Does anyone have any experience with trending VARs against changes in temperatures on a high-voltage high-resistance connection (such as a 500kv manual disconnect switch)? How can changes in VARs result in changes in hotspot temperatures, even while the resistance and current are being held constant?

I have never trended reactive power, but I do know that if there is an issue with it then temperatures will increase. Bad power factor will cause an increase in current and therefore any overheating components should increase in temperature even more. Normally, it is best to eliminate power factor issues anyway as they will cause an increase in costs.

Bob,

This is actually being measured in the substation of a power plant, so the power factor issues are a function of all the loads on the system (and, as you point out, those customers with poor power factors are being charged). However, in this instance, with the load (amps) remaining constant, the VARs are changing and the temperature seems to change with VARs in a constant load. Anyone else monitoring power plant substations seen this?

Rich Wurzbach
Maintenance Reliability Group, LLC

I have never heard of a specific dependence on vars.

The very tricky thing about power plant substations (as you know) is that you usually have multiple parallel paths for current to flow.

Sometimes determining the current in your hotspot can be very challenging, since you may have to use a vector calculation involving multiple lines depending on where you get your data from.

Some people have suggested that the resistance of the hot connection itself can significantly influence the current sharing between parallel paths as it changes resistance. I am a little skeptical that this effect would be significant since the connection impedance is resistive while the impedance of those large substation buses is primarily inductive. But you never know.

A few things come to mind:
1 - Are you sure you know the current directly through this disconnect and that it is remaining constant?
2 - I find it a lot easer to believe that the current distribution among parallel paths (either the normal distribution or the abnormal distribution created by the high resistance) might depend on the power factor in some way than to believe that a hot spot carrying constant current is affected by the power factor of a constant load current flowing through it.

More details would be interesting. Is it the VARs of the generator that are changing or the VARs directly associated with the connection . Is there a daily cycle associated with the change in vars? Perhaps you can describe how you determined the curretnt?

This message has been edited. Last edited by: electricpete,

Pete,

Thanks for your reply. This is actually a main generator disconnect, 3 phase, on the line from the generator step-up transformers to the substation ring bus. The amps load is being measured at the high-side of the main transformers. The VARs are measured at the generator.

Because the disconnect contact is made up across a set of 4 contact fingers, three is definitely the possibility of variations in total current flow between each of the fingers. In fact, one of the fingers is significantly hotter than the other three. My initial thought for temperature changes was in the variation of the contact resistance across the fingers, varying with thermal expansion, arcing past high resistance junctures, etc at the fingers. The plant, however, seemed to see a correlation between variation in VARs at the Generator that corresponded to the temperature changes. I remembered that the first first IR problem I discovered was on this same disconnect in 1990, and the substation engineer at the time had me plotting my temperatures against generator VARs, claiming that VARs changes would influence the temperature rise. I never understood that at the time, and haven't encountered many discussions about it since. But since I'm hearing engineers making this claim again, I want to see if there is some technical basis to it.

Rich Wurzbach
Maintenance Reliability Group, LLC

Aha. I was thinking of a disconnect associated with a breaker in substation. My discussion of parallel paths doesn't apply to the line from the generator (except possibly your idea about parallel fingers).

I honestly don't see any physical mechanism for the disconnect to "know" that the system power factor is changing, if in fact the current is the same.

Think about power factor, it is a phase lag between voltage and current. The disconnect contact surely knows that current, but what does it know about the voltage? Absolutely nothing! The voltage we're talking about is associated with the voltage difference to those other phases 10 feet away. Phase A disconnect contact doesn't care about the voltage at phase B. Note there is also a small voltage drop accross the contact itself (line side to load side) but that's not the voltage we're talking about. The angle between that contact voltage drop and the current is fixed (determined by the angle of the miniscule impedance of the contact itself). That voltage drop is miniscule compared to the system voltage and completely unrelated to the vars flowing in the system.

In summary, I cannot imagine a physical explanation why change in vars causes a change in hotspot rise for a constant current.

But some other possible angles:
1 - Is it possible that current is not truly constant, but real power (kw) is constant. Than as you increase the kvars, the current could increase slightly which of course could affect the hotspot.
2 - Is it possible that the system var loading changes to which the generator is being asked to respond have a daily cycle which might follow a daily cycle of wind or something?