Author Topic: how to measure 3 phase with oscilloscope  (Read 22608 times)

0 Members and 1 Guest are viewing this topic.

Offline karottoTopic starter

  • Newbie
  • Posts: 6
how to measure 3 phase with oscilloscope
« on: September 08, 2013, 05:21:28 am »
I am so confused. I want to measure the 3 phases output of an alternator on a 3 channel oscilloscope. What do I do with the ground wire of each scope? The 3 phase windings are not connected to any ground and there are only 3 wires (I removed rectifier and regulator. And while I am at it, what is the ground clip for anyway? I thought oscilloscopes are identical to DMMs except for the time factor. If that's true (measuring voltage) why is the ground clip so small and not equally prominent as the + lead (probe) is? Is it true that scopes measure voltage thus the ground lead always has to have a closed circuit with the probe otherwise no power flows though the scope? thanks much
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #1 on: September 08, 2013, 05:42:11 am »
Yes, scopes measure voltage and the ground lead is the other connection that completes the circuit with the probe. But scopes don't have the same kind of input isolation that DMMs have, so you need to be careful about the maximum voltages you feed into them to make sure you don't harm the scope.

The ground wires of all scope channels (on a bench scope) are connected electrically to mains ground, so for low frequency measurements you only need to connect one of them. And you must not connect the ground clip to anything that would make a short with mains voltages, or there will be a bang.

If your alternator only has three phase outputs and no neutral connection you may have to make one. You could get three large high value resistors and connect them in a star to the three alternator outputs to make an artificial neutral reference. Then you could connect the scope ground to the center of the star and measure each of the phase voltages relative to this point.
« Last Edit: September 08, 2013, 05:44:16 am by IanB »
 

Offline Kremmen

  • Super Contributor
  • ***
  • Posts: 1289
  • Country: fi
Re: how to measure 3 phase with oscilloscope
« Reply #2 on: September 08, 2013, 06:47:27 am »
Artificial ground can work but you will need to be careful about asymmetric loads. Those will skew the zero point giving you biased values for the phase voltages.
If you are only interested in the magnitudes of the voltage (after all, you know the phase shift is going to be 120 degrees) why not measure them one by one, cyclically assigning one leg to ground and one to live? That way you avoid all problems related to missing ground reference.
Nothing sings like a kilovolt.
Dr W. Bishop
 

Offline peter.mitchell

  • Super Contributor
  • ***
  • Posts: 1567
  • Country: au
Re: how to measure 3 phase with oscilloscope
« Reply #3 on: September 08, 2013, 07:38:42 am »
an easy-ish way is getting 3 small mains transformers, from plug packs or whatnot, putting them across the windings, then tieing one leg of each transformer together and using that as ground/neutral, and the other wire of each transformer as active

isolation and ground in one somewhat simple setup.
 

Offline G7PSK

  • Super Contributor
  • ***
  • Posts: 3861
  • Country: gb
  • It is hot until proved not.
Re: how to measure 3 phase with oscilloscope
« Reply #4 on: September 08, 2013, 08:36:27 am »
Is the alternator star or delta connection. If its star there will be a neutral inside the terminal box even if its delta it might well be reconfigurable to star most of the small to medium machines from reputable manufacturers will be made as six or 12 wire units so that they can be reconfigured as star, delta 3 phase or even single phase, or are you talking about an automotive alternator.
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 7377
  • Country: nl
  • Current job: ATEX product design
Re: how to measure 3 phase with oscilloscope
« Reply #5 on: September 08, 2013, 10:45:11 am »
I wouldn't connect the ground leads to anything in power electronics, or anything, which is capable of making 10+ Amp. You are only measuring 50Hz/60Hz, the grounding of the system is probably sufficient. Use averaging to get rid of the noise, if possible. And if you are planning to take a lot of these measurements, buy differential probes for the scope, and an isolation transformer, and be careful.
 

Offline karottoTopic starter

  • Newbie
  • Posts: 6
Re: how to measure 3 phase with oscilloscope
« Reply #6 on: September 08, 2013, 04:58:04 pm »
Thank you all for your replies. I still don't quite get it. How is it possible to see his signal if there is an open circuit? In basic electronics I was taught that no electricity can flow in an open circuit. So how come I see a signal although I don't have the ground clip connected to complete the circuit?

