Measuring Mains with Rigol DS1052E

[logicnibble]

Hi!
It's dangerous to measure mains voltage directly with an oscilloscope but since I've read many guys have done it, I'd like to know about the Rigol DS1052E and if there are any tips & tricks, like:

-Not connecting Rigol earth ground.
-Using diferential reading between channel 1 and 2 and not using the common aligator from both probes.

Thanks

[mikeselectricstuff]

If you have to ask this you probably shouldn't be doing it at all.
Lifting the scope ground is a very BAD idea. Diff probes often don't have sufficient range and/or common-mode rejection.
The normal way to do it is power the device under test from an isolating transformer so you can then ground the circuit ground (seperate ground- don't rely just on the scope probe ground).
Make sure your probes are suitably rated- you'd normally want a x100 probe for mains use. 100 probes are also handy to have for probing high impedance circuits like 32K watch crystals.

[Kiriakos-GR]

I think that logicnibble had set one interesting question.

And no it does not limited to the Rigol product range.

As a electrician I would like to have on screen other than volts and the wave form too.
So to be able to see and tell how perfect or clean is the AC wave at 220 or even at 380V.
Or even, to be able to check DC/AC inverters and see how they respond against a specific load.

Some people they do complains that the low quality DC/AC inverters cause noisy operation on small motors, usually small pumps.
I like to be able to see of what is happening with the wave forms.

If there is a tool capable to help me out there, I need to know which one is capable for the task.


   

[ejeffrey]

What are you trying to measure?  You can measure phase-to-ground and neutral-to-ground with an ordinary high voltage probe.  For phase-to-phase or phase-to-neutral measurements, hooking up two probes and doing math A-B on the scope has terrible common mode rejection, but it might be OK at 50 Hz.  Simply doing separate measurements of phase-ground and neutral-ground may be the most illuminating -- the neutral-to-ground voltage is indicative of the current.  Not enough for a true measurement, but if you turn on a reactive load you will probably see an out-of-phase voltage pop up on the neutral line.

If you just want to see the AC waveform, see if you can find a small AC output wall transformer with <24VAC output and measure the secondary voltage.  High harmonics will be filtered by the transformer, but if you purpose is to look for garbage on the mains caused by other devices you can still probably see plenty.

[logicnibble]

What are you trying to measure?

I'm trying to measure a uC circuit which is powered by a capacitive (transformerless) power supply.
This means all circuit is LIVE.
I also would like to measure a dimming circuit, which is LIVE too.

...an ordinary high voltage probe.

Does Rigol DS1052E have one?

[mikeselectricstuff]

What are you trying to measure?

I'm trying to measure a uC circuit which is powered by a capacitive (transformerless) power supply.
This means all circuit is LIVE.
I also would like to measure a dimming circuit, which is LIVE too.

In which case use a mains isolating transformer, and connect the circuit's 0v rail to mains earth so you can ground the scope to it.

[IanB]

In which case use a mains isolating transformer, and connect the circuit's 0v rail to mains earth so you can ground the scope to it.

Please humour me here as an electronics noob...but if the transformerless device already had it's 0 V rail connected to the chassis and the chassis was connected to mains earth, what might go wrong with connecting a high voltage scope probe under those circumstances?

I understand that if the low voltage supply was insulated from the chassis and completely floating then you would need to create a safe common ground for the circuit and the scope, and then I see why you would need the mains isolating transformer to do that. Do I have it right?

[Kiriakos-GR]

Well I believe too, that with a mains isolating transformer, you do not need any connection with the ground.

 

[jahonen]

For measuring normal "low-voltage" mains safely with an ordinary grounded scope:



Regards,
Janne

[IanB]

Well I believe too, that with a mains isolating transformer, you do not need any connection with the ground.

Well ground has two purposes: one as a common voltage reference, and the other for safety to ensure a low impedance return path for fault currents.

