains power.  Or course even an isolation transformer wouldn't have helped in that case since I believe it was the high-potential DUT that energized the scope.

Another good reason not to go floaty-scope is that it may not work very well--you may end up looking at a whole bunch of mains power artifacts.  In every case I've ever used isolation, isolating the DUT is the only reasonable way to do it.  The only exception I can think of outside of a high voltage laboratory--and I haven't run into it--is if you might need to float the scope with a low-voltage DUT that cannot easily be isolated from ground.  But that would be a more advanced topic with its own hazards, so I'll agree that in the current context, the answer to the idea of floating the scope should be a firm no.

[james_s]

I won't disagree with the overall notion that this is not an intro-level subject, but do keep in mind that the one known example (AFAIK) of an experienced, professional engineer getting electrocuted this way involved the apparently common but obviously dangerous method of simply breaking off or disconnecting the ground pin using regular mains power.  Or course even an isolation transformer wouldn't have helped in that case since I believe it was the high-potential DUT that energized the scope.

Another good reason not to go floaty-scope is that it may not work very well--you may end up looking at a whole bunch of mains power artifacts.  In every case I've ever used isolation, isolating the DUT is the only reasonable way to do it.  The only exception I can think of outside of a high voltage laboratory--and I haven't run into it--is if you might need to float the scope with a low-voltage DUT that cannot easily be isolated from ground.  But that would be a more advanced topic with its own hazards, so I'll agree that in the current context, the answer to the idea of floating the scope should be a firm no.

I would bet that it has happened more than once, and there have almost certainly been many more shocks resulting in various degrees of injury that were not fatal. I have floated a scope before and lived to tell the tale but I've also had a few close calls that were frankly too close for comfort and very easily could have been my last. It's one of those things that can have valid use cases but it should NEVER be attempted by anyone who doesn't fully understand what they are doing, and the OP clearly has not yet reached that level of understanding. It's best to just not even bring it up.

[rsjsouza]

I personally always disagreed with this. Anyone in this field with two neurons can think of isolating the earth to measure something. The subject must be mentioned at all times and the proper disclaimers must be emphasized.

Not to mention there are test equipment that are earth floated by design - especially the older models of oscilloscopes, function generators, etc. Knowing the pitfalls is very wise. 

You're entitled to an opinion, but just remember that competent, experienced professional engineers have been killed by floating a scope to take a measurement. We're dealing with a beginner who doesn't fully understand the concept of ground here, it would be negligent to even suggest floating a scope to them. Yes there is equipment that is designed to be floating, that isn't what we're discussing here.

Perhaps my writing was not entirely clear; I don't condone the practice of floating a test instrument. What I disagree is to prevent the subject from being discussed entirely for the sake of protection. This is not protection but instead an irresponsible way to leave the gap open for careless experimentation by newbies.

Anyone involved with this field is most probably curious, inventive and industrious - it doesn't take much from a newcomer to think that it is a practical idea how to circumvent something so simple without considering all the risks.

[tautech]

Anyone involved with this field is most probably curious, inventive and industrious - it doesn't take much from a newcomer to think that it is a practical idea how to circumvent something so simple without considering all the risks.

Exactly and for that reason alone we must do anything we can to discourage such dangerous practices....even talking in any detail about this sets curious minds off to investigate.   

If we are to be all properly responsible in these days of cheap differential probes we should all be shouting in unison........... DON'T FLOAT YOUR SCOPE !

[magic]

Nice, another scope floating war


So back to the topic,

I haven't watched the video in a long time, but regarding this...

First is Dave mentioned that providing the device doesn't have Earth ground on the IEC connector, then the scope ground lead can be connected anywhere.

It doesn't matter if a device physically has an earth pin or not. If you are trying to measure a mains circuit (i.e. on the primary-side of a transformer, or a circuit with no transformer at all), you cannot place the ground clip at any point which is not actually the building's earth. That's the only rule.

This is because earth and neutral are galvanically tied together somewhere in the building.

I think what he was saying was: If you are supplying your DUT from a power supply (i.e. wall wart) with only two AC input blades, the output (secondary-side of the transformer) is guaranteed to be galvanically isolated from mains. In this particular case, if your DUT is powered by this isolated supply, you can put your ground clip wherever you want in your DUT's circuit.

Emphasis added and this is right.

The part about neutral and earth being tied together is true, but you shouldn't put scope ground clips on neutral. Firstly, it may cause a fraction of neutral currents to return through the probe and scope's earth connection and trip the RCD in your building. Secondly, if you have no RCD or a broken one and neutral connection fails somewhere upstream, all your mains return current will flow through the probe and scope, possibly damaging them. Lastly, if both the neutral and earth disappear but live doesn't, you will have full mains voltage on all exposed metal of the scope.

You certainly can't connect the grounding clip to any arbitrary point of a circuit running directly on high voltage from mains. That's almost guaranteed magic smoke.

[tggzzz]

You're entitled to an opinion, but just remember that competent, experienced professional engineers have been killed by floating a scope to take a measurement. We're dealing with a beginner who doesn't fully understand the concept of ground here, it would be negligent to even suggest floating a scope to them. Yes there is equipment that is designed to be floating, that isn't what we're discussing here.

I won't disagree with the overall notion that this is not an intro-level subject, but do keep in mind that the one known example (AFAIK) of an experienced, professional engineer getting electrocuted this way involved the apparently common but obviously dangerous method of simply breaking off or disconnecting the ground pin using regular mains power.  Or course even an isolation transformer wouldn't have helped in that case since I believe it was the high-potential DUT that energized the scope.

Another good reason not to go floaty-scope is that it may not work very well--you may end up looking at a whole bunch of mains power artifacts.  In every case I've ever used isolation, isolating the DUT is the only reasonable way to do it.  The only exception I can think of outside of a high voltage laboratory--and I haven't run into it--is if you might need to float the scope with a low-voltage DUT that cannot easily be isolated from ground.  But that would be a more advanced topic with its own hazards, so I'll agree that in the current context, the answer to the idea of floating the scope should be a firm no.