As far as the alternator goes, I lied. I actually did open the alternator which is a star connection by the way, and soldered a wire on to where all three windings meet (neutral). And yes, when I hooked up my oscilloscope to the three windings and the ground clips to the "neutral" I did get a beautiful waveform display of three sine waves. But what bugs me is that how can it be possible that a built in three-phase bridge rectifier does not need a neutral wire to work perfectly (to receive the three sine waves), but my advanced oscilloscope cannot do that without a neutral wire. I just cannot wrap my head around this and any further explanation would be much appreciated. Thank you
« Last Edit: September 08, 2013, 05:11:58 pm by karotto »
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #7 on: September 08, 2013, 06:24:19 pm »
In general what you were taught was correct. You do need a closed circuit for current to flow and be measured. However, if you are only measuring voltage and not much current needs to flow, then some non-obvious return paths like parasitic capacitance and inductance can allow the signal to be seen.

As far as the bridge rectifier is concerned, your alternator has three output terminals. At any time the voltage on one terminal will be higher than at least one of the others, so the current can flow out from one of the three terminals and back to another terminal. The three-phase bridge rectifier is arranged so that whether the current is flowing from terminal 1 to 2, or 2 to 3, or 1 to 3, or 3 to 2, it always flows the same direction through the load. Thereby the AC output is converted to DC.
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #8 on: September 08, 2013, 06:28:01 pm »
Incidentally, if you connect the scope ground clips to one of the alternator terminals as a ground reference, and probe the other two terminals with two channels of the scope, you will see two sine waves relative to the reference terminal. There is a sine wave voltage between any two terminals on the alternator, as well as as between any one terminal and the neutral point at the center of the star.
 

Offline tszaboo

  • Super Contributor
  • ***
  • Posts: 7377
  • Country: nl
  • Current job: ATEX product design
Re: how to measure 3 phase with oscilloscope
« Reply #9 on: September 08, 2013, 08:48:49 pm »
Incidentally, if you connect the scope ground clips to one of the alternator terminals as a ground reference, and probe the other two terminals with two channels of the scope, you will see two sine waves relative to the reference terminal. There is a sine wave voltage between any two terminals on the alternator, as well as as between any one terminal and the neutral point at the center of the star.
I think you will see something evaporating in the wiring. The output is not isolated, if you connect the ground to one of the terminals, you basically shorted the terminal with your scope. You generally dont want to short anything with your scope.

karotto: The beauty of the 3 phase, that it doesn't necessarily need neutral connection. You still need earth for safety. The neutral and the ground is not the same thing, and again, you connected the two together with your scope. The star connection is the neutral, and the scope negative is the ground. There could be some volt difference between the two, driven by 0 ohm resistance. If there is a wiring error, expect up to 400 volt on the neutral.

karotto: What you are doing is NOT SAFE


mod: Ok, probably car stuff, not mains stuff.
« Last Edit: September 09, 2013, 12:03:57 am by NANDBlog »
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #10 on: September 08, 2013, 11:32:10 pm »
I think you will see something evaporating in the wiring. The output is not isolated, if you connect the ground to one of the terminals, you basically shorted the terminal with your scope. You generally dont want to short anything with your scope.

It's an alternator. Unless someone has grounded one of its outputs, it is completely floating. The ground clip of the scope is connected to mains ground. That doesn't mean it will short any floating external power supply.
 

Offline G7PSK

  • Super Contributor
  • ***
  • Posts: 3861
  • Country: gb
  • It is hot until proved not.
Re: how to measure 3 phase with oscilloscope
« Reply #11 on: September 09, 2013, 07:43:25 am »
But what will all this achieve. You will see 3 sine waves as already pointed out you know they are 120 deg. apart. If you want the rotation all you need is to look at the alternator, if you can't do that there are better devices for checking rotation.
If its a high voltage alternator it would be far better to look at each phase in turn using the probes in differential mode, but we now know from the posts that this is a car alternator (why the OP could not contain this information I do not know).
 