The question is whether the oscilloscope probe ground is connected to the mains earth or whether it is isolated and floating? I don't know much about oscilloscopes, but I suspect the scope ground must be connected to mains earth otherwise the probe would pick up all sorts of noise by inductive and capacitive coupling (much like a DMM generates random voltages on the display when the probes are not connected). If the probe ground is earthed, then connecting it to a point in a circuit that is not at earth potential is going to do bad things. The way to cure this problem is to ensure the circuit under test is completely isolated and floating, so that no current can flow between the circuit and the oscilloscope.

[ciccio]

Please humour me here as an electronics noob...but if the transformerless device already had it's 0 V rail connected to the chassis and the chassis was connected to mains earth, what might go wrong with connecting a high voltage scope probe under those circumstances?

The transformerless device cannot have an external chassis connected to mains earth: it's ground point is connected to mains live or mains neutral (if the plug is reversible),so it will be classified as a "dangerous to touch terminal".
A transformerless device  must be built inside an insulating case (old TV sets were made this way).
The headphone socket of my old TV set is driven by an insulation transformer.
The antenna connector is insulated through low value, high voltage capacitors (that limit the current that can be sourced from the antenna socket).

[IanB]

The transformerless device cannot have an external chassis connected to mains earth: it's ground point is connected to mains live or mains neutral (if the plug is reversible),so it will be classified as a "dangerous to touch terminal".
A transformerless device  must be built inside an insulating case (old TV sets were made this way).
The headphone socket of my old TV set is driven by an insulation transformer.
The antenna connector is insulated through low value, high voltage capacitors (that limit the current that can be sourced from the antenna socket).

That makes sense. But in the UK (where I'm from) reversible two pin plugs are found only with electric shavers. Everything else has a three pin fused plug with an earth connection. Devices can either be "double insulated", which means there must be no potentially live parts that can be touched, or the case must be securely connected to the earth wire in the mains plug. In theory (I think) you can connect the mains neutral to the mains earth inside the device since the two wires should be at the same potential. If someone reversed live and neutral in the mains lead then a short circuit current would flow and blow the fuse in the plug or trip the circuit breaker if someone bypassed the fuse. This idea of connecting earth to neutral inside a device is the bit where I may be off base I think...

[ciccio]

Are you sure that neutral on your socket is at ground potential?
Measure it, please.
Here in Italy neutral is grounded only at the electric company's medium voltage to low voltage transformer,
The one serving my home is about 500 meter from me, and in this moment  neutral is about 30 V respect to ground...
An old electrician told me: live is hot, and neutral is hot..

[Bloch]

Here in Italy neutral is grounded only at the electric company's medium voltage to low voltage transformer,
The one serving my home is about 500 meter from me, and in this moment  neutral is about 30 V respect to ground...

One thing to note is that every contry have there own mains standard 

Here in Denmark it is "normal" at every house to have you own ground point (a 2 to 6 meter long plug in the ground)

If i was you I would talk to the lokal electrician to check if you own ground point is ok. (My guess is not)

I dont have a englich page but try to see here http://translate.google.com/translate?js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&sl=da&tl=en&u=http%3A%2F%2Fcubus-adsl.dk%2Felteknik%2Finstallationer%2Fnulling.php

[IanB]

Are you sure that neutral on your socket is at ground potential?
Measure it, please.
Here in Italy neutral is grounded only at the electric company's medium voltage to low voltage transformer,
The one serving my home is about 500 meter from me, and in this moment  neutral is about 30 V respect to ground...
An old electrician told me: live is hot, and neutral is hot..

I'm not in the UK right now so I cannot measure, but I know there could be a potential difference between earth and neutral wires. In the UK the earth and neutral are either connected at the nearest substation as you describe, or are connected just before entering the home (in more modern systems). Either way, if there is 30 V between neutral and earth, it suggests there is some significant voltage drop where it ought not to be; for instance there might be an unexpected current flowing in the earth conductor?

I suppose though that I am wrong to think you could connect earth and neutral inside a device; most likely you should not do so and I imagine I would find that out if I could locate the right regulations to read...