"One known example"? I think you probably mean "one example posted on this forum". Absence of evidence (or other fatalities) is not evidence of absence.

Just to remind the OP, and others...

Example fatality from https://groups.io/g/TekScopes/topic/7632418#10795 with my emphasis

What makes it so dangerous to float the scope is that it is very easy for you to come in accidental contact with the floating scope chassis and receive a very bad shock, possibly lethal. One of my first customer contacts as a Sales Engineer for Tektronix was to call on the Sylvania Lighting Center in Danvers, MA and investigate a rumor about an engineer working there that was killed while using a Tek scope. I found it it was true. During lunch, one of the engineers was working alone in the lab on a lighting experiment that was using some 3 phase, 220 volt power. He needed to make some measurements between points none of which were at earth ground. So, he floated the scope . . . He even has the scope sitting on a scope cart with a sheet of insulation material between the bottom of the scope and the metal tray it normally sits in so the scope cart would not be "hot" with the scope. He also had a "tunnel" of plexiglas on both sides and over the top of the scope in a crude attempt to prevent anyone from accidentally touching the hot scope. The back was not covered with plexiglass in order to allow the fan to do its job and the front was not covered so the engineer could access the scope controls. This guy was WELL AWARE of the danger and took a lot of precautions to prevent shock . . . Bottom Line: He died anyway.

[tggzzz]

I'm thinking of getting a differential probe, but I'm uncertain when to use it, why I need to use it, and whether using it will be as accurate as using an expensive scope probe.

Make sure it is the right type of probe (or scope) for the job - and that includes the right type of differential probe; yes there are several.

FFI at
https://www.eevblog.com/forum/projects/isolation-transformer-for-scope/msg2259465/#msg2259465
https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

[magic]

He also had a "tunnel" of plexiglas on both sides and over the top of the scope in a crude attempt to prevent anyone from accidentally touching the hot scope. The back was not covered with plexiglass in order to allow the fan to do its job and the front was not covered so the engineer could access the scope controls. This guy was WELL AWARE of the danger and took a lot of precautions to prevent shock.

Emphasis inverted
Nice cargo cult safety there and the outcome doesn't surprise me.

[Psi]

I'm sure this has been asked many times, but I'm confused about probe grounding and how a scope can get damaged.

Usually it's the device/circuit you're testing that gets damaged from probe ground errors, not the scope.

The device gets a dead short across some of its circuitry.
The scope gets an unusually large current flowing through its earth connections. but that's normally BNC terminals to the shielding case and the case to the mains earthing wire. Not through chips or components so it usually doesn't damage the scope. There are of course exceptions if the current has to go through PCB traces to get to mains earth. But usually they can handle a fair bit of fault current, more than the device your testing can.
And even when it does blow out a trace on the scopes PCB it's an easy fix for anyone who has the skill to use a scope in the first place.

[magic]

There are of course exceptions if the current has to go through PCB traces to get to mains earth. But usually they can handle a fair bit of fault current, more than the device your testing can.

That's true until you come up with something like grounding the negative of an SPMS primary, which basically goes straight to mains live for half of each power cycle. Then it's a race for survival between the rectifier, SMPS PCB and scope PCB.

The scope input may have an advantage of ground planes

[Psi]

yep, there are exceptions.
But if the scope is built properly and has a metal cage around it with the BNCs attached and a earth wire terminal to the mains cable then it's pretty unlikely to take out your scope.

[tggzzz]

yep, there are exceptions.
But if the scope is built properly and has a metal cage around it with the BNCs attached and a earth wire terminal to the mains cable then it's pretty unlikely to take out your scope.

Have you tried that, or is that statement based solely on assumptions?

What are your other implicit assumptions? If made explicit, would they change the outcome? I suspect they would

[bostonman]

The scope gets an unusually large current flowing through its earth connections. but that's normally BNC terminals to the shielding case and the case to the mains earthing wire. Not through chips or components so it usually doesn't damage the scope.

I think my original confusion and why I initially posted the question was due to always hearing the scope can blow up. Normally I attach the ground lead to chassis ground, and, if it's obvious that is an isolated ground, I seek the negative/return. Most of my high voltage measurements have been at most AC input, but the majority are always low voltage secondary transformer side measurements. I have measured AC directly across the plug, but always make sure I know which side is neutral before doing so (I actually found a receptacle in my house that had a reverse HOT and NEUTRAL by way of one of those cheap LED testers that plug in and give you a sequence of lights that show the state of the receptacle - so it's not safe to assume that NEUTRAL is actually correctly connected).

My confusion came because I know the ground lead is Earth ground and touching a "hot" will cause a short circuit, but, the more I kept hearing that the scope can blow up, the more confused I got about the so called ground loop. Ignoring the scope probe blowing up in my hand, and blowing up the DUT (or hopefully the fuse first), I visualized a scenario where touching the ground lead to the wrong point will cause the scope to float so high that the internal electronics blows up due to nothing more than it floating. This doesn't seem to be the case. From my understanding, providing I keep the voltage below the maximum input voltage (400V ?), the scope isn't going to simply blow up because the ground lead floated (ignoring that I'm shorting a point to Earth ground and something will blow up).

As usual, maybe I over thought this and didn't need to ask the question. A very common situation is a power supply with three prongs, the scope with three prongs, scope probe ground lead touches a "hot" point, the fuse blows on the DUT, and nothing bad happens. A theoretical situation I imagine is that the instantaneous current is so bad that it blows open ground etches and pathways to ground in the scope causing the electricity on the point inside the scope that just opened to seek a path to ground and finds it through an adjacent IC chip thus "blowing" up the scope.

