Power Generation: Site Design, Installation, and Connection

[Genset]

Hello all,

I have submitted this question to another individual by PM, but wanted to post it in the forum as there may be others with experience in this regard.

I understand the issues and effects surrounding the use of standby/prime generators to support high percentage, non-liner (SCR based) loads. I have also read much about possible mitigation techniques (de-rating, adding a small resistive load (20% of total load requirement) etc) and that other inductive loads such as regular air conditioning motors may help to offset the effects.

The most sensible mitigation technique seems to be the addition of a three phase sensing True RMS AVR and also to filter the field voltage from the AVR, this dual solution prevents errors in sensing and enables the AVR to retain accurate control of the generator output.

I have access to Mecc Alte UVR 6 (3 phase sensing True RMS AVR) AVRs but am unaware of any filtering unit I can place in line with the field voltage wires, does anybody know of an AVR model/type which contains this as a feature, or if a separate filter for the field wires is available elsewhere? 

Finally, if a domestic customer has no three phase motors in his home, but chooses to install a three phase distribution board to take advantage of a three phase utility supply;  if his load requirement contains a high percentage of non-linear loads, would it be to his advantage (in respect to the generators ability to support the load) for him to choose a single phase generator over a three phase generator and if so, why?

Thanks or your input...

[mikel]

In general most AVR's supplied have limited ability to work with high harmonic loads.  Best mitigation technique is determined by actual distortion of the voltage waveform being sensed by the AVR, and resulting AVR response.  When power reduction modifications started being applied several years ago to sites where I worked, usual complaint was after this kind of work first test run on the generator resulted in over or under voltage and customer asking what was wrong with the generator.  Lighting retrofits and VFD installations usually being biggest culprits.  After struggling with this for awhile had a consulting engineer who got me started using an o-scope to look at the sensing coming into the AVR.  Once we determined where and how much the voltage sine wave was being distorted we could then figure out a could way to correct it.  In some cases we applied filter caps to the sensing lines, worked ok but in some cases affected transient response.  When CAT introduced the VR2 AVR (old card type) it was VERY tolerant of harmonics and performed very well in providing stable outputs on high harmonic loads. But it was expensive and hard to troubleshoot.  For many years I used Basler SSR regulators to replace not only CAT but many other manufacturer AVR's with good success.  The CDVR has better harmonic capability than most earlier CAT AVR's, but still has problems in some applications, although usually some tuning could make it work in most cases.

Hope that helps

[par38lamp]

We had a C18 that suffered from high harmonics when the customer went from a 12 pulse to a 6 pulse UPS, which was also the primay load on the generator.

We eventaully sapped the R448 voltage regualtor to a CDVR and solved the problem.

The next step was to provide 1:1 isolation x-fmrs on the voltage sensing leads, and even a cap in parallel to create a LPF and filter out anything above 60Hz.

[wtc2008]

A PMG will be helpful and select RFI between sensing lead and regulator.

The old VR3 in CAT genset features 20% harmonic tolerance.

The CDVR features 40% THD.

The stamford or LEROY SOMER should provide AVR with RFI function.