[Kiriakos-GR]

Are you sure that neutral on your socket is at ground potential?
Measure it, please.
Here in Italy neutral is grounded only at the electric company's medium voltage to low voltage transformer,
The one serving my home is about 500 meter from me, and in this moment  neutral is about 30 V respect to ground...
An old electrician told me: live is hot, and neutral is hot..

Yes and No.

The neutral can show voltage , when and only if there is extreme loads on the phase.
When the load drops, the neutral voltage gets lower too.
The voltage on the neutral commonly called as " returning currents ".
It is part of the energy that phase has, that it does not consumed by the electrical device,
and so one part of it returns on the neutral wire. 

The major problem that we have in the industry is that the loads are so large ones, and the neutral gets a significant amount of voltage.
You have no choice other than to ground it  when ever is possible.
But if you do so, there is no way to install protection relays afterward s. 

[alm]

Either way, if there is 30 V between neutral and earth, it suggests there is some significant voltage drop where it ought not to be; for instance there might be an unexpected current flowing in the earth conductor?

How about current flowing through the neutral conductor? V=IR, so if there is a lot of current and some resistance, there will be a voltage drop. 30V sounds like a lot though.

I suppose though that I am wrong to think you could connect earth and neutral inside a device; most likely you should not do so and I imagine I would find that out if I could locate the right regulations to read...

The only point were they may be connected is at the distribution level (may be in your own home, may be somewhere far away depending on the topology). Connecting them somewhere else may result in fireworks and will likely trigger the GFCI. They may also not be connected at all, the ground may also be in the middle between neutral and ground. A break in the neutral wire somewhere may also change the voltage on the neutral wire, since the equipment connected to the phases now forms a voltage divider, so a large unbalance between the phases will cause neutral to be far away from ground.

Equipment with the chassis connected to neutral has not been allowed for a long time, for good reasons. I'm not sure about the UK, but in many countries live and neutral are often wired incorrectly. Reversing ground and neutral is also somewhat common, but that gets noticed fairly quickly because it will trip the GFCI. Reversing live and neutral can remain undetected for a long time since most equipment developed for an international market doesn't care. Most safety regulations I'm aware of treat live and neutral exactly the same, both require the same clearance, creapage and isolation.

If you're not familiar with this, I think building something connecting directly to mains is a dangerous plan.

[Alex]

I think that logicnibble had set one interesting question.

And no it does not limited to the Rigol product range.

As a electrician I would like to have on screen other than volts and the wave form too.
So to be able to see and tell how perfect or clean is the AC wave at 220 or even at 380V.
Or even, to be able to check DC/AC inverters and see how they respond against a specific load.

Some people they do complains that the low quality DC/AC inverters cause noisy operation on small motors, usually small pumps.
I like to be able to see of what is happening with the wave forms.

If there is a tool capable to help me out there, I need to know which one is capable for the task.


 

Crest factor (http://en.wikipedia.org/wiki/Crest_factor) might be a useful indicator for what you are describing. Measurement of crest factor usually comes as a measurement in DMMs, oscilloscope and scopemeters. Of course, having a proper display of the waveform gives more insight, but a DMM with crest factor measurment should be able to tell you if you need to get the scope out. Just a thought.

[IanB]

How about current flowing through the neutral conductor? V=IR, so if there is a lot of current and some resistance, there will be a voltage drop. 30V sounds like a lot though.

I thought it seemed like a lot. A 30 V drop on a main conductor would seem to me something like a brown out.

If you're not familiar with this, I think building something connecting directly to mains is a dangerous plan.

Fair enough. I don't do electricity for my day job and I also don't like to just follow rules by rote without understanding them. But once I followed through the reasoning as you explained above I had no problem seeing the picture. My incorrect thinking was not something that was going to follow through into a real project without me conducting a proper investigation. Any exposed metal parts should be earthed and current carrying conductors should be properly insulated and should be connected only to the parts of the circuit that are meant to be energized.

[PetrosA]

There have been some good (and some not-so-good...) opinions on the 30V to ground measurement. In most modern systems where the neutral is grounded at or near the main disconnect or last transformer feeding a service, 30V could likely indicate one of a number of possible serious problems.

