Mains and the oscilloscope

[Neukyhm]

Hi, I'm new here, I hope you can help me with this. I already know the answer, so the reason I'm asking for this is because I want to make sure that my house won't burn.

I'm a physicist, but in my college the didn't teach us how to use a oscilloscope, just how to do some easy measurements.

The question is: is it safe to measure mains 220AC with A-B operation so probes ground clips are not used? (assume the probe is x10)

Again, I know the answer, it is safe. I just want to make sure, you guys are more experienced than me.

[Dubbie]

Not really safe. To be by the book you should use a high voltage cat rated probe.

Edit: across the web you may get very different responses to that given by this forum. Keep in mind that this forum has many experienced professional members and bad advice is very rarely given here. Just because others on other less rigorous forums have gotten away with bad practice does not mean you will.

[Brumby]

The right probes are critical to safety.

Watching this is a good idea as well....

[Ian.M]

You are under the illusion that your AC supply is a nominally 220V RMS, 50Hz sinusoid, maybe with some flattening of the crests due to loading by SMPSUs without power factor correction.  It is not.   

Firstly the voltage may be at the upper end of its EU harmonised tolerance range as your electricity supplier has no legal obligation to act on overvoltage within that range, so instead of  the 311V peak you expect, you could have up to 358V.  Then you must consider transients - see http://ecmweb.com/contractor/looking-sources-transient-overvoltages - and up to 1KV spikes riding on the sinusoid waveform are common to the point of being expected.   

Connect the tip of an ordinary scope probe to your mains and there is a high probability of it being exposed to spikes well over 1KV and it breaking down catastrophically.   

The situation is better if you have a good quality surge suppressing power filter (with transorbs, delta connected filter caps on input and output and at least one common mode choke) + an isolating transformer between the scope probe and the wall outlet - as the filter and transorbs considerably reduce the maximum spike amplitude, clipping ones on the peaks more heavily, and the transformer adds further low pass filtering as a typical fast transient doesn't have enough energy to significantly change the flux in its core, but I'd still want a CAT II rated probe with an adequate safety margin.  A 500V peak rated probe will only have a 40% margin - barely enough to be safe after such a filter + transformer.

[DimitriP]

I already know the answer, so the reason I'm asking for this is because I want to make sure that my house won't burn.

When you  "know", "you know". If you have to ask you don't "know".

What I don't know is the reason (or lack thereof) for the fascination with probing the mains voltage.

[BrianHG]

Only if you must.  Get a normal 240v to 24v ac transformer.  Make sure there is no direct connection to the mains and the transformer's secondary 24v side, use one with safety approvals like a wall-wart which has 24vac, or 12vac out.  Probe the signal on the 24v ac side with 1 probe normally and multiply your readings by 10.  The smaller the transformer, like 2 watts or less, the better the high frequency response.  It may be useful to add a resistor load on the 24v side like a 1k 2 watt resistor.  (The resistor will get warm)  An optional 100nf capacitor in parallel with the 1k resistor if you want to get rid of high frequency noise.

You should see your 50hz sine wave.

[tggzzz]

Hi, I'm new here, I hope you can help me with this. I already know the answer, so the reason I'm asking for this is because I want to make sure that my house won't burn.

I'm a physicist, but in my college the didn't teach us how to use a oscilloscope, just how to do some easy measurements.

The question is: is it safe to measure mains 220AC with A-B operation so probes ground clips are not used? (assume the probe is x10)

Again, I know the answer, it is safe. I just want to make sure, you guys are more experienced than me.

Engineers consider two things: how X works plus how X can fail. You have only considered how that can work.

The trouble is that there can be many many subtle and surprising ways that things can fail, and it is difficult for anybody (especially a beginner) to think of them all.

Back to your problem. If there isn't a simpler way of achieving your (unstated) goals, you should use the appropriate type of probe. I suggest you have a look at the safety references in https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

[danadak]

These might help -

http://www.tek.com/document/application-note/three-facets-floating-measurement-solutions

http://www.newark.com/pdfs/techarticles/tektronix/FFM.pdf


Regards, Dana.

[David Hess]

The question is: is it safe to measure mains 220AC with A-B operation so probes ground clips are not used? (assume the probe is x10)

Yes, it is generally safe but if it is something that you do regularly, better probes and maybe a better oscilloscope with appropriate safety ratings should be considered.  The peak voltage from 240VAC is lower than many probe and oscilloscope input ratings and with a x10 probe, it helps that the two are in series.  At least my older oscilloscope can handle 400 to 500 volts peak directly connected to their inputs.  Failure do to a severe overload will probably be limited to a blown input.

The common mode range of A-B mode is poor (you will be stuck at 50V/div) so the sensitivity will be limited.  A dedicated differential probe will have much higher sensitivity.

