measuring mains with oscilloscope

[nctnico]

here is what I do, when measuring stuff on a live mains:
power the scope via a special low capacity isolation transformator.
it was easy to make:
http://webx.dk/oz2cpu/230V-trafo-isolation-low-capacity/230V-trafo.htm

Don't try this at home  Get a job in a super market or flipping burgers and make money during the time spend creating dangerous contraptions so you can buy safe tools. Differential probes and current probes are very cheap nowadays. It doesn't pay to mess around.

[joeqsmith]

You could roll your own differential probe but I suspect it would cost more than just buying one.   I built a couple for the fun of it but I suspect the purchased ones would out perform them..
   

[oz2cpu]

tggzzz ??
it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

If the 1 ohm sense resistor brakes open, the scope prope will get full mains voltage, in 10:1 it can handle this too

[Simon]

The siplest way to do it is to remove the earth clips for the probes. set the scope to math subtraction mode and and use the two probes as "differential" probes. You still have to stay withing the range of the scopes inputs relative to ground but this way you can make sure the chasis is grounded and not worry about what you do with each probe.

[alsetalokin4017]

There are two questions I've noticed that frequently are asked by new oscilloscope owners and users. They are : "How can I measure the mains with my (grounded) oscilloscope?" and "why does my (improperly grounded) oscilloscope keep showing me 60Hz (or 50, whatever) noise from the mains?"
 

[Terry01]

It seems there is a couple of ways to measure the mains safely for beginners and a couple of ways to measure the mains using homebrew stuff that you should only use if you really know and understand what you are doing.

I fall under the "beginner using 2 probes" category but it is also good to learn the "other" ways too. It's always good to learn new stuff.

It's great how the guys who know better can teach the beginners new things on the forum and keep them from blowing themselves and/or their equipment up. 

 

[alsetalokin4017]

Most of the time my problem is the latter: I _don't_ want mains noise contaminating the signal I'm trying to measure. Ymmv of course!
(The "line" trigger option can frequently help one to identify weirdnesses caused by mains pickup in a higher frequency signal.)

But they say experience is the best teacher. Blow up one expensive scope by sticking things into places they don't belong, and you will probably never do it again.
 

[tggzzz]

tggzzz ??
it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

If the 1 ohm sense resistor brakes open, the scope prope will get full mains voltage, in 10:1 it can handle this too

Think about all the equipment you are connecting together, not just one bit. Include yourself in "the equipment".

The failure modes can be subtle and surprising, and can catch out even professionals that take precautions. For one real life fatality, see https://www.eevblog.com/forum/testgear/ground-of-oscilloscope-always-connected-to-earth/msg1967150/#msg1967150

[BravoV]

it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

If what you claimed is true, so this millions of dollars people poured in the measurement business either the manufacturers or the users are stupid and just wasting their time and money all this time ? 

[tggzzz]

it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

If what you claimed is true, so this millions of dollars people poured in the measurement business either the manufacturers or the users are stupid and just wasting their time and money all this time ? 

You can try to point people in the right direction, but the Dunning-Kruger effect is strong.

That's why I will never try to remotely indicate single specific problems in topics where there are many subtle dangers.

[Fungus]

tggzzz ??
it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

 

If you meant it's just as dangerous as a battery-operated 'scope then I agree 100%

ie. The danger isn't from the 'scope's power supply, it's from using a little crocodile clip to connect mains A/C to the chassis of a device with exposed metal all over the place.

If you can't understand where the danger comes from then my only advice is to wear rubber boots and gloves while you work. And don't sit on a chair.

[tggzzz]

tggzzz ??
it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

 

If you meant it's just as dangerous as a battery-operated 'scope then I agree 100%

ie. The danger isn't from the 'scope's power supply, it's from using a little crocodile clip to connect mains A/C to the chassis of a device with exposed metal all over the place.

If you can't understand where the danger comes from then my only advice is to wear rubber boots and gloves while you work. And don't sit on a chair.

And work with one hand in your pocket, tied there if necessary

[nctnico]

tggzzz ??
it is NOT going to fail, the isolation transformer I use is well made and with ultra low capacity
so there is no problems at all with the test setup i explain, it is just as good as using a battery opperated scope,

No it isn't. Even battery powered scopes come in two categories: expensive ones with isolated inputs (usually these have a plastic, non-conductive BNC connectors) and the cheaper ones which have a common metal chassis and metal BNCs. The latter are just as dangerous to use compared to floating a scope using an isolation transformer.

[n3mmr]

Regarding the proper way to measure inside the mains side of a power supply, i e a high voltage diff probe, which ones are actually good enough of the diff probes offered on Ebay?

I see a Micsig probe set for 5000V at about 180 USD, https://rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.de%2Fulk%2Fitm%2F113749875769 ,
and used testtec 9001 probes.
Are these any good? Dare one trust them?

[shakalnokturn]

I'm always astonished by how often this same question comes-up and even more by some of the answers given.

Although I have overlooked the problems of safe probing on line referenced equipment myself in the past, it is a really important point to understand.