1 - there is a bad (high impedance) neutral connection between your outlet and the neutral-earth bond
2 - you are connected to one of three phases (typical in EU) and there is a large imbalance between the phases
3 - you are connected to one of three phases and there are a lot of non-linear loads on the transformer causing an overload on the neutral (this is actually more likely now that the EU is forcing the use of compact fluorescent bulbs).
4 - the neutral from the transformer is undersized and overheating
5 - the ground is actually disconnected and you are measuring induced voltage on the ground wire to the properly grounded neutral

There are other possibilities, but these would be the main culprits.

If your service is protected by an RCD then you should be relatively safe. I would still have it checked by an electrician though since any of the neutral problems could potentially create a fire or shock hazard.

If there are large voltages like this between ground and neutral, please be VERY careful testing it because there's the possibility that connecting the two will shunt a large load (10s of amps) through whatever is jumpering them.

[ciccio]

Thanks PetrosA for your response:
I've verified this years ago with the Electric Power Company's technicians, after one of my many calls for service.
-The neutral to hearth bond is at more than 400 meters from my house (in the company's big MV to LV transformer). Lots of new building between.
-I use a single phase of a 3-phase network (the other 2 are used only occasionally by the condo's  common system (stair and external  lights, etc) and most of the time they are  unloaded. The other buildings may be connected to the same phase than mine, or to  one of the others. Nobody knows...
-The ground is connected to hearth  at my house, with 3  earthed poles  (verified with a Megger type instrument) so my RCDs work reliably.
-The neutral is connected to ground at the transformer.
-The neutral (and also the live) ables that supply my house are actually undersized, but the power dropped is acceptable to the company's standards.
The voltage between neutral and ground varies with the day of the week (different loads), with the hour of the day and with the season (summer: more air conditioners: more current draw).
Early this morning I has less than 5 volts, but in no case, even at it's maximum,  it could light a neon voltage tester..... If this happened,  the Power Company would surely intervene (or at least I was told so by one of the technicians)

This situation is "normal": when I was younger and was a touring "sound  engineer's helper", we had similar problems with power supplied to our stage's systems, and we never trusted both live and neutral.

[IanB]

It reminds me of a time many years ago when I was delivering and setting up some equipment on an industrial site on the other side of the world. We had some low voltage power supplies with earthed metal cages and when we happened to rest them on metal strips under the floor there were sparks flying. We measured voltages and the "earth" wire was at a potential of about 100 V. As I recall the situation, someone had wired up the supply wrongly at the distribution panel and at least two of the live, neutral and earth wires were swapped.

[vk6zgo]

Another way of looking at this is to say:-

"One side of my Oscilloscope goes to the mains earth,& the neutral line is ultimately connected to the earth,so I can get a fair idea of the 50/60 Hz signal by not connecting the probe earth to anything,or by connecting the probe earth to the earthed case of the device,& just probing the mains active lead."

Your Oscilloscope safety earth still protects you in this case,plus you can determine if your neutral & active connections are transposed.

The best way is, of course to use an isolation transformer so you can get a better idea of the mains waveform,as some HF stuff may be lost using the method referred to above.
In many cases,all you need is a rough idea,though.

On another point,I'm not sure why everybody freaks out about measuring mains voltages.
OK,the mains has a very low internal resistance/impedance,& can source huge amounts of current,but the
peak voltage of say,a 240v supply is less than 400v.

You can definitely look at voltages of this level with a Tektronix, HP,Siemens,Philips or BWD x10 probe!
Maybe the "El Cheapo" probes aren't as well rated!

VK6ZGO

[logicnibble]

So, the best way to make those measurements is to use an isolation transformer. Got it.

But if I had to measure it directly, can I use the Rigol DS1052E not connecting its mains plug earth connection?

If the oscilloscope is battery powered, I should no problem, right?

[Kiriakos-GR]

On another point,I'm not sure why everybody freaks out about measuring mains voltages.

I will agree that fear its not a good adviser, you are less concentrated and you can cause a mess due your own panic.
Please exclude the electricians from your list,  we love Mains ... LOL