[raspberrypi]

Bend the tines of a two forks and attach them together using a pickle. Stick the tines into your outlet forming a series circuit with the pickle. If the pickle burns evenly at both contact points you have an AC waveform, and at least 100 volts. This is the UL approved way to test this and meets ISO9001 standards. 

[Neukyhm]

What I don't know is the reason for the fascination with probing the mains voltage.

Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains.

[rstofer]

What I don't know is the reason for the fascination with probing the mains voltage.

Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains.

I too have an issue with measuring high voltages with a scope.  I don't even own the appropriate PPE (Personal Protective Equipment) like gloves.

Maybe x100 probes are a help.  They will certainly help protect the scope and they will probably have an appropriate safety rating.

A-B measurements have been a scope feature for generations.  It would be worth reading the User Manual for limitations.

It is my understanding that Residual Current Devices are common in Europe and that they protect the entire panel.  That certainly reduces the shock hazard.  In the US, we have Ground Fault Circuit Interrupters on branch circuits to various specified locations but they don't protect the entire panel and they trip at a very low number of mA.  But they are only used for outlets in wet or outdoor locations, not my spare bedroom.

Then there is the 'oops' bit.  How much energy is available on the circuit?  In the US, it will generally be less than 10,000A for residential services and probably lower than 5000A.  This is further limited by the branch circuit wire size but still, there can be a heck of an arc.

All in, I prefer to stay below 50V and more likely below 12V.  I'm especially fond of 5V.  And, no, I have no interest in SMPSs.  If I need one, I'll just buy it.

[tggzzz]

What I don't know is the reason for the fascination with probing the mains voltage.

Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains.

Oh, for heaven's sake! Why don't you ask the question you want answered, rather than encourage everybody to waste their (remaining) life giving useful answers to things that don't interest you.

How would you feel if I asked "is it OK to eat whelks" , and then said "listen, whelks are just an example of seafood I want to eat"?

[julian1]

I had a similar sense of trepidation about the A-B thing. Eventually I decided it made no sense not to buy the correct tool for the job - instead of relying on a workaround. A high-voltage differential probe, that's battery powered provides the necessary isolation. About  $200 second-hand on ebay.

[raspberrypi]

What I don't know is the reason for the fascination with probing the mains voltage.

Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains.

Oh, for heaven's sake! Why don't you ask the question you want answered, rather than encourage everybody to waste their (remaining) life giving useful answers to things that don't interest you.

How would you feel if I asked "is it OK to eat whelks" , and then said "listen, whelks are just an example of seafood I want to eat"?

What if I have a friend who may or may not have access to high voltages and may or may not own an oscilloscope, and say the probes accidentally touch the hot and the neutral...

[Ian.M]

Odds are, in North America, you'll get away with it,  Unless there is a floating neutral,. your domestic mains is under 200V peak (neglecting spikes) so for a typical 500V rated probe, that still leaves you 300V margin to handle spikes.  Its not ideal ore perfectly safe, you should  use a proper Cat II rated probe, and you *REALLY* shouldn't probe before the appliance fuse.

Clipping the probe ground to neutral is another matter - that's just dumb, and can melt the ground clip lead, the probe coax lead or even melt traces off the PCB inside your scope.   A GFCI breaker will trip but probably wont save you because it interrupts the live(s) not the neutral, and the problem is the potential difference between neutral and ground, usually caused by either poor bonding at the panel or heavy loads on the circuit.

[Kappes Buur]

See the adventure of ElectroBoom probing an AC circuit 

[Ian.M]

An excellent example of how *NOT* to do it.  e.g. 2:00 - 2:08 - floating a scope with scotch tape!

[highlight]/////////////// CAUTION /////////////// CAUTION //////////////// CAUTION //////////////// CAUTION ///////////////////[/highlight]

[grumpydoc]

An excellent example of how *NOT* to do it.  e.g. 2:00 - 2:08 - floating a scope with scotch tape!

[[highlight]]/////////////// CAUTION /////////////// CAUTION //////////////// CAUTION //////////////// CAUTION ///////////////////[/highlight]

I think ElectroBoom specialises in "How NOT to do it" videos. My main worry is that fact might not be obvious to everyone.

[Ian.M]

Yes.  We could make good use of a full width animated gif 'caution' black and yellow hazard tape 'smiley' for use above and below all ElectroBoom and some other videos!

Alternatively, if Dave installs the SMF hilight tag mod: http://custom.simplemachines.org/mods/index.php?mod=3825 we could simply do what I tried to above.

[BrianHG]

See the adventure of ElectroBoom probing an AC circuit 

It would have been funnier if after each time he had an explosion, the scope should have been blanked out, and after the next scene cut, you would see the scope on his desk to the right & a new one in use in front of him until that one is destroyed...

