Harmonic Analysis: sources to measure and for how long (and other questions!)

[tricard]

Hello,

I have been requested to perform an harmonic analysis for the very first time (always exciting to learn new stuff ). I have some training in this study area but to date have never applied these concepts. Specifically I am going to be performing an analysis at a water treatment plant operating some moderately sized motors, of which some are fed through VFDs and some are fed directly from the line. The contract requires that a study be done to ensure levels of total harmonic distortion are not exceeded as per IEEE 519 standards with the installation of a new VFD to feed one of the 4160V motors (350hp). Since the system is not set up yet, I was going to contact the VFD manufacturer to get the harmonic data on that new drive and then also measure the power quality throughout the remainder of the system.

This is what I was thinking of measuring. I will be using an old RPM 1650 power recorder that records up to the 40th harmonic:
- measure at the main 4160 bus which feeds two 600V VFDs (via 500kVA 4160/600V DY transformers), a 600V MCC via 300kVA transformer and a 4160V MCC (where the new VFD is being connected; all 4160 motors are presently fed directly off the line). Loading will be relatively constant, so I thought to measure it for a day or two
- measure at each individual 600V VFD
- measure on the 27.6kV plant primary (if instrumentation transformers allow it): the system is fed by a 2500kVA 27.6kV/4.16kV DY transformer (NGR grounded).

Are these reasonable measurement locations? Should I measure anything else? I can't see the 600V MCC producing huge harmonics, however I also don't know if there are any LV VFDs connected to it...

And finally: is there particular data I must request from the utility or just their short circuit MVA values (max and min).

I am using ETAP for the analysis (I know this isn't the best software to use for this type of analysis, but I have to use what I got!)

 

[MikeinOK]

One thing to keep in mind when you are measuring the primary is that instrument transformers can filter harmonics pretty effectively.
I don't know how much time you have before the study but the best reference I have ever read on harmonics and how to do a harmonics survey is Harmonics: A Field Handbook for the Professional and the Novice by Mark Waller. Link
Also, if you ever get the chance to participate in any seminars put on by Alex McEachern Link he is one of the world's foremost experts on the subject.
I know none of this really helps you in the short term but it can help a lot if you are going to be doing more of these type surveys.

 

[desrod]

Harmonics studies are tricky. I always use experienced teams to make measurements and I try to have measurements prior or without the new equipment to get a proper assessment of the background harmonics especially if I am designing the filters. Depending on the supply network, there is a risk of resonance and we prefer to get 2-3 bus level upstream to clear that. IEEE 519 defines harmonics limits to the 50th, check with your customer and supply utility is it is ok with them to only measure the first 40.

 

[tricard]

Thanks for the input both Mike and desrod. The harmonic study requested scope in the bid document only had about four sentences in it. I would imagine this is because harmonic studies haven't had the same level of scrutiny as arc flash studies in recent years (there were five pages of requirements for the coordination, short circuit and arc flash study yet only half a paragraph for the harmonic study) . I will send some questions to the customer and see if we can measure values before and after.