Line Drop Compensation 2

[powersmp]

We have 2000kft of 750 MCM (UG conductor) and 6000kft of 556 AL. It's a 3 phase, wye connected regulator set at a base of 114/110. Currently it's set to X=0, R=2 with a ct ratio of 400/.2 and pt of 120/1. It runs at 13.2kV. It's a 750/937 KVA SFR reg. After all was said and done, I calculted it out to be about X=15, R=4 for my compensation to produce the 114/110 V base where needed. My questions are:

Why does the regulator currently have more bucking than boosting and allow people to have low voltage at heavy loads with even a R=2 volt line compensation?

Why does everyone say that we shouldn't set X compensation? There is no straight answer other than "it's too hard to understand" and "because of cap banks fuses blowing we try to avoid that due to t he sensitivity of the X compensation".

Any help would be greatly appreciated.

 

[dpc]

Setting the line drop compensation basically just moves the point being regulated farther down the line, based on the R and X settings. Normally, this is set around the mid-point of the line in terms of load.

We have seen systems where the substation voltage was 13.8 kV, the regulator nominal voltage was 13.2 kV and the actual feeder voltage was 12.47, so the regulator is always in buck mode regardless of the load.

We also see a lot of regulators with NO compensation of any kind, because it's easier.

If the regulator has "First House" protection turned on, this will limit the amount of voltage boost regardless of the compensation settings.

What type of control do you have?

 

[ePOWEReng]

From my experience a typical setting for the feeder your describing would be R=2 and X=5, but that varies largely between companies and philosophies. DPC brings up a good point...if your VR has first house protection it allows you to be more liberal with your settings and not overvoltage customers that are close-in to the substation too much. In general, if you have a good model of your feeder you can approximate the voltage profile with different settings in your voltage regulators and see how it affects your voltage profile.

 

[ePOWEReng]

Sorry for the rushed typing...what I meant was playing around with typical loads, pfs, and possible VR settings is a good way to increase your confidence in your VR settings.

 

[powersmp]

It's an express feed, so there really are no customers close to the station (not within 8000 ft). I'm pretty sure we don't have a "first house" protection on because I talked to the individual that sets the settings and he said he doesn't know what that is, and I've never dealt with it engineering wise. Unless it's set on by default, I don't have it on. I would like to have no compensation, but in this case we need the load compensation since we tend to see low voltage once the load starts getting high.

But I'm still wondering why they would shy away from setting the X compensation? Any thoughts?

 

[jghrist]

Assuming that you really meant 2000 ft of UG and 6000 ft of OH instead of kft, I calculate an impedance of 0.304 + j0.802 ohms. With a multiplier of 400/120 = 3.33, the R and X settings would be R=1 and X=2.7.

I assume that by a 114/110 setting, you mean a voltage setpoint of 112 with a 2 volt bandwidth.

With 13.2 kV at the station (110 volts), you would need to boost the voltage at the station to allow for voltage drop on the line.

If you set X=0, you will undercompensate for the line voltage drop and get lower load voltages. The load voltage will vary depending on the load power factor because the regulator setting X/R ratio is different than the line X/R ratio.

 

[powersmp]

It should be a 2000/120 ratio. The station CT ratio is 400/.2 or a 2000/1.

You're suggesting not setting X is going to undercompensate for the line loss. Is there any way to raise the R to compensate for that? I'm told the X setting is too sensitive when cap bank fuses blow on the line.

 

[jghrist]

If you're using a Siemens MJ-X control panel, the compensation multiplier is k=1.0x(Current Transformer Rating/Voltage Tranformer Ratio). See Appendix D, Section D.4 of the control panel manual.

The capacitor bank will change the power factor of the line load, not the line impedance. If the R and X settings match the line impedance, then the load drop compensation will be correct regardless of the power factor (assuming that the capacitor bank is at the load center). You could raise R to compensate for setting X=0 for a particular load, but it would not give correct compensation for other loads. I don't see any reason for setting X=0. Can you get "everyone" to explain what capacitor fuses have to do with the X setting?

 

[powersmp]

jghrist - your question of trying to get them to explain it to me is exactly my problem. A lot of people here seem to just follow what's been done and don't try to figure out why. My understanding of the compensation is similar to the others that are posting, but I'm being told that it's not that way by people in a more senior position than I am. I was told that it was a 400/.2 or a 2000/1 ratio the same way I was told that we don't set X compensation.

I started this post simply because the things I'm being told was going against everything I thought and just wanted to get outside views on it because the answers I'm getting aren't making sense.

The initial setting though is a 114 volt setting on a 110 base with a 1.5 bandwith. You were right about 2000 ft instead of kft, sorry.

If there's anything else that you guys can say or suggest so that I can convince them to set the X value, that would be greatly appreciated. And thanks for the note about the K value.