[grumpydoc]

Yes.  We could make good use of a full width animated gif 'caution' black and yellow hazard tape 'smiley' for use above and below all ElectroBoom and some other videos!

Alternatively, if Dave installs the SMF hilight tag mod: http://custom.simplemachines.org/mods/index.php?mod=3825 we could simply do what I tried to above.

[glow] seems to work (in preview, anyway).

/\/\/\/\/\/ CAUTION /\/\/\/\/\/\/

[Ian.M]

It would have been funnier if after each time he had an explosion, the scope should have been blanked out, and after the next scene cut, you would see the scope on his desk to the right & a new one in use in front of him until that one is destroyed...

Rigol's Legal department should sent him a cease and desist letter requiring him to obscure all Rigol branding and model numbers on any equipment shown in his videos!      

Mehdi Sadaghdar is obviously a graduate of the 'Elect-a-Pope' school of engineering. He reminds me of 'Klaus' the fork-lift driver.

[rsjsouza]

What I don't know is the reason for the fascination with probing the mains voltage.

Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains.

Mains is a different beast when compared to, say, a local high voltage power source whose characteristics such as maximum voltage/current and source impedance are well known.

Mains has a low impedance that can yield currents at the order of kA as mentioned by rstofer, which will be only relevant if a breakdown or an arc happens between the power source and the test equipment (oscilloscope, probes, etc.). The chances of such event occur are defined by either a transient or an accidental mistake.

A transient is extremely hard to predict and can reach spikes at the order of kV which can easily exceed the probe safety specifications. Given the very low impedance of the source, any arc will be quite an event.

Accidental mistakes can be (but not limited to) either shorting the probes or inadvertently switching the probes from 10:1 to 1:1 (several regular oscilloscope probes have this tiny switch on them).

Probing a high voltage power supply is different than mains, given the fact that, even if it has a similar impedance as mains, it still has an advantage as spikes may not exist or be well characterized - thus eliminating the element of surprise.

All in all, kudos to you to double-check your procedure on this forum. In a very distant past I measured the voltage of the three-phase 60A bus bars feeders to my house and no surprises happened, but I was ignorant of all these factors at the time and did not take any precaution other than checking the maximum voltages of the 10:1 probes.

[madires]

The question is: is it safe to measure mains 220AC with A-B operation so probes ground clips are not used? (assume the probe is x10)

If someone is experienced with scopes and high voltage measurements, knows the limitations of the poor man's differential probe (A-B), and stays within specs of the probes and the scope, the answer would be "not recommended, but it will work at you own risk". In your case I'd strongly recommend proper differential probes rated for the voltages you want to measure plus a generous safety margin. Don't buy cheap knock-offs, buy a well known brand from a well known distributor. Yes, they are expensive, but your life is more valuable.

[Neukyhm]

I have read all your comments and summarizing, I can measure mains and any other thing with a similar voltage like a three phase motor I have with an inverter board. I have bought a probe in a store here, it says "Max input voltage: 600V CAT I, 300V CAT II (DC+peak AC)", so I think I can't use this probe if I have like 240RMS=340 peak, I will go for a diff. probe then.

[raspberrypi]

Mr Carlsons lab does a great example of how to plug a scope into the mains and how to do it right.

I know in america (not spain) since you have 120VAC you don't know what leg of the 220 your outlet is plugged into, that's why people always say its a 50/50 shot if you blow up your scope.

https://youtu.be/XBsQ3sZ45Fk

[mtdoc]

Nice video but...

I know in america (not spain) since you have 120VAC you don't know what leg of the 220 your outlet is plugged into, that's why people always say its a 50/50 shot if you blow up your scope.

No - The issue has nothing to do with which leg (L1 or L2 in the North American split phase 120/240V system) the particular branch circuit you are plugged into is utilizing.

[Neukyhm]

I know in america (not spain) since you have 120VAC you don't know what leg of the 220 your outlet is plugged into, that's why people always say its a 50/50 shot if you blow up your scope.

I know exactly the cable that is the neutral one, but the thing is, imagine there is a problem and neutral voltage is not 0 volts as it has to be, then I can't connect probe ground to neutral.

[Gyro]

Neutral will normally be NEAR ground, but depending on where they are commoned there will be a voltage between them, caused by Neutral conductor current. This will of course be a very low impedance source. It is not sensible or reasonable to expect a probe ground lead to short this current, regardless of any fault condition or not!

[Neukyhm]

Neutral will normally be NEAR ground, but depending on where they are commoned there will be a voltage between them, caused by Neutral conductor current. This will of course be a very low impedance source. It is not sensible or reasonable to expect a probe ground lead to short this current, regardless of any fault condition or not!

I have like 40 volts between neutral and ground, that's what I'm talking about xD and that's why I want to measure mains, because something is wrong with my electrical installation.

[madires]

A scope is the wrong tool for that kind of measurement. Ask an electrician, he's got the right tool for that.

