Harmonics and Power Factor Correction issues

[BokD]

Let me preface this with the comment that 'I am not an electrical engineer', so apologise if this does not make much engineering sense or is a bit vague.

I have a mining client who operates a copper mine in the DRC. The associated SXEW consists of about 8MW of Electrowinning (rectification) and approx 2MW remaining is linear loads. The mining client had a consultant provide the design for the Harmonic Filters and the PFC. Typically I am used to seeing the harmonics and the PFC being dealt with separately (as they are completely different problems, but both solved using capacitance) but in this instance the consultant has stated that he has 'designed' a combined capacitor bank that he suggests deals with both harmonics and capacitance. According to the consultants calculations, harmonics problems are expected on the 5th and 12th harmonics and he has designed a 5.2MVAr system to eliminate the 5th harmonic issue and a 9MVAr system to eliminate the 12th harmonic problem. He states that both designs will also get the PF above 0.97. By my calculations his solution results in switching 14.2MVAr to eliminate 5th and 12th harmonics on a total 10MW system??? This just seems wrong - commissioning tests have highlighted that there are 9% over-voltage and leading PF issues when we switch these banks in - most protection circuits downstream trip on overvoltage. The consultant who designed this is convinced that this is not a sizing issue but rather a grid supplier issue. Any thoughts?

Additionally, is it normal to combine a harmonics and PFC capacitor design or are these two very separate issues typically dealt with by the installation of two separate systems?

Thirdly, what is the current thinking around capacitor banks with online generators? Should cap banks (PFC and Harmonic Filtering) be turned off to ensure that the generator alternators are not damaged due to sustained leading PF's that may occur? Has anyone ever managed to get a co-generation (grid and generators) system successfully up and running with PFC and Harmonic Filtering?

 

[wroggent]

Before the capacitors, if the site apparent power is 19.5MVA (or p.f. is 0.51 lagging), then 14.2MVAr would be needed to get the p.f. up to 0.97 lagging. Was the load reduced during commissioning?

 

[PHovnanian]

Additionally, is it normal to combine a harmonics and PFC capacitor design or are these two very separate issues typically dealt with by the installation of two separate systems?

That actually makes sense. Just slapping some power factor correction capacitors on a system with harmonic loads can cause problems.

 

[VTer]

Look up displacement and distortion power factor. The total PF is a combination of both. Depending on the system, the consultant may be designing the harmonic filter to improve the distortion PF therefore improving the total PF. I do not fully understand the comment on the 12th harmonic. Typically even harmonics are not concern in power systems. 5th and 7th are a typically highest magnitude harmonics with six pule rectifiers with 11 and 13th being less significant. If you have a 12 pulse rectifier then 11th and 13th will be the larger magnitude harmonics. I believe that significant 12th harmonic would imply that the waveform is not symmetrical around the axis with some sort of DC offset.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic ù and this we know it is, for certain ù then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". û Nikola Tesla

 

[magoo2]

Rectified loads generate harmonics with the most common significant ones being 5th, 7th, 11th and 13th. Harmonics are drawn to capacitors because they offer a low impedance path. So plain capacitor banks and rectified loads quickly lead to problems with overloading the capacitor banks and high voltage. I don't believe you'd get any 12th harmonics.

What you'd like to do is tune the capacitor banks to trap the main harmonics. A 5th harmonic filter is simply a capacitor bank with an inductor which tunes it to the 5th harmonic. That usually takes care of the 7th as well. You also have another harmonic filter tuned to the 11th harmonic to deal with the 11th and 13th harmonics.

The capacitors used in filters normally have a higher voltage rating than those for plain capacitor banks because the harmonics will be drawn to the capacitors and as I indicated before this will raise the voltage. So it is standard practice with harmonic producing loads to have harmonic filters to trap the harmonics and raise the power factor.

The trick now is to see what system voltage is supplied to the load. If it is 11 or 12 kV, then I suspect you'll have a noticeable voltage rise as you indicated. The other possibility is resonance which depends mainly on the characteristics of the main transformer and the capacitor ratings in the filter. To deal with these two issues you'd have to get your engineer to supply a lot of the system information.

At worst, you might have to go with a smaller filter and live with a lower power factor. There are more exotic devices to explore like an SVC but these are much more expensive and require a thorough system study.

I apologize if I put too much engineering terms in my response. You might mention what part of the world this is in.

 

[magoo2]

Looking back, you did mention DRC, so I got my answer. The Congo.

 

[MatthewDB]

A passive harmonic filter does produce VARs and will help power factor. But you've found out the problem with this is the inability to adjust the PFC correction.

Typically only a fraction of the total PFC comes from the tuned filters. The rest comes from de-tuned cap steps under control of a PF regulator. (De-tuned banks are tuned below the lowest harmonic, often 4.3.)

I also don't get the 12th filter.

As for the generator: Typically the PF controller has an offset when on generator. The controllers have time delays to reduce cycling of steps. When holding .95 PF they can go leading at times. To prevent problems on generator, the target will be set down to something like .85 using a contact or modbus input.