As for floating the scope, electronics that have ground points floating above ground, high voltage power supplies, etc... these are most things I stay away from because I don't grasp enough about grounding to feel safe gambling with anything (or being around high voltage especially if I'm alone). Sometimes "Earth" ground makes sense to me, other times I get confused. In some equipment, such as a weed wacker, Earth ground is a secondary grounding source to avoid being electrocuted should the neutral open but you won't get a shock should the third prong go missing. In other stuff, such as maybe a refrigerator, with the third prong broken off, the device will give you a shock. Rhetorically asking, why does one pose a shock risk and the other doesn't?

A silly question: why not install a PICO fuse in the ground lead so it opens before (hopefully) any injuries occur or damage to the scope?

Regarding differential probes, I assume they are not bullet proof meaning measuring something you shouldn't can cause an arc thus taking out the scope. In any case, it seems using them is certainly the safest method, however, are they reducing signal quality/bandwidth?

If my scope probe is some high end HP going into a high frequency scope (my four channel is 600MHz), and I use a diff probe, then I'm reducing signal quality.

What about measuring voltages without the ground lead connected, can this cause the scope to blow? Sometimes I've measured signals where it didn't make any difference whether ground was connected or not, but I assume the connection was through Earth ground.

[bdunham7]

"One known example"? I think you probably mean "one example posted on this forum". Absence of evidence (or other fatalities) is not evidence of absence.

The example you've repeated is commonly repeated and is the only one I'm aware of.  Are there others?  Perhaps, but I don't know and nobody else seems to know of any, so they're unknown to me,  you and the rest of us.  Unless you know of another, in which case please enlighten us.

[tggzzz]

The scope gets an unusually large current flowing through its earth connections. but that's normally BNC terminals to the shielding case and the case to the mains earthing wire. Not through chips or components so it usually doesn't damage the scope.

I think my original confusion and why I initially posted the question was due to always hearing the scope can blow up. Normally I attach the ground lead to chassis ground, and, if it's obvious that is an isolated ground, I seek the negative/return. Most of my high voltage measurements have been at most AC input, but the majority are always low voltage secondary transformer side measurements. I have measured AC directly across the plug, but always make sure I know which side is neutral before doing so (I actually found a receptacle in my house that had a reverse HOT and NEUTRAL by way of one of those cheap LED testers that plug in and give you a sequence of lights that show the state of the receptacle - so it's not safe to assume that NEUTRAL is actually correctly connected).

My confusion came because I know the ground lead is Earth ground and touching a "hot" will cause a short circuit, but, the more I kept hearing that the scope can blow up, the more confused I got about the so called ground loop. Ignoring the scope probe blowing up in my hand, and blowing up the DUT (or hopefully the fuse first), I visualized a scenario where touching the ground lead to the wrong point will cause the scope to float so high that the internal electronics blows up due to nothing more than it floating. This doesn't seem to be the case. From my understanding, providing I keep the voltage below the maximum input voltage (400V ?), the scope isn't going to simply blow up because the ground lead floated (ignoring that I'm shorting a point to Earth ground and something will blow up).

As usual, maybe I over thought this and didn't need to ask the question. A very common situation is a power supply with three prongs, the scope with three prongs, scope probe ground lead touches a "hot" point, the fuse blows on the DUT, and nothing bad happens. A theoretical situation I imagine is that the instantaneous current is so bad that it blows open ground etches and pathways to ground in the scope causing the electricity on the point inside the scope that just opened to seek a path to ground and finds it through an adjacent IC chip thus "blowing" up the scope.

As for floating the scope, electronics that have ground points floating above ground, high voltage power supplies, etc... these are most things I stay away from because I don't grasp enough about grounding to feel safe gambling with anything (or being around high voltage especially if I'm alone). Sometimes "Earth" ground makes sense to me, other times I get confused. In some equipment, such as a weed wacker, Earth ground is a secondary grounding source to avoid being electrocuted should the neutral open but you won't get a shock should the third prong go missing. In other stuff, such as maybe a refrigerator, with the third prong broken off, the device will give you a shock. Rhetorically asking, why does one pose a shock risk and the other doesn't?

A silly question: why not install a PICO fuse in the ground lead so it opens before (hopefully) any injuries occur or damage to the scope?

Regarding differential probes, I assume they are not bullet proof meaning measuring something you shouldn't can cause an arc thus taking out the scope. In any case, it seems using them is certainly the safest method, however, are they reducing signal quality/bandwidth?

If my scope probe is some high end HP going into a high frequency scope (my four channel is 600MHz), and I use a diff probe, then I'm reducing signal quality.

What about measuring voltages without the ground lead connected, can this cause the scope to blow? Sometimes I've measured signals where it didn't make any difference whether ground was connected or not, but I assume the connection was through Earth ground.

There is no point in people quickly and poorly rehashing information in high quality sources.

Most of those points are addressed in the references given in the "Safety and Praxis" section of https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/ Other more general information is in the "Formal" section. Have you read those? Naturally they aren't the only sources of decent information.

As to the fuse in the ground lead, firstly inductance, and secondly Murphy's law indicates that the fuse will be protected by semiconductors.

[TimNJ]

Maybe I'm just adding to the noise by giving yet another explanation. But, really, it's nothing really to do with any wishy washy concept about "ground" or about "what is ground, after all?". It's really just a matter of basic circuit analysis.

If your scope's ground clip is tied to your building's "earth node" (and it is, via third prong of the inlet), and if you place the ground clip in a circuit which also contains the building's "earth node", then the clip better be placed on the earth node or else you have a logical error (a.k.a. a short circuit!)

[magic]

The point of having no fuse in the grounding lead is to hopefully blow up the DUT with a spectacular smoke before you connect the probe tip and blow up the scope's frontend

(Which is to say, even if the fuse is not protected by semiconductors, you are still on a straight path to immediate destruction when probing like that).

[nctnico]

1) I personally feel floating of a scope should not even be mentioned, especially not to someone who is slowly learning about "earth" "zero volt" and the like. The idea should not even be enterained untill those concepts are clear as day.