[tggzzz]

Neutral will normally be NEAR ground, but depending on where they are commoned there will be a voltage between them, caused by Neutral conductor current. This will of course be a very low impedance source. It is not sensible or reasonable to expect a probe ground lead to short this current, regardless of any fault condition or not!

I have like 40 volts between neutral and ground, that's what I'm talking about xD and that's why I want to measure mains, because something is wrong with my electrical installation.

And that, ladies and gentlemen, is an excellent example of why you should determine a potential client's actual problem before discussing their "solution".

First the question was "The question is: is it safe to measure mains 220AC with [a scope]... Again, I know the answer, it is safe.... "
That mutated into "Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains."
And we now find that all he wants to do is find a gross fault in his mains installation. Nothing subtle.

A scope is the wrong tool for that problem.

Given what you have said so far, I strongly recommend you get a properly qualified electrician to examine your installation. Don't risk hurting other people.

[mtdoc]

I agree with what madires and tggzzz have said.  An oscilloscope is the wrong tool and it's best to consult an electrician.

But - if you have a DMM (with proper CAT rating) that has a lo-Z setting you can check whether that 40V is really of any consequence. For example, see this nice Fluke application note on ghost voltages.

On the other hand if your neutral is not properly bonded to a proper earth ground, it may be real..

If you're not absolutely sure you can test this safely - call an electrician.

[Neukyhm]

First the question was "The question is: is it safe to measure mains 220AC with [a scope]... Again, I know the answer, it is safe.... "
That mutated into "Listen, asking if I can measure mains voltage is my way to ask if I can measure hundreds of volts of anything with the oscilloscope (assuming that voltage is below probe & oscilloscope limits). It's not like I feel fascination for mains."
And we now find that all he wants to do is find a gross fault in his mains installation. Nothing subtle.

First the question was "measure mains" because that's what I want to measure wheter my installation is good or not.
That mutated into "measure hundreds of volts" because  I want to measure an inverter three phase motor board output.
And then I said that I won't connect probe ground to neutral because a problem in my installation forbids me to do that.

Please, I have not spent 400 euros to just measure mains.

[tggzzz]

If you're not absolutely sure you can test this safely - call an electrician.

Of course, that doesn't work where the Dunning-Kruger syndrome is present.

[Neukyhm]

If you're not absolutely sure you can test this safely - call an electrician.

Of course, that doesn't work where the Dunning-Kruger syndrome is present.

Do you think that just because I want to learn? Because I'm here for that, to Learn (did you learn or you were born knowing electric engineering?), it's very sad that you laugh at unexperienced users that want to learn, denota mucha falta de humildad.

[DimitriP]

If you're not absolutely sure you can test this safely - call an electrician.

Of course, that doesn't work where the Dunning-Kruger syndrome is present.

Do you think that just because I want to learn? Because I'm here for that, to Learn (did you learn or you were born knowing electric engineering?), it's very sad that you laugh at unexperienced users that want to learn, denota mucha falta de humildad.

A quien no pide consejo, darlo es de necios.

You did ask.

[tggzzz]

If you're not absolutely sure you can test this safely - call an electrician.

Of course, that doesn't work where the Dunning-Kruger syndrome is present.

Do you think that just because I want to learn? Because I'm here for that, to Learn (did you learn or you were born knowing electric engineering?), it's very sad that you laugh at unexperienced users that want to learn, denota mucha falta de humildad.

It was a general statement, not specifically aimed at you.

However, in your original post you did write "...Again, I know the answer, it is safe...". I do appreciate English is probably not your first language, just as Spanish is not mine.

[raspberrypi]

Nice video but...

I know in america (not spain) since you have 120VAC you don't know what leg of the 220 your outlet is plugged into, that's why people always say its a 50/50 shot if you blow up your scope.

No - The issue has nothing to do with which leg (L1 or L2 in the North American split phase 120/240V system) the particular branch circuit you are plugged into is utilizing.

Care to elaborate what the issue is? Usually when you tell someone they are wrong you explain why.

[raspberrypi]

Neutral will normally be NEAR ground, but depending on where they are commoned there will be a voltage between them, caused by Neutral conductor current. This will of course be a very low impedance source. It is not sensible or reasonable to expect a probe ground lead to short this current, regardless of any fault condition or not!

I have like 40 volts between neutral and ground, that's what I'm talking about xD and that's why I want to measure mains, because something is wrong with my electrical installation.

My neutral has 40 volts on it too when I put the meter between it and ground. Is this caused by coupling from the wires all running next to each other? How much current roughly would this 40 volts put out? Could I run a 5W light bulb off it or even an LED? Or would it disappear as soon as you put any load on it.

[Neukyhm]

However, in your original post you did write "...Again, I know the answer, it is safe...".

