Well, first let me say, that I 100% concur that the primary use of the tap changer is to correct voltage.
That being said, in the case of our facility, we have a small power plant with a 7.5 MW generator, tied in with the city utility.
So no matter what our generator is making, relative to what we are buying, the VARS demand is the same of course as long as load is constant.
Now, the utility, as miserly as they are don't want to supply any more VARS than they have to. So our load tap changer (LTC) is controlled by a VARS controller. The VARS controller does NOT tap relative to voltage. What it does though, is tap to maintain the utility supplied VARS at less than 1 MVAR. And because of this small voltage shift, if the LTC drops city supplied voltage, in Isochronous mode on our generator (if I remember correctly) the generator will then assume VARS from the city. On the triangle, if you increase VARS, you increase VA, and therefore lower PF, for our side.
So one might say it is a form of PF control for the city, in this instance because it's sole purpose is to lower VARS for the utility, which increases PF, albeit a very small amount because of the total load they have. But it does noticably affect our PF.
An example of the voltage drift occurs at night when demand decreases. The LTC being in a happy steady state mode with <1 MVAR on it will see VARS decrease as evening progresses due to load shutdown, which of course causes voltage to drift from say 7400V up to 7500V (on a 12.5KV system, our meters must read phase to ground?). System voltage will stay there until one of several things happen.
1. Daytime load requires enough VARS that the LTC must correct, thus shifting voltage back down, but not necessarily to the normal 7400.
2. An operator intervenes to adjust voltage manually using the LTC, but we never do this.
3. An operator "drives" votage down using the voltage regulator on the turbine. Note that this does not directly affect voltage because the city determines that. However, the lower of field current in an isochronous generator will lower VARS on the generator, which will migrate to the city, which may or may not cause the LTC to move. Sometimes you need to send a good amount of VARS to the city to do that.
In the end, the LTC will tap sending VARS back to our generator at a new steady state voltage.
Please keep in mind that this is based on what I believe is the correct application of power factor. Of course, as one of my co-workers likes to point out, I am wrong from time to time, so if I am wrong somewhere in here, I'd like to correct my understanding.
[This message has been edited by rat4spd (edited 09-18-2005).]