Anyone wanting to work on live equipment should go to the trouble of first understanding how electricity is delivered to wherever they are using it, the role of each wire used up to the socket, the existing protection equipment and it's condition if something does go wrong.
Beyond the financial aspect of a failure the question you want to ask when working in your lab or garage is: "If I touch this how much current will I be taking?" When "this" is your oscilloscope Ext. Trig. BNC or handle the question should become worrying.

If you aren't worried at that point you may as well take the reasoning one step further and double wrist-strap yourself with a nice low impedance to the local ground, it has potential to roast your fingers off, but may at least save your chest.

There are well known good solutions for probing on 220V, they do have a low cost compared to an oscilloscope or funeral.
Is it really worth putting your relatives to all that bother for a bloody 9$99 power supply that should have been better designed in the first place?

[vk6zgo]

For Pete's sake, everybody, buy a bloody differential probe, & those of us with enough knowledge to do it safely in other ways can just quietly die out, without running into hundreds of iterations of this thread  everytime we log on throughout our declining years!

[Fungus]

I'd like to know what information people hope to gain by poking at mains with an oscilloscope as opposed to, eg., plugging in a lamp.

For Pete's sake, everybody, buy a bloody differential probe, & those of us with enough knowledge to do it safely in other ways can just quietly die out, without running into hundreds of iterations of this thread  everytime we log on throughout our declining years!

You don't even need that for a quick test. Just pop the ground clips off your probes and set the 'scope to "AC" mode.

(Use two probes and "difference" as necessary).

[Simon]

I'd like to know what information people hope to gain by poking at mains with an oscilloscope as opposed to, eg., plugging in a lamp.

For Pete's sake, everybody, buy a bloody differential probe, & those of us with enough knowledge to do it safely in other ways can just quietly die out, without running into hundreds of iterations of this thread  everytime we log on throughout our declining years!

You don't even need that for a quick test. Just pop the ground clips off your probes and set the 'scope to "AC" mode.

(Use two probes and "difference" as necessary).

Yep I said that way back but it's not sexy or dangerous enough.

[maginnovision]

I logged a week on an oscilloscope the mains voltage at my house. Captured large voltage dips that likely fried my refrigerator compressor and my old AC compressor. I showed the logs to SCE who paid for those failures after finding a loose connection on the pole I get power from. I then logged another week to make sure it didn't happen again.

You can't do that plugging in a lamp.

[tautech]

I logged a week on an oscilloscope the mains voltage at my house. Captured large voltage dips that likely fried my refrigerator compressor and my old AC compressor. I showed the logs to SCE who paid for those failures after finding a loose connection on the pole I get power from. I then logged another week to make sure it didn't happen again.

You can't do that plugging in a lamp.

Exactly !

Connecting a scope to mains is not for the inexperienced but it can safely be done.
Wise to use 100x probes or a differential probe for safety reasons though.

[tggzzz]

I logged a week on an oscilloscope the mains voltage at my house. Captured large voltage dips that likely fried my refrigerator compressor and my old AC compressor. I showed the logs to SCE who paid for those failures after finding a loose connection on the pole I get power from. I then logged another week to make sure it didn't happen again.

You can't do that plugging in a lamp.

Just so, but there are simple easy cheap and safer ways to achieve that than connecting a scope to the mains, e.g.:

• use an ac voltmeter that can capture and upload the values to a PC
• connect a standard 240V/12V (etc) transformer to the mains, and connect the scope to the 12V output

With a little thought and imagination, it is usually possible to avoid the potential dangers of connecting a scope directly to the mains.

[joeqsmith]

I'd like to know what information people hope to gain by poking at mains with an oscilloscope as opposed to, eg., plugging in a lamp.

Normally, I am looking for some sort of transient condition.  The scope isn't really a tool I would use for that.

[BravoV]

I did use my HV diff. probe + scope, in measuring AC mains voltage & current (thru shunt) for capturing, and tweaking the soft start circuits that built for AC inductive load, like my bench drill, variac and other AC rotary cutting tools and etc.

As these inductive loads can have nasty current spike (sometimes 10 to 50 times at rated current) at few 1st cycles when powered on at the right/wrong AC phase, and also caused serious voltage sag or trip the current breaker.

In this specific case, just can't use slow AC logger like DMM (own Fluke 287) as its not fast enough to capture that events.

[Fungus]

You can't do that plugging in a lamp.

I could point a camera at a lamp much more easily than I could record a week's worth of waveforms on my 'scope.

As these inductive loads can have nasty current spike (sometimes 10 to 50 times at rated current) at few 1st cycles when powered on at the right/wrong AC phase, and also caused serious voltage sag or trip the current breaker.

ie. A light bulb would flicker if it's connected to a nearby socket.

[Fungus]

Connecting a scope to mains is not for the inexperienced but it can safely be done.
Wise to use 100x probes or a differential probe for safety reasons though.

I never said it can't be done, I even said how I'd do it a couple of posts up, ie. remove ground clips from probes (to avoid temptation), use probes.

The issue at stake is floating oscilloscopes so you can attach them to mains wires without blowing your mains fuses (as if that should be the thing that's worrying everybody).