That's because I know the theory, and the theory says that the voltage I want to measure is below probe and oscilloscope limits.
But practice is other story, you guys are more experienced and that's why I'm asking, because I do know that I can measure that IN THEORY, but in practice, anything I have not take care of could go wrong.

A quien no pide consejo, darlo es de necios.

I don't know what you mean with that, I'm here for any advice I can get. I just don't want anyone to laugh at me (an absolut newbie with oscilloscopes) for this, because what if a new  user willing to learn comes here and asks something elementary like how to turn on a led with a 3V battery? I would be sad if I see people laughing at him for that.

I have no problem to say that I know almost nothing compared to you in this topic.

Aaaand talking again about it, what if I build a voltage divider with a few megaohm resistors?

[Brumby]

With mains power, things get a little more dangerous.  It's not about simple voltages any more, it is about serious energy levels.  10KV around a CRT tube will wake you up in a hurry and perhaps fry some types of component, but 230VAC mains can make things explode.  Add a transient or two and things can get really dicey, really quickly and unexpectedly.  Your very first post indicated you were not really aware of this, as you stated:

Hi, I'm new here, I hope you can help me with this. I already know the answer, so the reason I'm asking for this is because I want to make sure that my house won't burn.

....

The question is: is it safe to measure mains 220AC with A-B operation so probes ground clips are not used? (assume the probe is x10)

This line in particular ...

Again, I know the answer, it is safe. I just want to make sure, you guys are more experienced than me.

... demonstrates you didn't know.  This is the sort of thing that makes a lot of experienced people very anxious for your safety (and why some people bring up the D-K syndrome thing).

To your credit, you did have the presence of mind to bring your question here, before sticking probes into a mains environment.  As a result of that, you have, indeed, learned something - and something that has the very distinct potential to save you from injury (or worse) - without exposing yourself to the real risk.

To be fair to those who may have come across as overcritical, nobody here has any real idea of who you are, your background or your capabilities.  You've only been a member for a couple of days - and the only thread you have posted on is this one.  As time goes by and we have the chance to discuss things with you, we will develop a better understanding - but at this point in time, you are somewhat of an unknown to us.

While there are a lot of things we can help you with, sometimes it is better if advice like the following is given:

If you're not absolutely sure you can test this safely - call an electrician.

Not only does an electrician have the qualification for this sort of work, they also have experience.  Sometimes, it is this experience that can interpret the readings and/or follow a testing path that can identify the fault - because faults in the real world are sometimes non-intuitive, until all the necessary information is at hand (which might take a lot of time, reasoning and measurement) ... or it has been encountered previously.


We are certainly in favour of helping and educating here, but we don't want to encourage wrecklessness - and until we get to know a member a bit better, we are likely to err on the side of caution.

One last word .... there is a wide scope of members here - with different backgrounds, different experience, different cultures - and, of course, different native languages.  It's a global community.  There are bound to be communications issues.

[Brumby]

However, in your original post you did write "...Again, I know the answer, it is safe...".

That's because I know the theory, and the theory says that the voltage I want to measure is below probe and oscilloscope limits.
But practice is other story, you guys are more experienced and that's why I'm asking, because I do know that I can measure that IN THEORY, but in practice, anything I have not take care of could go wrong.

You omitted the words "IN THEORY" with your original post.  If you had included those, then a whole lot of the messy part of the discussion would not have happened.

If you thought we should have understood that, then you need to understand the engineering mind - it works on the basis of what is presented to it.

[Brumby]

Aaaand talking again about it, what if I build a voltage divider with a few megaohm resistors?

That idea is how a high voltage probe is designed - but there are a few more considerations than just simple voltage calculations....

The physical layout of the divider is important, as is the substrate on which it is mounted.  If using a PCB, then there has to be proper consideration given to isolation for each point.  Then there is a case which will need to keep the operator safe when transients are encountered.  As well, you will need to consider the voltage rating of the resistors.  There are probably some other considerations that people with better knowledge on this subject could add.

These issues are not trivial, especially if you don't have the practical experience.  Better to buy a commercial HV probe.

[Ian.M]

Neutral will normally be NEAR ground, but depending on where they are commoned there will be a voltage between them, caused by Neutral conductor current. This will of course be a very low impedance source. It is not sensible or reasonable to expect a probe ground lead to short this current, regardless of any fault condition or not!

I have like 40 volts between neutral and ground, that's what I'm talking about xD and that's why I want to measure mains, because something is wrong with my electrical installation.

My neutral has 40 volts on it too when I put the meter between it and ground. Is this caused by coupling from the wires all running next to each other? How much current roughly would this 40 volts put out? Could I run a 5W light bulb off it or even an LED? Or would it disappear as soon as you put any load on it.