And then the idea should stay an idea as in: DON'T EVER FLOAT AN OSCILLOSCOPE!

[nctnico]

"One known example"? I think you probably mean "one example posted on this forum". Absence of evidence (or other fatalities) is not evidence of absence.

The example you've repeated is commonly repeated and is the only one I'm aware of.  Are there others?  Perhaps, but I don't know and nobody else seems to know of any, so they're unknown to me,  you and the rest of us.  Unless you know of another, in which case please enlighten us.

Messing with mains is dangerous anyway. I lost a classmate / friend during middle school because he was tinkering with electricity. His parents found him in his bedroom.

[rsjsouza]

Anyone involved with this field is most probably curious, inventive and industrious - it doesn't take much from a newcomer to think that it is a practical idea how to circumvent something so simple without considering all the risks.

Exactly and for that reason alone we must do anything we can to discourage such dangerous practices....even talking in any detail about this sets curious minds off to investigate.   

This is not protection but instead an irresponsible way to leave the gap open for careless experimentation by newbies.

Or do you think that someone simply told not to do something will refrain from doing so just because of that? Don't you remember how human curiosity works?

Anyhow, my PSA is:
<psa>
Never remove the protective ground or use an isolation transformer to power any test equipment (oscilloscopes, bench DMMs, signal generators, spectrum analyzers, etc) that were not designed for that and has exposed metal surfaces and terminals (also called "floating" a device). The metal will become elevated to the potential of the test equipment and will be an open circuit. If the operator (you) touches it, the circuit will be closed and current will flow through your body, potentially at elevated levels that will be enough to provoke severe burns, accidents or cardiac arrest (= potential death). This is not fearmongering.

The Portable DMMs, portable scopes, portable spectrum analyzers, portable wattmeters, portable whatever are "floated" by design and should be used as such. Regardless, always be careful during operation and never touch the metallic tip of one of the probes if the other probe is connected to the circuit - the effect might be the same as the previous case as the operator (you) can also close a circuit and suffer the consequences listed above.
</psa>

[nctnico]

The Portable DMMs, portable scopes, portable spectrum analyzers, portable wattmeters, portable whatever are "floated" by design and should be used as such.

Not even that! As a rule of thumb: Only equipment which has isolated inputs is designed to measure potentially lethal voltages. All other equipment will need special (isolated / CAT rated) probes in order to make them safe for measuring potentially lethal voltages.

Portable spectrum analysers, oscilloscopes, etc are typically NOT designed to be used on high voltages! You can touch all kinds of metal parts on these devices that are connected directly to the inputs or shielding.

[tautech]

Anyone involved with this field is most probably curious, inventive and industrious - it doesn't take much from a newcomer to think that it is a practical idea how to circumvent something so simple without considering all the risks.

Exactly and for that reason alone we must do anything we can to discourage such dangerous practices....even talking in any detail about this sets curious minds off to investigate.   

This is not protection but instead an irresponsible way to leave the gap open for careless experimentation by newbies.

Or do you think that someone simply told not to do something will refrain from doing so just because of that? Don't you remember how human curiosity works?

Can I say this again:
If we are to be all properly responsible in these days of cheap differential probes we should all be shouting in unison........... DON'T FLOAT YOUR SCOPE !

[Psi]

1) I personally feel floating of a scope should not even be mentioned, especially not to someone who is slowly learning about "earth" "zero volt" and the like. The idea should not even be enterained untill those concepts are clear as day.

And then the idea should stay an idea as in: DON'T EVER FLOAT AN OSCILLOSCOPE!

Maybe, "DON'T EVER FLOAT AN OSCILLOSCOPE UNLESS YOU ARE 50+ YEARS OLD AND HAVE A LARGE GRAY BEARD"
I think that gets the point across without lying to newb's, which I don't really like doing.

[Ice-Tea]

A silly question: why not install a PICO fuse in the ground lead so it opens before (hopefully) any injuries occur or damage to the scope?

It's important that you understand the reason why metal surfaces are earthed: it's an important safety measure. It ensures that no dangerous potential develops between a conducting, easily accesible surface and "earth" (which is also easily 'touchable' in many ways and forms). If you start fusing earth connections, you may save your equipment but loose your life.

[bostonman]

I had a spectrum analyzer at my old job where the Earth ground pin on the plug end got loose and wasn't making good contact. My point is, I can see an accident where someone has a floating scope by accident.

One thing to install would be a GFCI outlet. Also, I have a foot pedal, and, if I am measuring some voltages during a repair, I may use the foot pedal in hopes that my foot jolts off the pedal thus killing power to the DUT.

Anyway, I feel the need to emphasize not to float a scope has been stated enough.

The point of my initial question may have come off as not understanding short circuits, but it was more about does one short circuit actually cause the entire scope to go up in smoke. The video I watched helped show the pathway the scope probe ground will force a DUT to take, and certainly gave me a better visual on what exactly is going on.

The "blowing up the scope" portion of my question has also been answered and that is: the ground path(s) can blow up and open thus "blowing up the scope". The electronics MAY not get harmed, and the blown scope may be simply fixed by fixing the pathways to Earth ground that got damaged, however, electricity can arc or do whatever it wants to find the least resistance to ground thus maybe taking out components in the scope.

As for a differential probe, I assume diff probes can't replace regular probes due to whatever limitations they have such as (I assume) BW, rise time, etc...  In other words, I shouldn't put my scope probes on the shelf and use exclusively diff probes for everything, only when working on Earth ground circuits?

Is there a preferred affordable diff probe for home hobbyists?

[mag_therm]

When setting up a 2 channel differential measurement, first connect both probes to the same point of the two points to be measured and run DUT to check the CM rejection. Then move one probe to the other measuring point.

[james_s]

Differential probes, like any probes have limitations, and it's important to understand the limitations of the specific probe that you're using. They ALL affect the signal you are trying to measure to some degree, that is all probes, not just differential probes. Also some of the fastest scope probes you can get are differential however they are not high voltage rated, they exist because many high speed signals in modern equipment are themselves differential. I have a 100MHz differential probe and use it for applications that need the HV rating, for more general purpose stuff I use a general purpose passive probe. In practice, I rarely need to scope anything that has high voltage but on those occasions I do, the HV probe has no substitute.

[tggzzz]

1) I personally feel floating of a scope should not even be mentioned, especially not to someone who is slowly learning about "earth" "zero volt" and the like. The idea should not even be enterained untill those concepts are clear as day.

And then the idea should stay an idea as in: DON'T EVER FLOAT AN OSCILLOSCOPE!

Maybe, "DON'T EVER FLOAT AN OSCILLOSCOPE UNLESS YOU ARE 50+ YEARS OLD AND HAVE A LARGE GRAY BEARD"
I think that gets the point across without lying to newb's, which I don't really like doing.

There are old bikers probers and there are bold bikers probers. But there are very few old bold bikers probers.

Once after plugging a scope in, I discovered someone else had floated it and not unfloated it.

I was not a happy camper, especially as nobody came forward to say they had floated it and what voltages had been applied.

[tggzzz]

As for a differential probe, I assume diff probes can't replace regular probes due to whatever limitations they have such as (I assume) BW, rise time, etc...  In other words, I shouldn't put my scope probes on the shelf and use exclusively diff probes for everything, only when working on Earth ground circuits?

RTFReferences you have been given.

Is there a reason you trust hastily written stuff from random people more than considered statements from experts?

[tautech]

As for a differential probe, I assume diff probes can't replace regular probes due to whatever limitations they have such as (I assume) BW, rise time, etc...  In other words, I shouldn't put my scope probes on the shelf and use exclusively diff probes for everything, only when working on Earth ground circuits?

This is a large topic that really only ones observant experience can answer for you as another user may have very different requirements however general basic knowledge is required for all that are to use a scope and there are several industry standard papers that can easily be found with a Google search for 'differential probing techniques'.

The most common and simplest used is with 2 probes, both with Reference leads removed and Math Minus engaged to display the differential measurement between them.

Is there a preferred affordable diff probe for home hobbyists?

These days several and at reasonable prices however hobbyists need learn their investment in a scope doesn't stop there and imagine they can do everything on the cheap and mostly with some ingenuity you can however there are times when only the proper tool can safely do the job.

Unlike investment in a scope, a differential probe may outlast several scopes so this need be factored in as does the value you place on your life and the risk on it by taking shortcuts especially when dealing with lethal voltages.

[NiHaoMike]

Unlike investment in a scope, a differential probe may outlast several scopes so this need be factored in as does the value you place on your life and the risk on it by taking shortcuts especially when dealing with lethal voltages.

That's only likely if you buy a differential probe of vastly higher bandwidth than the scope, very unlikely for hobbyists.

[tautech]

Unlike investment in a scope, a differential probe may outlast several scopes so this need be factored in as does the value you place on your life and the risk on it by taking shortcuts especially when dealing with lethal voltages.

That's only likely if you buy a differential probe of vastly higher bandwidth than the scope, very unlikely for hobbyists.

Yes well ask yourself what usage types and frequencies are differential probes most used for.

Commonly well under 100 KHz are the normal requirements......pretty basic needs however it still needs to be done safely !

[bostonman]

RTFReferences you have been given.

Is there a reason you trust hastily written stuff from random people more than considered statements from experts?

I think that was uncalled for. Although your point is correct, and maybe some of what I've asked should have just been researched rather than asking so many questions on here. Sometimes input from a group provides more insight (or at least I feel) than an individual PDF online, and who better to provide feedback on diff probes for hobbyists than the many hobbyists on here who have had experience with the pros and cons?

The takeaway I've got from the input on this particular topic is actually emphasis on safety. Although I'm knowledgeable about grounding and safety, nor would I attempt to run anything ungrounded, it shows that many people on here rather take time typing about safety rather than just blurt out some quick answer to a question.

Maybe I should have never started this thread and just read PDFs online, however, also think about how many people are going to come across this thread and say to themselves: stop! I better avoid haphazardly taking measurements on power supplies. A good example: again, although I'm understanding of grounding, I'm giving more thought to grounding the probe lead than I did before.

Also, before, I thought a differential probe was used to do a subtraction between two voltages when there is a point higher than ground and not for safety purposes. Now I have learned the actual advantages of a diff probe, a visual on Earth ground path through a scope thanks to the video referenced early on in this thread, and a newer level of concern for safety.

[tautech]

Just when you think you have a handle on this a job like below comes up, a 230VAC 180A dual thyristor phase control welder using UJT's for phase control timing.
Each thyristor controls opposite halves of the mains sinewave and their phase control MUST be balanced for smooth operation, for the benefit of both the weld quality and to protect the welding transformer from any mechanical stress due to it running rough.

The dual gang 100k pot is for welding current control which results from the period of phase control and the two 100k trimmers are to both set the balance and max output.

Tell us how you might scope this and with what probes to make these adjustments safely. 

[bdunham7]

Tell us how you might scope this and with what probes to make these adjustments safely. 

How about putting a small load on the output and then connecting the ground and tip of a standard 10X probe to each lead with just about any old scope?

Or, assuming the N input is ground-referenced, just use the 10X probe with no ground lead at the top of the thyristors where they are connected.  That will look a bit different, but I think you could make the adjustments OK.

[tggzzz]

Unlike investment in a scope, a differential probe may outlast several scopes so this need be factored in as does the value you place on your life and the risk on it by taking shortcuts especially when dealing with lethal voltages.

That's only likely if you buy a differential probe of vastly higher bandwidth than the scope, very unlikely for hobbyists.

My differential probes both have a bandwidth that is a fraction of the bandwidth of my slow scopes. Nonetheless they are useful, used, and necessary in some circumstances.

Or perhaps you would like to narrow your statement to one of the different classes of differential probes?

[tggzzz]

RTFReferences you have been given.

Is there a reason you trust hastily written stuff from random people more than considered statements from experts?

I think that was uncalled for.

After you have twice ignored the solid references and persisted, IMNSHO it was called for. This is a serious subject that could damage equipment - or cause serious injury to you or someone that reads the thread later.

https://www.eevblog.com/forum/projects/why-can-not-using-a-differential-probe-blow-up-scope/msg4369297/#msg4369297
https://www.eevblog.com/forum/projects/why-can-not-using-a-differential-probe-blow-up-scope/msg4369855/#msg4369855

There's a UK proverb: "you can lead a horse to water, but you can't make it drink". But where safety is concerned, you have to make it drink!

Although your point is correct, and maybe some of what I've asked should have just been researched rather than asking so many questions on here.

My point is correct, and you should done the research for the points I mentioned earlier.

Sometimes input from a group provides more insight (or at least I feel) than an individual PDF online, and who better to provide feedback on diff probes for hobbyists than the many hobbyists on here who have had experience with the pros and cons?

By all means read random posts and gather what you can, but they are not a subsititue for considered expert statements and understanding.

Over the past few years politicians in several countries have mouthed a manta "don't trust the experts" (but do trust me). That's because they want gullible people to believe nonsense that benefits them personally. I will fight against that way of thinking (and I use that word loosely) spilling over into technical spheres. I have no intention of elaborating; this forum is a politics-free zone.

The takeaway I've got from the input on this particular topic is actually emphasis on safety. Although I'm knowledgeable about grounding and safety, nor would I attempt to run anything ungrounded, it shows that many people on here rather take time typing about safety rather than just blurt out some quick answer to a question.

Would you have it any other way?

Maybe I should have never started this thread and just read PDFs online, however, also think about how many people are going to come across this thread and say to themselves: stop! I better avoid haphazardly taking measurements on power supplies. A good example: again, although I'm understanding of grounding, I'm giving more thought to grounding the probe lead than I did before.

Starting this thread was just fine, and in no way a problem. Threads like this are a superb way to get pointers to what is necessary; they can help avoid lots of dribbling opinions and marketing fluff/dross, and be quickly guided to the core information.

We all start out as beginners, whether or not we know it. I've been doing this kind of thing for 50 years, and I know how much I don't know, how easy it is to delude myself that I understand it, and how easily nasty mistakes can be made.

One simple example: the concept of "ground" is a fiction or analogy that has some value in a limited number of circumstances. No more. In reality, in many practical cases all that matters are the electromagnetic fields.

Also, before, I thought a differential probe was used to do a subtraction between two voltages when there is a point higher than ground and not for safety purposes. Now I have learned the actual advantages of a diff probe, a visual on Earth ground path through a scope thanks to the video referenced early on in this thread, and a newer level of concern for safety.

If you read the references, you would realise that there are some differential probes just like that. You will also realise their limitations, and when they should/shouldn't be used.

[bostonman]

I'm uncertain if this helps, but I pieced together (what I think) are various configurations of grounds and where one may place the probe Earth ground to provide feedback to not only me, but to others who question placement.

I guess maybe take a look and comment on whether the placement would be safe/correct.

A recent real life configuration that I came across was during repairing my WES51 soldering station which is somewhat mimicked in example C of the attached. Earth ground is screwed to the metal plates of the transformer, and the wire from the outside shell of the soldering iron is also screwed to the transformer. The twist is that the thermistor seems to be welded to the soldering iron shell because each side of the thermistor to Earth ground measures 1 ohm to Earth ground on one side of the thermistor and 2 ohms on the other side of the thermistor to Earth ground.

In example D, I think connecting the scope probe Earth ground to the case is safe, but would the scope measure correctly or would it be wrong because the grounds are different?

[bostonman]

Anyway, another method for you to probe, without having to cut off the earth from the scope mains plug...would be to supply your DUT via a mains isolation transformer, (MIT) and be sure that this M.I.T   doesnt connect earth to  either of the secondary outputs. Then you can put your probe ground anywhere on that DUT,  and it doesnt have to be a diff probe in that case. Obviously still dont touch the high voltage isolated outputs.

I'm resurrecting an old thread because I plan to purchase a Micsig diff probe. At the moment I've been tinkering with their current probe and I'm impressed with it. I'm sure these are not top of the line items, but, for the price, seems they are good for my needs.

Anyway, I don't plan to do high voltage measurements such as high voltage switch mode power supplies, but, should I come across the need, I have a question about adding another layer of protection such as an isolation transformer.

First off, the Micsig diff probe and current probe both get their power by USB and the suggestion is to power it from the scope USB which seems counterproductive since a failure could cause a short from the probe to the scope thus blowing up the USB port. Seems to me buying a 5V wall wart would be the safest approach, but those wall warts are usually not very stable. Is it probably better to power it with something else?

About isolation transformers, I am not strong in the area of floating grounds, isolation, etc... so I want to double check my thought on this. Technically I understand floating grounds, but then I sometimes overthink things and confuse myself.

I'm assuming an isolation transformer varies from a standard one because it isolates the primary and secondary with a magnetic field with no common Earth ground. Now if a power supply is powered by the isolation transformer, is the power supply floating in a fashion that would be similar to cutting off the Earth ground on the oscilloscope (I'm going to repeat the advice of everyone which is NEVER FLOAT THE OSCILLOSCOPE - I'M JUST USING THAT AS EXAMPLE - BUT I WANT TO MAKE SURE I'M STATING NOT TO CUT THE THIRD PRONG).

