Oilfield Harmonics 1

[s22j26]

Hi All,
I'm working in a west Texas oilfield. The motor that drives the pumpjack is 460V, 3-ph, 60 HP. It is powered directly from the grid: no generators or VFDs, etc. Looking at the current to the motor on a scope, I can see a lot of 5th-harmonic (300 Hz) content distorting the current waveshape. I then used a Dranetz power analyzer which reported that the harmonic content constituted 7 A out of 42 A of RMS current. This is a lot more than I have seen at other fields.

This motor had power-factor correcting capacitors on it. I disconnected them to see if they made any difference, but I didn't see any.

Could the use of a VFD or PF-caps elsewhere in the field affect the current observed at this motor? If there were any, they would be at least 1/2 mile away, and not on the same transformer. Any ideas on how to find the source of the distortion?

Thanks!

 

[waross]

Try looking at the quality of the incoming power with the motor offline. If there is voltage distortion on the incoming lines it may cause current distortion in the motor current.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Kiribanda]

s22j26,
While agreeing with Waross,the FLC of your 60HP/460 motor = 78 A(rms).Your measured current is only 42 A(rms).That means the motor is loaded approximately 50%.It means slightly higher than the no-load current.If your 3-P input voltages are balanced,then donot you think that this low current contains harmonics?Is there any way to load it to its FLC=78 A and then check the harmonic contents of the input load currents?

 

[Skogsgurra]

I wouldn't worry at all. A harmonic current contents of around ten percent is quite normal even if it isn't seen everywhere.

Did you find this as a result of searching for a specific problem with that motor? Or did you just 'happen to see it'?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[electricpete]

The possible causes:
1 - Caused by power supply harmonics as mentioned by Bill
2 - Caused by motor saturation as suggested by Kiribanda.

I tend to think #2 is more likely since you describe no sources of pollution to your power. As Kiribanda mentioned your motor is operating at low load which means you are seeing primarily the magnetizing component of the current. The magnetizing component will show the effects of saturation (i.e. harmonics) most.

Which harmonics do we expect? Analysing a single phase L/R circuit, we'd expect to see a boost at the + and - peak in the current waveform. The single phase circuit waveform has quarter wave symmetry which gives odd harmonics 3/5/7/9/11/13 etc. In a 3-phase wye circuit, a balanced 3rd and 6th harmonic (zero sequence) cannot flow so your left with 5/7/11/13 getting smaller as the harmonic number increases. So we expect 5th to be the most prominent and that's what you've got.

This may simply be inherent in an aggressive near-saturation motor design (and operating at low load which makes harmonics a higher fraction of total). It can also be exaggerated if the motor is exposed to higher than nameplate voltage. Less likely it can be exaggerated by unbalanced voltages although then you might see some 3rd harmonics.

If you post current and voltage waveforms you may get more/better guesses.

I agree with Gunnar and others it is certainly not a cause for concern at this loading level. If you'd like to discuss derating as you approach full load, that's another question.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Any ideas on how to find the source of the distortion?

If you have a daily cycle of voltage hi/low... check the current once at high and once at low. If current THD is following voltage magnitude, most likely motor saturation.

Current and voltage waveforms may provide some clues.


=====================================
(2B)+(2B)' ?

 

[electricpete]

Also some motors can have different ways of connecting them (dual voltage.. wye-delta start etc). Misconnection can cause the same saturation effects as high voltage

=====================================
(2B)+(2B)' ?

 

[s22j26]

All,
Thank you very much for your replies.
- I will return to the well to record the voltage with the motor turned off, as Bill suggested.
- The pumpjack motor has a cyclical load. It varies between no-load and 120% of full-load with a period of about 8 seconds. Whether lightly loaded or heavily loaded, the harmonic RMS content is about the same. In percentage terms it is up to 20% when lightly loaded and less than 10% when heavy.
- I believe that the nameplate voltage is 460V (will check). The applied voltage is around 485V when lightly loaded and is pulled down to around 450 when heavy.
- I stumbled on this observation, then went back to check wells at other sites and found that the harmonic content was significantly less at those sites.
I'll post more information in a few days.
Thanks again, happy holidays.

 

[s22j26]

Hi All,
As suggested, I turned off the motor and used the Dranetz to measure THD of the voltage. I'm seeing about 39 V in the 300 Hz component out of a total RMS of 480 V. At another well in the same area (although maybe 2 miles away), the 300 Hz component is only 21 V out of 480 V. I don't know whether this difference is significant but I'd like to understand it. - What are the possible sources that might be creating this component?
- How close would the source need to be to still have this effect?
Thanks

 

[Skogsgurra]

The load that produces the fifth harmonic (300 Hz) can be anywhere in the system. Distance has very little influence on low order harmonics.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[HornTootinEE]

Some of the oil field distribution systems up here in the ND oil fields have run into these types of problems, but they have stemmed from the use of 6-pulse VFD drives and no filtering that really muck with your Voltage waveform. But typically they see alot of 3rd harmonic as well, so not sure if that would actually be your issue or not. Something to look around for though. Somebody may have a 6-pulse VFD and are attempting to filter the harmonics with some sort of low-grade passive filter