How to measure 3phase 480VAC with oscilloscope

[jzhvymetal]

I have been investigating if this can be done with “Pseudo-Differential method” aka “A minus B”.    Since the scope I’m using is a 4 channels scope is it possible to safely display all three phases by using the following three math functions CH1-CH2, CH1-CH3, and CH2-CH3.   Is there potently any safety issues doing this?  This would be done with 100x probes without the ground clip from the probe connected. See attachment for A-B measurement. Also below are some links I found on different methods.


Fundamentals of Floating Measurements and Isolated Input Oscilloscopes
http://info.tek.com/rs/tektronix/images/3AW_19134_2_MR_Letter.pdf

Emerson-Technical Note CTTN #136 Oscilloscope Safety (Must be logged with free account to download)
http://www.controltechniques.com/CTDownloads/SharePoint/Download.aspx?SiteID=4&ProductID=82&DownloadID=2091&VersionID=2342

Do not use floating
http://www.cbtricks.com/miscellaneous/tech_publications/scope/floating.pdf

[alm]

Connecting the probes without ground lead to the individual phases should be safe (though with limited common-mode rejection ratio), as long as you observe the overvoltage category rating of both the probes and the scope. Since 3 phase power is usually used for high power residential or even industrial loads, you might need CAT III 600 V or even CAT IV 600 V rated probes and ideally a scope rated for CAT III/IV (though not necessarily at 600V because of the 100x attenuation of the probe). I believe the Multi-Contact IsoProbe series might contain some suitable probes (also sold under the Fluke brand for their ScopeMeters). Not sure about scopes rated for CAT III/IV beyond the $$$ Fluke stuff.

Of course you might have a low-energy CAT I 3 phase circuit, for which any CAT I 600V probe and pretty much any 1 Mohm scope should be safe. If you only care about the low-frequency content, a suitable 3 phase transformer might improve safety in absence of CAT III/IV probes.

[jzhvymetal]

If you use 100x CATIII probes is it still required to use a CATIII scope?

Do you have a suggestion for a small 3 phase transformer?

[alm]

Technically I guess the scope should be rated CAT III 6 V with a 100x probe . I believe the transient test for even CAT IV 600 V calls for 'only' a 8 kV transient from a low-impedance source. This should be attenuated to 80 V as long as the probe does not fail. So I would argue that pretty much any decent scope should be sufficient assuming you have a solid probe. The probe is your only defense, however. If it would fail short, the scope could fail catastrophically and possibly injure the operator. So no probes made by the manufacturers that make some of those cheap 'CAT III 1000 V' multimeters with glass fuses and pretty much zero input protection .

I am not sure where you would find a small 3-phase transformer. A quick Digikey search did not yield any results. Maybe look at places more aimed at industrial electronics?

[albert22]

I think that three separate transformers. One for each phase will be simpler and will measure each phase separately.
EDIT_
Only if you have access to the neutral wire.

[madires]

And another one Get differential probes of a known brand from a known distributor. Make sure the probe is rated for the voltages you like to measure (don't forget to add a proper safety margin).

[joeqsmith]

I would have no problem tossing that Hioki I showed on a bus like that.   It's designed for mains work and does a good job.  You would have all three current and voltage waves plus an extra voltage and current input (which comes in handy).   I think it can sample at 2Msps in transient mode. You can trigger on just about anything.  Not too expensive if you have a need for it. 

[tautech]

If you use 100x CATIII probes is it still required to use a CATIII scope?

Do you have a suggestion for a small 3 phase transformer?

A viable option might be a pair of Siglent's ISFE
http://siglentamerica.com/prodcut-fjxx.aspx?fjid=404&id=1488&tid=1&T=2
$135

Some info on it here:
https://www.eevblog.com/forum/testgear/siglent-isfe-scope-front-end-isolator-any-opinions/

[Gregg]

You could find two or three control voltage transformers that are very common in industrial motor control etc applications.  Something like this: http://www.ebay.com/itm/Siemens-KT8050P-Industrial-Control-Transformer-50-VA-/131203740829
If you connect them in an open delta, you only need two.  If you want to derive a lower voltage neutral, connect three in a Y configuration and ground the center [neutral].  Transformers might suppress some of the spikes and there is inductance to consider, but the voltage can be brought down to something reasonable.  24 VAC is a common control voltage in the US as well as 110 VAC.

[alm]

A viable option might be a pair of Siglent's ISFE
[...]

No mention of any IEC 61010 overvoltage rating on their website, let alone independent testing (e.g. UL, TÜV). The included leads do not exactly scream 'CAT III' to me either. Not something I would use for even CAT II (e.g. household single-phase mains), let alone a high-energy three-phase circuit.

[MrW0lf]

Seems Pico has crafted sharply focused product for differential/line stuff with 9-pin D-sub active probe connectors, 14bit ADC:

https://www.picotech.com/oscilloscope/4444/picoscope-4444-overview



Interesting field for active probe modding!? Attach D-sub and throw batteries away, fool some simple ID-system?

Think this is how scope wars will go - specialization! Hard to stuff right stuff for all tasks into single scope.

[tautech]

A viable option might be a pair of Siglent's ISFE
[...]

No mention of any IEC 61010 overvoltage rating on their website, let alone independent testing (e.g. UL, TÜV). The included leads do not exactly scream 'CAT III' to me either. Not something I would use for even CAT II (e.g. household single-phase mains), let alone a high-energy three-phase circuit.

You're right.
Pulled mine out and looked at the documentation and on the unit to see if there was any test markings but sadly no. 
Even their current Probes datasheet doesn't have them listed while an older version I have on file does. More 

I'll shoot them an email to get all this updated to a better standard.
Thanks for pointing this out.


BTW, alm, are you the same alm of old having returned to the forum or a new one ?

[rstofer]

I think that three separate transformers. One for each phase will be simpler and will measure each phase separately.
EDIT_
Only if you have access to the neutral wire.

I agree with three transformers but I would connect them delta-wye and ground the neutral point of the Y connected secondary.  The transformer primaries would be across the three phases in a delta configuration.

I would want the secondary voltage as low as I could get.  480-24V transformers are about $50.  We used instrument rated potential transformers in the electrical business although the secondary voltage was often 120VAC.  Serious potential transformers are likely to cost a good deal more money.