 

[dpc]

I think the reluctance to set the X compensation is based on scar tissue acquired from past problems of overvoltage when the settings were not correctly made. Since the X component of the line impedance is generally a lot greater than the resistance, X is normally set higher than R and dominates the operation of the line drop compensation.

The voltage drop changes as the power factor of the load changes and you need both R and X settings to get it to operate correctly.

"First House" protection is found on newer regulator controls and allows a maximum voltage limit to be set that will override the Line Drop Compensation to avoid overvoltage at loads close to the substation.

 

[jghrist]

The only thing I can suggest is to get a copy of the control panel manual and go to them and say "I really can't figure this thing out - please explain to me why the setting shouldn't be like it says in the manual."

What is the system Ø-Ø and Ø-N voltage?

 

[tinfoil]

You should talk to the regulator mfr (if they are still in business). They are often VERY helpful (particularly if you hint that you can get THIS one working, you might be interested in more units)

Some manufactures want the X and R values in ohms, some use some cryptic pseudo-P.U. thing that is impossible to figure out on your own.

The reason why the X bit is left out is becasue it is 'hard' to compute properly. As already mentioned, this is almost always the dominant factor for compensation, and if you do not apply it correctly, you will get wrong results. It is for this reason that some POCOs avoid it altogether.

But it IS worth the trouble to accurately model your line impedances/capacitances to get your X values. If you follow consistant construction standards, you only have to work this out once.

 

[apowerengr]

Regulator buck or boost is not just a function of your settings but also transmission line performance and power transformer no-load tap settings. Your utility might set its no-load tap based on some transmission line contingency where they know the input voltage to the substation is 5-10% below the "system-normal" voltage. If this is the case, then normally the regulator would have to buck in order to output the voltage you programmed the control for. In other words, a bucking regulator does not mean low voltage.

Strategies for R and X settings vary greatly and some examples were presented in earlier postings. If you have low (not just lower) service voltage on an express feeder during high-load times you can either raise the base voltage setting normally, increase the R and/or X compensation settings, add capacitors close to the load, or add an additional stage of regulation near the load. Recognize the X setting for overhead lines is a function of phase conductors size and spacing, if the line has several different construction types in the overhead segments you need to take the time to make detailed calculations of the X component of the impedance. Also, load power factor does affect the regulator output voltage if the X compensation is set to something other than 0 volts, something which was incorrectly stated earlier. So, large point loads with large reactive load swings or switched capacitor banks can affect your voltage regulator's output voltage if you use X compensation - something not many people want to do as there are too many variables to try and troubleshoot when you have a problem. Ask one of your peers if they have a copy of GE's old Omnitext for voltage regulation - the discussion of Line Drop Compensation should clear up any additional questions you might have.

 

[jghrist]

So, large point loads with large reactive load swings or switched capacitor banks can affect your voltage regulator's output voltage if you use X compensation

This is exactly what load drop compensation is supposed to do. If the R and X are set to the R and X of the line (with the proper compensation multiplier), then the line voltage of the regulator will increase or decrease such that the voltage at the load center will stay at the voltage set point (subject to bandwidth) regardless of the load. If the load power factor increases, say because a capacitor bank switches on, then there will be less voltage drop in the line and the regulator line voltage will reduce to compensate for the different load.

It may be simple to set X=0, but it isn't a very good approximation of the line reactance. The load voltage will change with load swings if you set X=0 (or any other bad approximation of the line reactance). If you want to use load drop compensation, why not set it so that it works as designed? There are good arguments for setting R=0 and X=0 with the setpoint voltage equal to the maximum allowable. With these settings, you will attain maximum possible voltage at all customers, but will have large voltage swings at the load center.

 

[apowerengr]

Powersmp - at 20 MVA of feeder load (460 amps) you only have 2.5 volts drop on the primary at 120 V base. If you want 114/110 basis at the customer set R=X=0 and set your base voltage at 116.5. Was your question about bucking regulators and low service voltage answered? You have never specified what "low" voltage was and where it was measured. There are some causes of low service voltage which regulators are not designed to correct - perhaps yours is one such instance?

jghrist - 2.5 volts does not look like a large voltage swing at the load center. Setting any level of reactive line drop compensation can cause voltage regulators to do unintended things from improperly operating capacitor banks for example. This causes service voltage imbalances which as we all know cause motor performance issues. That's one historical reason why many utilities don't use it; cap banks don't get maintained, fuses blow, oil switches hang up or fail and the results are service voltage complaints. At least that's been my experience.

 

[powersmp]

Thanks for all of the feedback. The "low" voltage we're seeing is 5% less than the service voltage we state in our service agreement, which is 120V. So, at the customer, we are seeing less than 114. Actually, their service is 277/480V and they're seeing 260V. This has only happened recently, and there really hasn't been any kind of load shift, but the voltage at the end of the express feed has been noted to be 118V - which is a 6V drop at heavy load (420 a in this case)