Tolerable harmonic distortion level - 480 V 2

[dpc]

We have a pump station application where 95% of the load will be VFDs (480V). The interface with the utility is on the primary of the main transformer, so harmonics there are buffered by the transformer impedance.

I'm wondering what your experience is regarding the magnitude of harmonic distortion that would be tolerable on the 480 V side without causing problems for PLCs, computers, and the drives themselves? We're trying to determine what type of filtering would be adequate (for lowest possible cost, of course.).

Based on past experience, I wouldn't expect problems if voltage THD is 10% or less, but thought I'd check with the experts.

TIA,

Dave

 

[Skogsgurra]

Dave,

I regard you being one of the experts. Anyhow, here is my contribution.

I once went to a paper mill to show some new instruments I had bought. A Fluke harmonics analyzer and some other gear. We checked the outlet in the chief engineer's office - 14 % THD! Hmm...

Took out the Dranetz PowerPlatform 4300. Yes, we had those 14 % all over that part of the factory. Lots of seventh harmonic. Turned out to be resonance in power factor correction capacitors and stray inductance of feeding transformer.

No one had ever noticed. And no one had any problems. There was the usual mix: drives, SCR and VFD, computers, instrumentation, lighting, direct-on-line async motors.

So, that kind of answers part of your question.

The European standard says 10 % THD in industrial grids. And 15 % for shorter periods - I think 30 seconds.

I would be concerned about the higher frequencies - especially if you are installing VFDs with AFE (active front end) to get a better power factor. They (at least some makes) are known to generate lots of HF conducted pollution. And that is a threat to computer and other low-power PSUs.

Gunnar Englund

http://www.gke.org

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

 

[ScottyUK]

10% voltage THD is getting high - how big is the total load and how stiff is the local utility supply? That will influence how much disturbance your load causes to the utility system, and potentially other customers.

The following link relates to the Electricity Association's Guidance Note G5/4 which originates in the UK but I've posted it because some of it might be of interest.

http://www.gambica.org.uk/pdfs/G54-1.pdf

----------------------------------

Sometimes I only open my mouth to swap feet...

 

[CJCPE]

IEEE 519 recommends 10% maximum total harmonic voltage distortion for a distribution system that has only drives connected to it. For general purpose distribution systems, the recommended maximum voltage THD is 5%.

The biggest worry would probably be resonance with any power factor correction capacitors and excess capacitor current.

The transformer is not necessarily going to prevent complaints from the neighbors and/or the utility. At the utility connection point, IEEE 519 limits the current distortion rather than the voltage distortion. The question may not be whether you cause a problem but whether you distort the current more than your share. You should do an analysis to see if you are within IEEE 519 limits.

 

[dpc]

Thanks,

Yes, we will do what is necessary to meet IEEE 519 5% limit (and related current harmonic limits) at the point of common coupling. The system is fairly weak and the transformer is small. There is nothing on the transformer except two VFDs and the necessary controls to run them, plus a little lighting, etc. No capacitors. It's a similar situation to running a pump station on a standby generator.

My experience has been the same as Gunnar's. I've measured quite high voltage distortion levels at facilities where no one had ever had a problem.

Supplier originally proposed active filters to get done to 5% on the secondary side, but the cost was shocking to the owner. I'm thinking that reactors or maybe a broadband filter might do the trick for less cost, if they can live with somewhat higher distortion on the 480 V side.

Cheers,

Dave

 

[DickDV]

It can be beneficial to split the VFD load and connect half to a delta transformer secondary with the other half on the wye transformer secondary. Has many of the benefits and characteristics of 12 pulse drives without the cost (except for the dual secondary transformer, of course)

Any good answer for this has to involve a lot more specifics like lead lengths, transformer impedances, and drive hp ranges.

I guess that's why drive manufacturers use computer simulations to predict network harmonics. I'd use that much before any seat-of-the-pants estimating.

 

[dpc]

Dick,

I can calculate (or estimate) the distortion, but I guess my real question is at what level of THD should there be a concern regarding the impact on other equipment? At what level do you start getting worried.

 

[CJCPE]

If you don't have any capacitors to worry about, I think that 10% THD voltage will be OK for other loads.

 

[Marke]

Power factor correction capacitors are greatly affected by harmonics voltages on the supply, but most other equipment is relatively immune to them with the exception of rotating machines.
I have often seen the "weighted THD" as being a better indicator of what will cause problems. A high level of third harmonic is not nearly as bad as the same magnitude of say 7th harmonic.

Where you have motors operating of the supply and there is a lot of harmonic voltage distortion, you will have slip losses for each of the harmonics and these slip losses will be dissipated in the rotor of the motor.

If there is significant even order harmonic, this will cause serious problems with any iron cored devices, particularly transformers and motors. Even harmonics are bad, real bad!!

Distortions that occur around the voltage zero crossing, are more likely to affect electronic equipment, particularly SCR and triac based equipment where the synchronisation can be interfered with. Notches in the voltage waveform can cause SCRs and triacs to delatch or commutate early and in some equipment this can be a problem if continuous firing techniques are not used.

I do not think that there is a magic figure that you can apply, but I would look for a weighted THD figure rather than a standard THD figure as this is more indicative of potential issues with other equipment, plus the potential to interfere with audio circuits.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[ozmosis]

Dave
Even though the VFD load would equate to 95%, it would depend on the ratio between the actual load and actual rating of TRX rating. As a general rule, once the non-linear load is above 20% of the TRX rating, then I would look seriously at harmonic mitigation techniques.

 

[dpc]

Thanks Marke. I agree it is hard to make a general rule that will always work. The drives appear to be standard 6-pulse PWM drives from Eaton, so I assume there will be a high percentage of 7th and 5th.

sed2,

The two drives will be about 70% of the transformer rating. I've already recommended a larger transformer because they have visions of adding a third drive in the future. But they don't want to replace it now.

Thanks all,

Dave