RSO tests

[edison123]

Can anyone explain what is Recurrent Surge Oscillograph (RSO) test of Generator Rotor and how it is done ?

 

[electricpete]

RSO (recurrent surge oscillation) tests look for shorted turns based on the principle of time domain reflectometry. Time Domain reflectometry as you may know works on principle of creating a traveling wave (steep voltage wavefront with high time rate of change) and looking at the timing of reflections which are created when the wave hit's a change in impedance (for example a shorted turn). I have seen TDR done at our plant looking for instrument cable connection problems... we used equipment from CHAR services. Sorry I don't know much more about it.

 

[busbar]

An example of related test equipment may be

http://www.bakerinst.com

 

[electricpete]

That is an interesting comment busbar and started me thinking about the difference between RSO (for generator rotor) which I am not too familiar with and the motor surge test for stator turn insulation condition (Baker) which I am more familiar with.

The surge test is potentially destructive. It uses relatively high voltages... on the same order as the dc hi-pot test voltages.

I believe the RSO is non-destructive and uses much lower voltages (perhaps 10 volts or less?) although still short rise times.

Someone posted a link with lots of info on rso in response to edison's question on another board

http://www.epri.com/attachments/224581_ShortedTD.pdf

Both tests use a principle of comparing two identical windings... look for similar traces. My casual observation from the above link is that the curve doesn't look like any surge tests results I have seen... surge tests usually have more oscillation. I'm not sure why that would be and whether there are other substantial differences beyond the voltage used. Anyone else care to comment on how a rso test is similar/different than a motor surge test?

 

[UKpete]

ep, I think it's horses for courses. Having read the epri paper on my journey to work this morning (it's quite a long journey) it appears that what they are trying to do is locate the physical position of a shorted turn which is already there, or at least appears when they spin the rotor. This is probably only of interest on machines where a partial rewind is physically possible and economic.

RSO (or TDR, time-domain reflectometry) is low voltage so it will not find any weaknesses in the inter-turn insulation that may at some future stage turn into a full short. Surge testing however does provide that "punch", the electrical stressing of the inter-turn insulation. Hence I think it is generally more applicable to electrical machines except as mentioned above.

The big question with surge testing is: how hard do you punch, too much and you may damage a healthy winding, too little and it won't find the weaknesses. Also, does it propagate sufficiently through all the turns?

 

[jbartos]

Suggestion: Visit

http://www.spectrum.ieee.org/WEBONLY/publicfeature/feb01/wire.html

for Time Domain Reflectometry (TDR), Frequency Domain Reflectometry (FDR) and Standing-Wave Reflectometry (SWR)

 

[busbar]

I like Paul (A.S.) Gill’s perspective in his book on electrical testing: before starting a critical test, make sure there are spare parts immediately available.

 

[edison123]

Somewhere, I have read that RSO tests use very low step DC Voltages (about 12 V)between one end of rotor winding and rotor body and measure the results with digital scopes. Any comments?

 

[jbartos]

Suggestion to busbar (Electrical) Jun 18, 2003 marked ///\\I like Paul (A.S.) Gill’s perspective in his book on electrical testing: before starting a critical test, make sure there are spare parts immediately available.
///This is certainly a sure way to do it. However, considering nowadays trends, e.g. Wall Street approach to electrical enginering and testing, the extra requisites tend to be discarded if some harmless testing methods and instruments are applied.\\\