[bdunham7]

First off, the Micsig diff probe and current probe both get their power by USB and the suggestion is to power it from the scope USB which seems counterproductive since a failure could cause a short from the probe to the scope thus blowing up the USB port. Seems to me buying a 5V wall wart would be the safest approach, but those wall warts are usually not very stable. Is it probably better to power it with something else?

You an use a small battery pack like the ones they sell for portable charging.  But keep in mind that your differential probe is not completely isolated from either the oscilloscope ground or the USB connector!

I'm assuming an isolation transformer varies from a standard one because it isolates the primary and secondary with a magnetic field with no common Earth ground. Now if a power supply is powered by the isolation transformer, is the power supply floating in a fashion that would be similar to cutting off the Earth ground on the oscilloscope (I'm going to repeat the advice of everyone which is NEVER FLOAT THE OSCILLOSCOPE - I'M JUST USING THAT AS EXAMPLE - BUT I WANT TO MAKE SURE I'M STATING NOT TO CUT THE THIRD PRONG).

Actually most (but not all) transformers are isolation transformers in that sense.  However, a typical isolation transformer setup does not 'break the ground', but rather passes it through if it has one.  An isolation transformer is not equivalent to cutting off your ground pin and a typical isolation transformer on its own will not 'float' an oscilloscope or anything else.  What the transformer does is provide an output that is not ground referenced.  Your typical mains power outlet has one side (neutral) that is bonded to ground at your service panel and stays pretty close to ground potential.  The other one (hot) has the full 120VAC relative to ground potential.  That's the one that zaps you if you touch it, since you are usually at ground potential.  An isolation transformer breaks this ground reference and neither side is referenced to ground, so touching one side or the other will not cause significant current to flow through you. 

Using an isolation transformer properly and safely takes some care and thought.  If you provide a ground reference for one side--or even any part of the DUT circuitry--then other parts may then have significant potential to ground and be a shock hazard.  If you clip two or more probe ground clips to parts of your DUT that have different potential, you will still blow up your oscilloscope.

[bostonman]

A battery would probably be a good approach, but not sure I want to deal with battery cost; or maybe a rechargeable. I'll have to see. I was thinking about building a linear 5V regulator, but many more quicker options exist for such a low voltage/current.

Anyway, seems like an isolation transformer is a safe approach but could result in more safety issues if one isn't careful.

Currently I'm browsing Hammond isolation transformers because I'm curious what options exist.

[bostonman]

I meant to ask, one of the Micsig differential probes (the higher voltage one) has x50 and x500 setting (I assume x50 covers one input voltage range and x500 has a higher input voltage range).

I haven't used an oscilloscope with x10 in quite some time. Most scopes I've dealt with are 1M and 50ohm. If I remember correctly, if setting the channel to x1, 1V/div was just that. If x10 was selected, it would change the 1V/div to 10V/div.

If I buy the diff probe with x50 and x500, and my scope doesn't have an option to select x50 and x500 (let's assume just x1 and x10), does this mean a 1V DC input signal will show as 50V (assuming I select x50) or 0.02V (1/50)?

If so, does this mean I can't use these diff probes with an older scope that doesn't have a menu option to select (what I consider) oddball attenuation settings?

[bdunham7]

does this mean a 1V DC input signal will show as 50V (assuming I select x50) or 0.02V (1/50)?

If so, does this mean I can't use these diff probes with an older scope that doesn't have a menu option to select (what I consider) oddball attenuation settings?

The latter (x50 implies 50X attenuation, not amplification) and on scopes without a 50X selection (which was most of the older ones) you simply have to use mental math on the fly.  Or post-it notes.

[vk6zgo]

There are of course exceptions if the current has to go through PCB traces to get to mains earth. But usually they can handle a fair bit of fault current, more than the device your testing can.

That's true until you come up with something like grounding the negative of an SPMS primary, which basically goes straight to mains live for half of each power cycle. Then it's a race for survival between the rectifier, SMPS PCB and scope PCB.

The scope input may have an advantage of ground planes

The RCD in the bench feed will win the race.
Classic Oscilloscopes didn't rely upon a PCB track for grounding, but modern ones do, though I would still put my bets on a dedicated RCD.

The only time I ever got caught probing SMPS was when somebody borrowed the DUT isolating transformer, & I forget for a scond.
Big flash, "snap" went the RCD, & no more power to the DUT.

If all the person wants to do is to see if there is mains at the input to the DUT, a "quick look" with the 'Scope ground clip removed will do the job safely, as will the Quasi differential method using two channels, but if you need to critically monitor the voltrage across Live & Neutral at the DUT power input, you are stuck with buying or borrowing a differential probe or isolating the DUT with an appropriate transformer.

[bostonman]

The latter (x50 implies 50X attenuation, not amplification) and on scopes without a 50X selection (which was most of the older ones) you simply have to use mental math on the fly.  Or post-it notes.

Sorry, I realized the x went in the wrong spot after too late.

So, it's not going to attenuate the signal by 500? I was thinking if I got a 500X diff probe and put 500V DC into the scope, I'd only see 1V DC.

If this was the case, I feared not being able to see small signals because the scope couldn't turn up the V/div enough. Say if it were a 500mV DC signal, what would I see on the scope?

As for the attenuation, from what I'm seeing, there are two or three different models with different levels of protection. One model has 10X and 100X, and a high isolation voltage model has 50X and 500X. I was going to purchase the 10X and 100X model (with lower isolation voltage) because I had concern I wouldn't see small signals if I used say an old 465 oscilloscope.

[bdunham7]

I use 'x50' and '50X' interchangeable, perhaps that isn't right.  Yes, your 500X probe will give a 1V output to the scope if you put 500V on its terminals.  You aren't going to have much luck with small signals with these probes--just look at the specifications for noise and CMRR.  If you are referring to the Micsig DP10013, the input referred noise alone is 230mV_rms on the 500X range.

[bostonman]

Well, the x in front or after the 50 (or 500) does imply the same thing unless I'm just too focused on algebra. Maybe it's also why the spec for the diff probe is confusing.