Offline karottoTopic starter

  • Newbie
  • Posts: 6
Re: how to measure 3 phase with oscilloscope
« Reply #12 on: September 10, 2013, 01:43:59 am »
This is an educational endeavor. I'm trying to demonstrate to me and a few friends. I'm sorry but I still don't quite understand. in star configuration two phases are connected together. But then why wouldn't the phases combine and enhance or cancel each other out at any given position? Where do those nice graphs of 3 sine waves 120º apart come from which my alternator is supposed to deliver (without a neutral!)? My alternator is supposed to produce this and I *want* to measure it *without* a neutral because my alternator does not use a neutral to supposedly send these 3 sine waves to the rectifier. Sorry to be so dense but I just can't wrap my head around this. thank you
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #13 on: September 10, 2013, 03:18:28 am »
Number the three alternator outputs 1, 2 and 3.

Connect the ground clip of a scope channel to output 1 and probe output 2. You will see a sine wave.

Now move the ground clip to output 2 and probe output 3. You will see a sine wave again.

Lastly move the ground clip output 3 and probe output 1. Again a sine wave.

Here are your three sine waves. Each will be 120 degrees out of phase with the other, but unfortunately the only way you will be able to see each of them on the same display at the same time is with differential probes on the oscilloscope, because otherwise there is no common ground or neutral point to connect to.

Each of the output terminals has a sine wave voltage relative to the other two terminals, but the voltages go round and round in a circle.

You could say that terminal 1 is the neutral reference for terminals 2 and 3, but then you cannot measure the voltage on 1 because you have made it your neutral reference, so it is by definition at zero.

However, you could make terminal 2 your neutral reference and then you could measure sine waves on terminals 1 and 3. Which ever terminal you pick as your reference, you will measure sine waves on the other two terminals.

It's like one of those M.C. Escher drawings where everything is relative to everything else and there is no ground:

 

Offline Kremmen

  • Super Contributor
  • ***
  • Posts: 1289
  • Country: fi
Re: how to measure 3 phase with oscilloscope
« Reply #14 on: September 10, 2013, 07:30:12 am »
[...] in star configuration two phases are connected together. But then why wouldn't the phases combine and enhance or cancel each other out at any given position? Where do those nice graphs of 3 sine waves 120º apart come from which my alternator is supposed to deliver (without a neutral!)?
OK, basics. Your alternator is a _generator_ right? Works by magnetic induction as do all generators of this kind. So what you will get is an induced voltage in each of the 3 coils. Forget for a moment that the coils are connected to anything, and just think of them standing free with both coil ends unconnected. So you have 6 wires coming out. Measuring at the ends of any one coil, you will see a sine wave whose amplitude is proportional to the induction and frequency to the rotating speed of the generator. Lets call it Vcoil for later reference. This should be obvious. Assume you have 3 scopes (or 3 diff probes or whatever, so that you can see the coil voltages all at the same time. Further assume that your probe polarities are all the same (say probe ground on the anticlockwise end of each coil). You will now see those same sine waves, shifted 120 degrees to each other. Your scope may have a "ground wire" in the probe, but the generator does not. Every one of those coil ends is just a coil end.
Ok, leave the probes aside for a moment. If you now connect all coils into a loop (thus creating a delta connection) basically nothing changes. Remember that the voltages generated by the coils are at mutual 120 degree phase shift. If you know how to do the Kirchhoff voltage analysis across the loop, you will see that there is no net voltage, thus no idle current circulating in this loop. Or you can take my word for it :). If you now measure the voltage across one coil, you will see that nothing has happened - the voltage is the same Vcoil as it was when the coils were unconnected. Please understand that the word "ground" is just a reference point. The real world doesn't need that for anything. Thus this circuit configuration is perfectly happy without one - it is just the voltages between circuit nodes that are visible/measurable. You can call one of those "ground" if it floats your boat, but it doesn't change anything. (But it doesn't really make much sense either in this symmetric circuit).