Putting a load of more than  a couple of mA between a 'high' neutral and ground will trip any RCD or GFCI on the circuit, unless the ground wire is broken somewhere, and in that case it will put a dangerous voltage on the exterior metalwork of all grounded equipment plugged into the same circuit.

If in doubt, call in a qualified inspecting electrician.

Assuming you have a CATIII or better rated DMM with fused test leads, if you really want to find out what's going on, you'll need to drive a temporary ground rod into moist ground, and bring an insulated wire from it inside to get a true Earth reference to measure against.  If using thin hookup wire, you'd better put an inline fuse holder with a 250mA fuse on the end of it in case you accidentally touch it to a live terminal or other live metalwork.  Do not assume that any pipework is properly grounded.

By testing between the true earth wire and socket ground you can detect if the ground wiring has a problem.  You can also check the voltage between true earth and neutral.

CAUTION: if there is a fault, lethal voltages may exist between the true earth wire and any supposedly grounded metalwork in the room. Treat the earth wire as-if it is live until you are certain no dangerous voltage differences are present.  Do *NOT* attempt this sort of testing with a CATII or lower rated meter, with unfused test leads or if you are unsure what you are doing.

[mtdoc]

Nice video but...

I know in america (not spain) since you have 120VAC you don't know what leg of the 220 your outlet is plugged into, that's why people always say its a 50/50 shot if you blow up your scope.

No - The issue has nothing to do with which leg (L1 or L2 in the North American split phase 120/240V system) the particular branch circuit you are plugged into is utilizing.

Care to elaborate what the issue is? Usually when you tell someone they are wrong you explain why.

*In North American split phase electrical service, there are 2 "hot" legs at the service entrance (L1 and L2) and one neutral which is bonded to earth ground.

*Both L1 and L2 are 120V with respect to neutral/ground. Each individual 120V branch circuit coming off of the main service panel utilizes either L1 or L2 - but not both. (A 240V circuit utilizes both).

*From the point of view of the end user of these 120V circuits at the home electrical outlets, it makes no difference which leg (L1 or L2) is supplying any particular outlet.  It is only relevant to the electrician wiring the service panel and wanting to balance loads between legs.

HERE is a good, concise reference for this. See figure 1.

*From the point of view of oscilloscope use and safety and understanding how not to "blow up your oscilloscope" and use or miss-use of isolation transformers, the distinction or even knowledge of which leg is supplying an outlet is irrelevant and has nothing to do with any supposed "50/50 chance" of blowing up your oscilloscope.

*Please watch Dave's video ("How not to blow up your oscilloscope") and the video you posted again to better understand why and how measuring mains supplied circuits (or measuring mains itself) with an oscilloscope can result in disaster without great care and understanding.  It's not an easy subject to fully grasp and Dave does a much better job in his video of explaining it that I could do in a brief forum post.

[Elf]

My neutral has 40 volts on it too when I put the meter between it and ground. Is this caused by coupling from the wires all running next to each other? How much current roughly would this 40 volts put out? Could I run a 5W light bulb off it or even an LED? Or would it disappear as soon as you put any load on it.

When wired to code, the neutral is bonded to ground at the panel where utility service enters. Neutral typically rises above ground potential due to the current flowing through the resistance of the wire(s) back to where ground and neutral are bonded. For example if that were 2 ohms back and 10 volts N-G then 5 amps would be returning over the neutral. If the load were removed (0 amps draw), then neutral should be at ground potential. You are essentially just measuring the voltage across the length of the neutral wire from you to the bonding point, via the ground wire (as a long extension of your multimeter probe).

40 volts seems very high for a N-G voltage. You may have bad connections (high resistance) or perhaps it is floating entirely and not bonded. I would get the attention of an electrician.

[tggzzz]

In typo veritas

...but we don't want to encourage wrecklessness...

The rest of your post is spot on, too.

[David Hess]

Putting a load of more than  a couple of mA between a 'high' neutral and ground will trip any RCD or GFCI on the circuit, unless the ground wire is broken somewhere, and in that case it will put a dangerous voltage on the exterior metalwork of all grounded equipment plugged into the same circuit.

Putting more than a couple of mA between neutral or hot and any other point will trip a RCD or GFCI whether ground is connected or not.  The RCD or GFCI does not require ground to work.  The other point could be another phase, neutral, hot, or anything which draws a current including a different ground.  The RCD or GFCI is a two wire device.

[Ian.M]

OTOH if the ground wire is broken you probably wont be able to get enough current to trip the RCD. .You cant get a L-N imbalance if the current's got nowhere to go!

[David Hess]

OTOH if the ground wire is broken you probably wont be able to get enough current to trip the RCD. .You cant get a L-N imbalance if the current's got nowhere to go!

If that is the case, then you also cannot shock yourself except between the line and neutral which a RCD will not help with anyway.  Often the problem involves the sink or tub and for that, the RCD will work just fine.  At least here in the US, areas around water like the kitchen and bathroom are where RCDs are required.

[Neukyhm]

I have finally decided (until I buy better probes) to build a voltage divider with two big 10 Mohm resistors.

By the way, the oscilloscope I bought is the Siglent SDS1102CML+

[tggzzz]

I have finally decided (until I buy better probes) to build a voltage divider with two big 10 Mohm resistors.

So, before you do that you should consider the ways that could fail. And if you don't understand them all, you should reconsider.

Hint: it is unlikely that a beginner will be able to simply think, and reproduce all the hard-won experience gained over the decades. Hells teeth, I doubt I could think of all the ways.

[Neukyhm]

I have finally decided (until I buy better probes) to build a voltage divider with two big 10 Mohm resistors.

it is unlikely that a beginner will be able to simply think, and reproduce all the hard-won experience gained over the decades. Hells teeth, I doubt I could think of all the ways.

That's why I want to do things. As I said, I'm a physicist. In college, I learnt how transformers work but no idea about how to build one, so I had to learn that by myself. I have built some 100KV transformers, and I have blown up some capacitors in the way (it's fun) and I'm proud of it (the capacitors, I mean)

But with the oscilloscope is different, it's my first one. That and the high power of mains have led me to this forum so I can be extra sure that I don't break the oscilloscope or kill myself. Hey, I know what's 340V AC through my body for 10 seconds (it wasn't my fault, the inverter circuit of an automatic door was broken and the RCD too, it didn't jump).

I may have no experience with oscilloscopes but I'm into electric stuff.

PS: guys, you really have to test your RCD once a month at least xD

[Brumby]

Well, you have obviously dabbled a bit - and no doubt understand our concerns ... so I will finish by saying:  Stay safe.



But we are still a bit worried....

[tggzzz]

Hey, I know what's 340V AC through my body for 10 seconds (it wasn't my fault, the inverter circuit of an automatic door was broken and the RCD too, it didn't jump).

There are many types of "RCD" and different installations. However, in the UK I would expect an RCD to trip if you had >30mA through your body to ground. It wouldn't trip if the current was from live to neutral, nor if <30mA.

[Neukyhm]

Hey, I know what's 340V AC through my body for 10 seconds (it wasn't my fault, the inverter circuit of an automatic door was broken and the RCD too, it didn't jump).

There are many types of "RCD" and different installations. However, in the UK I would expect an RCD to trip if you had >30mA through your body to ground. It wouldn't trip if the current was from live to neutral, nor if <30mA.

But the thing is that my RCD was completely broken. I was 11 years old then (now 22) and I almost die. An electrician found the RCD to be dead, it didn't respond to the test button.

Well, you have obviously dabbled a bit - and no doubt understand our concerns ... so I will finish by saying:  Stay safe.

But we are still a bit worried....

I appreciate that, you guys are worrying about somebody who you don't even know.

[StillTrying]

By the way, the oscilloscope I bought is the Siglent SDS1102CML+

Why did you choose a CML+ ?
Did you find an online review anywhere ?

[Neukyhm]

By the way, the oscilloscope I bought is the Siglent SDS1102CML+

Why did you choose a CML+ ?
Did you find an online review anywhere ?

Is it a bad oscilloscope?

[StillTrying]

Is it a bad oscilloscope?

The jury (of about 1) is still out.  https://www.eevblog.com/forum/testgear/siglent-cml-firmware/

[tautech]

Is it a bad oscilloscope?

The jury's still out.  https://www.eevblog.com/forum/testgear/siglent-cml-firmware/

Yours is ~3rd that I know of that has had a problem, one of which I sold. 
The Plus version does seem to have some issues but I am very confident Siglent will fix them in short time. They must....there are many of this model in the marketplace.

SDS1102CML+ is my best seller.

[StillTrying]

Yours is ~3rd that I know of that has had a problem, one of which I sold. 

After I'd gone away, I was trying to remember which thread I'd just posted that in because I remembered that I'd forgotten to say that currently the CML+ jury consists of only about 1. 

The Plus version does seem to have some issues but I am very confident Siglent will fix them in short time. They must....there are many of this model in the marketplace.
SDS1102CML+ is my best seller.

In normal modes, being able to trigger on parts of a waveform 1mV or less high is 20 times better than my aging CRO, so I won't be throwing the CML+ away just yet!

[raspberrypi]

I think this is where I got the 50/50 idea from:

Never remove the Neutral (White wire) from the Neutral t
erminal Bar in the Service Entrance panel /
load Center / Distribution panel if the phase conductors are energized — it could be part of a Multi-
wire Branch Circuit, and removing it could damage the connected electrical loads / cause a fire.
removal of the Neutral wire will amount to the two 120 VAC loads being connected in series across
240 VAC. If the impedances of the two 120 VAC loads are not equal (have different Watt ratings), the
load with the higher impedance (higher watt rating) will see a dangerous over voltage condition >
120 VAC and may reach 240 VAC in case the other load is shorted (please see example in Fig 2 on
page 3).

[Neukyhm]

I have finally tested the oscilloscope, brand new and calibrated in factory last month, so I suppose it has the latest firmware. I have also measure mains with a 2x 10Mohm resistors and I'm still alive!

Just one thing, I have tested the "-" math function, it displays with a white cursor the result but nothing more, I mean, where is the voltage measured? I don't know how to make it display for example the Vrms of the subtract operation.

[tautech]

I have finally tested the oscilloscope, brand new and calibrated in factory last month, so I suppose it has the latest firmware.

In Utility/System info you can check the firmware version is the same as this one:
http://siglentamerica.com/gjjrj-xq.aspx?id=5074&tid=15

Just one thing, I have tested the "-" math function, it displays with a white cursor the result but nothing more, I mean, where is the voltage measured? I don't know how to make it display for example the Vrms of the subtract operation.

Spend some more time to become more familiar with the scope and time reading the manual on the CD.
IIRC you can turn on something in the Measurements menu to get it up on the OSD.
If I get a chance later I'll get one out and have a look.

[StillTrying]

Just one thing, I have tested the "-" math function, it displays with a white cursor the result but nothing more, I mean, where is the voltage measured? I don't know how to make it display for example the Vrms of the subtract operation.

Congratulations, it's your first bug find!
Ignore me, I've got all the math functions working somehow!

[Neukyhm]

Just one thing, I have tested the "-" math function, it displays with a white cursor the result but nothing more, I mean, where is the voltage measured? I don't know how to make it display for example the Vrms of the subtract operation.

Congratulations, it's your first bug find!  I'll add it to the other thread soon, unless you want to.

You can subtract two waveforms by using '+' and having one channel inverted, but these math functions are very rough.

With "white cursor" I ment that I can see the result waveform in white, but I can't or don't know how to display information about that waveform.

In Utility/System info you can check the firmware version is the same as this one:
http://siglentamerica.com/gjjrj-xq.aspx?id=5074&tid=15

I confirm it's the latest firmware.

[StillTrying]

With "white cursor" I ment that I can see the result waveform in white, but I can't or don't know how to display information about that waveform.

I think you can only go by the grid/graticule divisions, (no math auto measurements).
Perhaps your math waveform is hiding the "Math scale:" bit. (Edit: It is a daft place for them to have put it.)

The waveforms are just my finger touching the end of the probe tips - both connected together.

[Neukyhm]

Exactly the same, I can see the white waveform but not any information about it.

[tautech]

Exactly the same, I can see the white waveform but not any information about it.

I'll have a fiddle with a unit and a real simple setup, say 2 different DC voltages and various Math functions with a view to reporting your findings (when confirmed) to the factory.
Probably tonight as I have a heap on today. Factory closes for the weekend at 10pm my time.

[tautech]

Simple Math test.

Both channels 1V/div and and all Ref levels (0V) set to -2V, Ch1, Ch2 and Math.
Math scaling set to the same as active channels: 1V/div.
Supplied voltages:
Ch1 5VDC, Ch2 2VDC.

Math Ch1 - Ch2.



The result for Math is as expected: 3 div above Ref so 3 V.
The highlighted function button allows the Multifunction control to adjust the Math scaling (V/div).
The uppermost function button allows the Multifunction control to adjust the Math trace Reference level.

There is no Math OSD measurements in the UI.

Should we ask Siglent if this can be implemented ?

[Brumby]

Congratulations on getting your scope up and running - but as has been said above, I would also suggest you spend some time doing that weird thing - RTFM  (Read The Flippin' Manual).


As for this:

I have also measure mains with a 2x 10Mohm resistors and I'm still alive!

I had little doubt you would survive.  The chances of something going wrong are low - but they're not zero.

Just do me this one little favour ... imagine what would happen if 5kV or 10kV were to hit your setup.  Then, take steps to contain the aftermath ... and lastly, whenever you use it, expect that to happen.

You may never have the event - but it is better to be prepared than have to deal with the consequences.


I know I'm going on about this a bit - but you can't count on getting second chances.  (I'll step away from the soapbox now.)

[tggzzz]

You may never have the event - but it is better to be prepared than have to deal with the consequences.

Just so. Engineers (i.e. not hackers or amateurs) consider how things fail to be as important as how they work.

We'll see if the OP appreciates that, but so far the evidence is against it.

[Neukyhm]

Should we ask Siglent if this can be implemented ?

That would be very nice. As I said, I thought that I would see the option to measure whatever math function you are using in "Measure" menu, I expected to find that option just where you select to measure CH1 or CH2, I mean, I expected it to also measure "Math".