I have a 465 scope and a digital one (not sure if it goes up to 500 scale though). Regardless, I confused myself because I haven't dealt with x1 or x10 scopes in years; and I only used x10 when I had a probe connected or x1 for a straight coax.

Does setting it to x1 or x10 actually change the attenuation? An old scope had x1 and x10, if I connected a probe while it was on x1, the scaling would be off, but I couldn't blow up the scope (ignoring whether I measure a voltage that exceeds the front end spec). If it was on x10 and I connected a straight coax, the scaling would be off.

In the case of my digital scope, it has high Z and 50ohms. It's rare I ever use 50ohms because I don't want to risk blowing up the Agilent hybrid chip, so I'll use a 50ohm terminator with the scope in high Z.

I'm assuming keeping my digital scope set to x1 or x10 will not have any affect on whether the diff probe is set to x50 or x500, it will just give me the wrong voltage level reading.

With this being said, I was going to purchase the x10 and x100 model, but maybe it's best to go with the x50 and x500 even though my measurements may only be for 120V AC?


Edit:

If you are referring to the Micsig DP10013, the input referred noise alone is 230mVrms on the 500X range.

Sorry, I should have included that I was ignoring the noise floor. I was using basic math to understand things better.

[bdunham7]

Does setting it to x1 or x10 actually change the attenuation?

With this being said, I was going to purchase the x10 and x100 model, but maybe it's best to go with the x50 and x500 even though my measurements may only be for 120V AC?

If you look at your old 465, there are two gray squares under the number ring on the vertical scale knob.  The one on the left is the actual V/div setting or what you would read with a direct (1X) connection.  The square on the right is what you read with a 10X probe.  Clearly you aren't making any change to the scope itself when you change probes.  A more modern scope works in an interactive way, but it still comes out the same.  If you set the scope to 10X and 10V/div, it is actually measuring 1V/div at its input.  What happens when you change the probe scale setting may vary a bit, but typically it changes the displayed V/div, not the actual sensitivity.  In any case, just remember that the actual scope scale is displayed V/div divided by probe scale factor.  10V/div with a 10X probe scaling = 1V/div.

The 10x/100x DP10007 is fine for 120VAC.  It reads up to 700V, but is protected to at least 1400V AFAIK.  And it has lower input referred noise.  Apparently the CMRR still sucks, as per Dave Jones.

[bostonman]

Actually, I just found a spare (i.e. not being used) x50/x500 model at work and ran some tests with the results being as expected.

Measuring the calibration output (5Vp) with a scope probe, at x10 setting: 5Vp.

Using the diff probe without changing anything on the scope, setting the Micsig to x50, I measured 1Vp; and it looked fairly clean.

Setting the Micsig to x500, I measured 100mVp on the scope, but it was very noisy.

Obviously once I set the scope to x50 or x500, the V/div matched the input correctly and I got the correct reading.

but typically it changes the displayed V/div, not the actual sensitivity.

This actually what I was getting confused over. The x1 and x10 on the scope wasn't necessarily the confusing part, but more about whether I inject say 100Vp DC voltage through a scope probe (assuming the scope is capable of handling such a voltage - but I'm just using 100Vp as a round number) while the scope is at x10, and accidentally hit x1, do I blow up the scope or is it just going to display 10Vp?

As for the Micsig diff probe, looking at the differences between the three (https://www.micsig.com/Differential%20Probe04/), I may be better off with the x10/x100 because I don't plan to measure anything above 700V and I'll end up having too much attenuation on the x50/x500 if I want to measure just 20V DC; it will just be 400mV at x50 with lots of noise due to the noise floor.

[Ground_Loop]

You should watch Dave's video on exactly this topic.

The other question is regarding somewhat my original question. How does the scope physically blow up by touching the Earth ground lead to a high voltage/current source? I understand the physical path current takes (more now after watching the video), but I imagine any current is just pulled to Earth ground bypassing all the electronics the scope until (hopefully) a circuit breaker or fuse pops or the scope probe blows up in your hand because the thin ground wire couldn't handle the high current thus immediately heating until it melts off.

In other words, I touch the scope ground lead to a mains, and the probe can blow up in my hand and/or the ground wire on the probe can melt until it's an open circuit. Simple enough to understand and certainly not a good position to be in because of injury risks. Assuming the ground lead and/or probe doesn't blow up in my hand to open the circuit (and ignoring whatever damage can happen to the unit I'm measuring. i.e. opening a trace, popping the fuse, etc...), I'd assume the only "damage" to the oscilloscope would be a ground wire opening inside the scope which would (hopefully) be a simple fix (open the scope, solder a new wire from the BNC to Earth ground, and remove the old burnt wire).

Speaking from experience here, not with a scope, but a bench multimeter (HP 3457).  While probing in a circuit plugged into mains with a non-polarized two prong plug, I connected the negative meter lead to the hot chassis.  Several traces literally evaporated from the multimeter board and also took out numerous components.  I was able to repair it, but I now have a differential probe.

[bostonman]

Until I began this thread, I had a grasp on probe Earth ground, now I don't.

Before I would just connect the probe ground to the chassis. If I didn't see a direct connection to the chassis, I'd clip on to the negative leg of maybe a capacitor.

Now that I've given thought about things, if the PCB is tied to chassis ground, then the probe ground doesn't need to be connected at all. If the PCB is isolated from Earth ground, then would I be causing ground loops by connecting the secondary isolated ground to Earth ground?

In any case, I'll be ordering a diff probe, and may even get both the x10/x100 and x50/x500.

[bdunham7]

secondary isolated ground to Earth ground?

Da who?!? 

Draw some pictures!  Simple schematics can often help you sort things out.  There should be no reason to be terrified of grounds or chassis.  Anything with a 3-prong plug, a metal chassis and a printed circuit board probably is grounded and should remain so at all times.