OK, go back to where all the coils were separate. If you now connect all say anticlockwise coil ends together, you obviously have a star connection. The central connection point can be called the "ground" or neutral but you don't need to connect it to anything. Assume for a moment that it is invisible. Since one end of each coil is hanging in the air, there obviously won't be any circulating currents, but what about the voltages you will see between the loose ends? If you now connect a scope probe between any 2 free ends, what will you see? Assuming the generator is still the same and the speed is the same as before, you will obviously see the same frequency sine wave. Only this time the amplitude will be sqrt(3) * Vcoil. Why - because there are now 2 coils in series between any 2 free ends, and the 120 degree phase shift in the induced voltages will give rise to this multiplication factor. It is almost trivially provable by drawing the voltage vectors on graph paper and measuring the length of the line connecting the free ends. Think of the Mercedes Benz logo with the 3 point star. Each arm of the star is a coil voltage. What you measure is the distance between the end points of the star. Distance = sqrt(3)*length of arm. Same thing.
Quote
My alternator is supposed to produce this and I *want* to measure it *without* a neutral because my alternator does not use a neutral to supposedly send these 3 sine waves to the rectifier. Sorry to be so dense but I just can't wrap my head around this. thank you
And produce it, it does. Your challenge is just that you cannot have one coil end act as the live wire in one measurement and the ground in another, at the same time. That would be short circuiting the generator via your instrument. Like IanB points out, the only way to do this with a scope is to use differential probes. Their point being that the probe doesn't really have a ground. It just measures the voltage difference between the probe electrodes, exactly as the generator produces its voltages between coil ends. No ground anywhere in sight, unless you decide to call one point ground.
If this still doesn't light a bulb, the only additional thing i can recommend is to brush up about the basics of polyphase generators. Google should be your friend, as usual.

Re Escher: Gotta love the man. An artist after every engineer's heart.  ;D I have at least 3 books of his art in the shelf and have visited several exhibitions displaying his original art, including many of the best know pieces. He wasn't much appreciated in his creative years but later disciplines such as topology, fractal math, chaos theory etc etc have found plenty of resonances with his art. For anyone interested, a recommendation: if you got the chance, do read Doug Hofstadter's "Gödel, Escher, Bach: an Eternal Golden Braid - a metaphorical fugue on minds and machines". As the name says, it prominently features Escher's art, including "Relativity" in IanB's post.
Nothing sings like a kilovolt.
Dr W. Bishop
 

Offline G7PSK

  • Super Contributor
  • ***
  • Posts: 3861
  • Country: gb
  • It is hot until proved not.
Re: how to measure 3 phase with oscilloscope
« Reply #15 on: September 10, 2013, 07:31:33 am »
The alternator does add the voltage of two coils for the power output when in star mode just not full coil voltage its Volts of one coil to neutral X square root of three to find the three phase voltage, hence a 240 volt single phase gives you 415 volt three phase and its divide by square root of three to find the single phase voltage, if you alternator is producing 15 volts at 3 phase each phase to neutral is 8.66volts.

There is one method that you could use to see all three signals on one scope, switch your scope to AC and then connect your earth leads via a small capacitor ( 100nF ) as you are looking at ac you will see the sine wave but any voltage measurement might well be out, but you can check that with a DVM.
« Last Edit: September 10, 2013, 07:39:45 am by G7PSK »
 

Offline karottoTopic starter

  • Newbie
  • Posts: 6
Re: how to measure 3 phase with oscilloscope
« Reply #16 on: September 12, 2013, 03:21:35 am »
thank you much. I still can't quite wrap my head around why the rectifier "sees" what my scope cannot see but I know that all youre explanations are as good as any explanations will ever be. I will go back to the basics and learn about AC before I dig more into generators. thanks again
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 11885
  • Country: us
Re: how to measure 3 phase with oscilloscope
« Reply #17 on: September 12, 2013, 03:43:32 am »
But your scope can "see" what the rectifier sees, only not all at the same time.

If you connect your scope between any two of the three output terminals of the alternator, you will see an AC sine wave on the screen. If there is an AC sine wave then diodes can rectify it. To do this you need two diodes, one for the "out" and one for the "return".

When there are three output terminals, there are three pairs of terminals you could connect the scope between: 1-2, 2-3 and 3-1.

Each of these gives a sine wave, and each sine wave can be rectified.

Since each output pair needs two diodes to rectify it, you need six diodes in total for the three phase rectifier. If you examine the rectifier, that is what you should find. Three "out" diodes, and three "return" diodes.

If you are puzzled about why there is no neutral, just think of a battery. Any given battery has no neutral terminal: it has a +ve terminal ("+" is not neutral) and a -ve terminal ("-" is not neutral either). Circuits don't need a neutral to work, they just need a "+" and a "-". The three phase alternator is cunningly contrived so that the three terminals are alternately "+" and "-" in such a way that no short circuits happen. It's a kind of mathematical magic.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf