VFD's & IEEE Std 519

[rwilsond]

I'm involved in the selection of VFD's for some large AC motors (up to 450HP). The spec that was given to the supplier imposes the requirement that these VFD's "shall meet all requirements of IEEE Std 519".

I have IEEE Std 519 ("Recommended Practices and Requirements for Harmonic Control in Elect Power Systems"), and have read through it. But I do not see "requirements"; it seems to be nothing more than suggested guidelines.

So, what is there in IEEE Std 519 that can be considered to be REQUIREMENTS?

 

[sberbece]

Usually VFD-s produce harmonics. What harmonic orders and magnitudes depend on the VFD type and the manufacturer.
However, you shall be sure that methods of harmonic mitigation are in place and the total harmonic distortion (THD) meets the requirements of IEEE 519, which is TDH ? 5%.

In addition, if your power distribution system contains large capacitor banks (i.e. for power factor correction), you have to undertake a harmonic analysis/study to see if resonant condition will be created by installing the VFD and to install harmonic filters in order to eliminate the resonance condition.

Regards,
Stefan

 

[dpc]

There are limits for voltage and current harmonics at the point of common coupling with the utility. That is what is being referred to. I don't have my copy in front of me, but the requirements are definitely in the Standard.

 

[stevenal]

I see your point, and I suppose that legally the supplier has an easy out. The intent is clear enough, though, the supplier is expected to meet the recommendations. The spec converts the recommendation into a requirement.

 

[magoo2]

IEEE 519 has a table of current distortion values that must be met by a site so that the resulting voltage distortion typically is less than 5% THD for total distortion and 3% for a single harmonic. It is not a recommended limit to be applied to individual equipment. To apply it in this manner would be ultra conservative and would typically result in the total distortion at the site being less and probably much less than the limits of 519.

The table of current distortion limit are given for different ratios of short-circuit/load current. In effect, a stiffer system, one with higher ratios of short-circuit/load current, should be able to tolerate higher amounts of current distortion.

A more reasonable approach for your predicament might be to consider meeting IEEE 519 where the VFDs are some percentage of the today facility load like 25, 50 or 75. Right now you're considering it being 100% of the load which is the most limiting from an IEEE 519 perspective.

 

[jraef]

Looks to me as though the spec writer took a lazy way out of doing a proper job.

What I usually see is a spec requirement that the contractor shall demonstrate, upon completion of the installation, that the power system, after having had the VFD installed and operating, meets the recommendations in IEEE-519. That means that whomever takes on supplying the equipment must do a site analysis of the existing equipment (if a retrofit) or of all other connected equipment WITHOUT the VFD, and do a Fourier analysis of the VFDs contribution to confirm that with the VFD connected and operating, it does not cause the system to exceed those recommendations. That really is the only legit way to do it, because VFD manufacturer has no control of what might have otherwise already been there or distorted caused by other connected loads.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[rwilsond]

Thanks for the replies; they were very useful.

FYI - the two largest VFD's (for 450 HP motors) will be part of a 44-ton capacity bridge crane. I'm simply providing a 5 kV feeder to the 1600 kVA transformer that rides with the bridge (this transformer supplies all power to the bridge crane). The feeder runs several hundred feet to my MCC. I will consider my MCC bus to be the Point of Common Coupling (PCC). So, I will understand that once all is installed and in operation, the harmonic distortions, at this PCC, should be within the limits of Chapter 10 of IEEE Std 519.

Feel free to let me know if I got it wrong.

 

[dpc]

Seems reasonable. We generally specify what is to be considered the Point of Common Coupling.

But this all assumes that the VFDs in question are the only harmonic sources.

If the voltage distortion is already 4.5% without the drives running and 5.5% with the drives on, does this meet the requirement?

