TRIPPING OF HARMONIC FILTER 2

[reactive]

We have a 6,6kV 6MW 6/12 pulse thyristor Mine Winder with a 3,1 MVAr 3rd order damped filter (Resistor and decoupling capacitor across the reactor) tuned at 4,8p.u. connected to the same busbar (no other loads). Acceleration of the winder peaks the load at about 15MVA.

Measurement at approx. 500 ms averaging shows normal harmonics for the winder with the filter out of circuit. With the filter in circuit acceleration of the winder results in immediate tripping of the filter on harmonics (the filter is rated for 90% harmonics at 5th).

Parallel resonance below the tuning frequency occurs at exactly 4p.u. We measured the filter harmonics during the trip and managed to see one set of readings as the winder accelerated before the filter tripped. The 4th harmonic was at around 350A on all phases (it could be and is likely, based on the levels that the relay saw, that the peak 4th is much higher).

The 4th without the filter in circuit is 4-5A giving an amplification factor of anywhere between 90 and 300.

Simulation of the network shows a max. possible amplification of 25. What on earth is going on?

Is it possible that some interaction is occuring between the filter and the winder and is it related to the lack of damping on the busbar?

In South Africa we have relatively high levels of harmonics due to the many mines and therefore are very experienced in the field of harmonics. However, this is difficult to explain - the max. amlification that we have seen in 20 years is 40x and that situation was theoretically explained.

 

[busbar]

Long-shot possibility from ieee.org
Paul C. Buddingh, Even Harmonic Resonance—An Unusual Problem IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 39, NO. 4, JULY/AUGUST 2003 p.1181

Abstract—This paper is a case study examining the cause, effect, and solution of an actual “noncharacteristic” even harmonic problem in an electrochemical plant. While issues concerning “characteristic” harmonics are well documented in the literature, limited information is available describing how noncharacteristic harmonic currents create operational problems. The author has not found any papers describing a case of even harmonic resonance and would like to share the experience of resolving a 4th harmonic resonance problem in the hopes it will benefit others.

 

[caided]

The tuned frequency of a harmonic filter is related to the kVAR of the capacitance and the reactor rating. In the case of the capacitance, the operting voltage and the rated voltage are different due to the design approach. Therefore, there is a possiblity that the filter tuned frequency is lower than the expected 4.8. If you can provide all the parameters of the filter I can verify this specific issue.

 

[reactive]

We are specialist consultants. The filter tuning frequency is definitely 4,8.

What we believe now is that the filter parallel resonance (which is spot on 4th)is amplifying the small 4th (4A) 25 to 30 times (100A - 120A), causing synchronising problems with the thyristors. This in turn is causing the winder to generate more uncharactaristic harmonics (4th) and we have a snowball.

Does this sound feasible?

 

[jbartos]

Suggestion/Questions:
1. How do you perform the filter tuning?
2. Where is the 4th harmonic coming from?

 

[reactive]

1. V=6,6kV; C=220uF/phase; L=1,92mH (the damping circuit will not affect the tuning, although for interest the resistor is 5ohms and the decoupling cap 8uF).
2. Any thyristor converter will generate some uncharactaristic harmonics. This winder draws some 2000A during acceleration with peak 4th harmonic currents of 4-5A. This is a very small harmonic but we believe that with amplification the resultant voltage distortion is causing some firing problems.

 

[jbartos]

Suggestion to the previous posting:

1. 4A of 4th harmonic is 5A/2000A x 100%=0.25% which appear to be negligibly small

2. 6-pulse conversion will generate 5th, 7th, 11th, 13th, 17th, 19th, 23rd, 25th, ... characteristic harmonics

3. 12-pulse conversion will generate 11th, 13th, 23rd, 25th, 35th, 37th, ... characteristic harmonics

If the 5th harmonic is present and filtered then the 6-pulse conversion is applied

4. The untuned harmonic filter can absorb an excessive amount of harmonic current.

5. References:
5.1 Steve Ashmore, Manager, Harmonic Filter Systems, High Voltage Capacitor Division, "Capacitor Banks and Harmonics," 1977 ABB Electric Utility Conference, Subject VII-3

Among other contents, it states that the tuning frequency has to be accurately controlled by design, ....

5.2 Derek A. Paice "Power Electronic Converter Harmonics, Multipulse Solution for Clean Power," IEEE Press, 1995

 

[jbartos]

Small encore:
1. The ABB may be contacted for more info:

http://www.abb.com/GLOBAL/ABBZH/ABB...1256A94003313B4/$file/Saldanha+A02-0176+E.pdf

http://www.abb.com/global/ABBZH/ABB...2569740043321D/$file/Degerfors+A02-0115+E.pdf

http://www.abbsales.com/pdf/AC1502.pdf

etc.
2. How high is the 2nd Harmonic?
3. Have you tried to filter out the 4th harmonic?

 

[reactive]

4A is small untill it is amplified 30x.

All thyristor converters generate some uncharacteristic harmonic.

The filter can handle 90% 5th harmonic. It is however tripping on about 1500% harmonic overload.

2nd harmonic is not being significantly amplified and is therefore negligible.

Filtering out the small 4th is not an option. Avoiding significant amplification is, and if the insertion of some damping load does not sort out the problem we will probably convert the filter to a 2nd order filter.

 

[jbartos]

Suggestion: Apparently, the present filter design is amplifying the 4th harmonic due to its resonance.

 

[ScottyUK]

jbartos,

Redwing stated this in his initial post. Have you just come to this conclusion?