 

[ozmosis]

rwilsond
Firstly, get the terminology right:
“Total harmonic distortion” is defined as either harmonic voltage or current distortion factor. Therefore, the equation given for distortion factor is used to calculate Total Harmonic Current Distortion (THID) and Total Harmonic Voltage Distortion (THVD).
“Total demand distortion (TDD)” is defined as the “root-sum-square harmonic current distortion in percent of the maximum demand load current”. The equation for TDD is different to THID. The difference is that the total load current must be used in the denominator. This is the total fundamental load current resulting from all loads connected to the point of common coupling (PCC), not just the fundamental current resulting from the non-linear load(s)
The calculation of TDD is therefore a measurement of the system’s performance, not of a specific product. To calculate TDD, it is necessary to know the total fundamental load current of the system (IL).
If the total load current is unknown, a best estimate of TDD can be made by using the rated load current of the supply transformer as IL in the calculation. Using this in the calculation would assume the transformer is fully loaded.
I would say from the initial information provided, that some form of harmonic mitigation would be required, depending on the type of VFD you are using.
I did a quick calculation using just 2 x 450hp VFD's, based on our range that have built-in DC link chokes. The THvD came to 10.38% and the TDD (Total Demand Distortion) came to 17.2%. Just using these 2 drives indicated that 49% of the transformer load was harmonic load.
There are numerous methods of harmonic mitigation, I used a passive harmonic filter in the same calculation and this brought the THvD down to 2.34% and TDD to 3.12% at the transformer.
Note: these are only 'finger in the air calcs' based on the basic information provided and is assuming no other linear load and no other non-linear loads on the PCC.
The supplier of the VFD's should be able to provide some sort of calculation to simulate the their product, but as already pointed out by others-if the background harmonics are already over the limit, even supplying VFD's with zero harmonics would not meet IEEE519.
It is a system guideline not a product standard!

 

[davidbeach]

ozmosis, I'll take a bit of issue with "if the background harmonics are already over the limit, even supplying VFD's with zero harmonics would not meet IEEE519". If you can add enough load with zero harmonics you can bring down both the THID (and therefore the THVD) and the TDD. You can always correct a harmonics problem by adding enough resistance heat.

 

[ozmosis]

David
I agree with your comment. I was referring more to the mindset of specifications that seem to indicate that IEEE519 is a focus on individual products rather than the system. Any solution would require knowledge on what is there already and therefore the ability (wish) to ensure you do something about it prior to installing any additional non-linear load.

 

[rwilsond]

The bridge crane (including its 1600 kVA transformer and VFD's) will be a very small part of this yet-to-be-built plant (which will use 100's of MVA overall), and I doubt it would have a significant contribution on the overall power system distortion.

So, I must somehow simplify this issue with the crane bidder, and focus on a localized area about the PCC (i.e., MCC bus). I am considering providing them with some estimated info: the available short circuit at the PCC; the size & length of the 5 kV feeder from this MCC to their 1600 kVA transformer; and the demand load of this MCC on its upstream transformer.

Would it then be meaningful and reasonable to ask this vendor to provide their estimated contributions to distortion at this MCC (disregarding the contributions from other sources, since those contributions are beyond their scope)?

 

[ozmosis]

The point of common coupling (PCC) is defined as the point in the network where the consumer’s network connects to the power utility network. This could be on the primary or secondary side of the low voltage transformer depending on the installation. However, it also defines that within an “industrial plant, the PCC is the point between the non-linear load and other loads.” For example the PCC could be at a common distribution bus bar within your power supply network.
On this basis, your requests on the vendor would be meaningful in my opinion.
If you know the maximum short circuit current at PCC and the maximum demand load current (fundamental component) at PCC, then you know the ratio. Using the ratio you can determine a guideline as defined in IEEE519 (table 10.3)for "maximum Harmonic Current Distortion in % of demand load current for the harmonic spectrum and therefore the TDD.
Ratio<20 would require a TDD of 5%
Ratio 20<50 =8% TDD
etc, but it sounds like the ratio will be small.

 

[thinker]

We supply high power multiple VFD based dynamometer systems, and our customers now include in their specs our suggested wording on IEEE519 compliance. We confirm that the CONTRIBUTION of the SINGLE VFD to voltage and current distortion will not exceed values recommended by the Standard, if short circuit ratio at PCC is greater than XX. We urge the customer to undertake a full harmonic study after all VFDs have been installed to address multiple drives harmonics interaction issue, and come up with the filtering solution on the common AC bus level.
Also, IEEE519 is quite obsolete since it addresses mainly 6-pulse conversion technology. Most of modern high power regenerative drives are based on PWM active front end with very different voltage and current harmonic contents.