 

[caided]

For the given value of C = 220 MFD/phase and L = 1.92 mH, the calculated resonant frequency is 245 Hz. This is 4.9f for the 50 Hz system and 4.08f for the 60 Hz system.

What is the voltage rating on the capacitor unit?
What is the kVAR of the given unit?
How are they connected (how many in parallel and how many in series). This is required to see how much derating is to be applied on the capacitor unit.
Voltage rating of the system = 6.6 kV (line-line)
Voltage rating/phase = 3.81 kV
The available kvar from each unit will be much less than the rating of the capacitor unit.

 

[reactive]

Yes, 4.9p.u. is correct. 4.8 was a typo but makes no difference.

Filter output is 3,14 MVAr, cap is single phase 44uF - 5000V, 1 series - 5 parallel, reactor is 440A, Vcap at nominal is 3976V.

I am battling to see what this has to do with my problem as I know how to work out bank ratings. It is the interation with the winder that I am interested in.

 

[jbartos]

Suggestion to ScottyUK (Electrical) Sep 2, 2003 marked ///\\jbartos,
Redwing stated this in his initial post. Have you just come to this conclusion?
///No. There appears to be a peculiar situation. The unwanted harmonics are usually filtered out or prevented to happen. To filter out 4th harmonic is not an option. See reactive (Electrical) Sep 1, 2003 for the statement. To prevent 4th harmonic to exist, one would have to stop its generation that is allegedly by thyristor converters. Now, there are not that many options left, are there? It looks like some engineering mission impossible subject to further research. It may or may not be for free then.\\

 

[caided]

There are five capacitors in parallel in each phase with the following rating. 44 MFD/phase, 200 kVAR/unit, 5 kV/unit, five of them in parallel. The expected reactive per phase is 1,030 kVAR. The operating voltage of the capacitor is 3.81 kV per phase. The delivered reactive power is 632 kVAR/phase. The corresponding per phase current at fundamental frequency is 165.8 A. The total capacitance per phase is 138.6 MFD. The resonant frequency will be 6.17f. The design resonant frequency is 3.9f. The winder may or may not have a resonant frequency at 4f. But the dominant harmonic frequency at 5f due to the six pulse converter is able to excite at 4f. This analysis is not taking the resistance in to account.

There are ways to correct this condition. In order to evaluate the overall condition, the harmonic analysis program has to be used at various conditions. You may contact Practical Power Associates using the e-mail practical_power@yahoo.com. The owner of this company has extensive experience with filters for arc furnaces, dc converters and utility applications. He is the author of the book: Ramasamy Natarajan, Computer Aided power system analysis, Marcel Dekker, New York, 2002. In this book there are chapters on load flow, short circuit, stability, harmonic filtering, switching surge, lightning surge etc. written using recent computer programs. You can obtain this book through internet.

 

[jbartos]

Comment: My suggestion to filter out 4th harmonic that is significantly increased due to resonance implied a modification of the current filtering scheme. Details of the filter parameters have not been posted.

 

[reactive]

I don't know how you work out that 5x44=138.6. Hopefully you don't design filters.

As I have already stated I don't need help with the filter parameters, I know what they are. I have designed over 200 filters.

Jbartos, there is no way to get rid of the uncharacteristic harmonics i.e. 4th. The problem is that we are amplifying the absolutely normal, almost insignificant 4th.

The solution is to modify the filter to a 2nd order filter by removing the decoupling capacitor. This will increase the damping at lower frequencies. The problem is that the resistor will then be underrated.

My question was really to do with the interaction of the amplified 4th with the winder.

 

[jbartos]

Suggestion to the previous posting marked ///\\reactive (Electrical) Sep 5, 2003

Jbartos, there is no way to get rid of the uncharacteristic harmonics i.e. 4th.
///Yes, I agree that traces of even harmonics are generated by switching scheme and devices that have to be applied.\\ The problem is that we are amplifying the absolutely normal, almost insignificant 4th.
///Agree.\\The solution is to modify the filter to a 2nd order filter by removing the decoupling capacitor. This will increase the damping at lower frequencies. The problem is that the resistor will then be underrated.
///This is what I also meant by "filter parameters (including design circuit) have not been posted." in my previous posting.\\My question was really to do with the interaction of the amplified 4th with the winder.
///What options are left for an engineer and designer of over 200 filters?
1. Prevent 4th harmonic generation - This is not possible due to hardware selection and design.
2. Filter out 4th harmonic - This is not possible as stated by the designer.
3. Inject 4th harmonic of opposite phase.
4. Perhaps, switch to Forum of "Electrical Eng-Tips Magicians"\\\

 

[ScottyUK]

JB,

I like your idea of antiphase current injection. I was involved with some development work on active harmonic filters for LV applications quite some time ago, and had not made the connection with this application. Interesting idea, especially at this sort of power level.

 

[reactive]

I think that the easiest and a relatively low cost solution is to install a higher rated resistor and remove the decoupling capacitor, resulting in much better damping at lower frequencies.

There are literally hundreds of these type of filters installed on winders without problem (although busbars with no other damping loads are rare).

My question wasn't really to do with the solution it was to do with the interaction of the amplified 4th and the winder which results in tripping of the filter.

Even if the 4A 4th is amplified 30x this only gives us 120A 4th which, although not great, the filter can handle (damping loads would decrease this significantly). However, we are seeing 4th harmonic currents in excess of 350A which indicates (my understanding) that there is interaction with the winder (possibly causing problems with the synchronising) which is in turn causing the winder to generate more 4th, in turn being amplified, and so on............

I wanted to know if anyone had had experience with similar interaction due to uncharacteristic harmonics.

Thanks

PS There is no black magic in electricity.