Distribution Transformers in same Area blowing same phase 2

[hd2000]

Greetings all, was normally browsing here as a guest but have now decided to join this great group

I have a problem in a distribution network I maintain. It is in an area with:

Main 33kV D-D step down to 11kV sub-station(fuse protected both sides HV & LV)

Then 11kV outgoing auto-recloser (1 feeder)

Then multiple 11kV/0.4 D-Y transformers on the single feeder.

In this area we have constant blowing of ONLY yellow phase HV fuses on the transformers, sometimes even on the main 33kV (about 1 a week) on different TXs each time without a set pattern

What could be the possible problem in the system to cause this? The main 33kV TX has been changed several times also.

 

[GEORGE2311]

What are the transformer kVA and vector groups? What are the fuse ratings? Do you have any disturbance recordings from a protection relay? When you say hv fuses on the transformers you mean the 11kV side of the 11/0,4kV trafos? It seems very interesting.

G.F.

http://www.engineering-services.gr

 

[waross]

Have you tried a scope on the problematic phase? The old discontinuous ground fault comes to mind.
Are your transformers connected so that excess current through a lightning arrester will blow the fuse?
If you have something generating transients on that phase, the lightning arresters may be discharging and blowing the fuses.
Alternately, if the voltage is low or high on one phase, motor loads will try to correct the voltage and may cause excess current to flow.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

Are there any grounding transformers on the 11 kV?

 

[hd2000]

Thanks for the feedback

There a several transformers on the feeder, a mixture of some 50kVAs, 100kVAs, and 200kVAs. There are all of DYn-11. There is no "intelligent" protection on the system. It is just a fused sub-station. Yes the HV fuses on the 11kV side of the transformers keep blowing yellow phase. The fuse ratings have been checked several times and are correct.

We have not yet tried a scope, but I think that sounds like the way to go forward. Yes all the transformers lightning arrestor are connected to that the fuse will blow as it arrests. I was actually thinking of changing all the lightning arrestor in the whole area. Transients on the phase sounds right, but I cannot think of what would be causing such. True about the motor loads, one phase might be out high/low...

No grounding transformers in the system, it is a very straightforward distribution network. I think putting a scope on things sounds like a good way forward for now.

 

[Skogsgurra]

I would recommend a transient recorder instead of a scope. That way , you can set triggers in many ways that usually cannot be done with a scope. You will also have all events stored for later review - much like in a modern protection relay. Also, a scope seldom has more than four channels and you need at least six - three voltages and three currents.

There has been some discussion about "sympathetic inrush" here lately. The mechanism is that a DC load (the inrush component) affects neighbouring transformers so they also get more or less DC saturated and trip the primary protection.

In your case, it probably isn't about "sympathetic" inrush, but a DC load somewhere that "contaminates" the system.

I have had such problems with thyristor soft starters. When the thyristors don't trigger correctly you get a heavy DC component in the transformer so that it saturates and primary current goes way beyond expected inrush.
Primary current once got so high that the primary proection current coil caught fire. It was like living in Lucas Land - complete darkness.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Marmite]

With an ungrounded system you could have all sorts of stray voltages developing between your system and true ground which could cause insulation to break down. Are the transformers designed for use on an ungrounded system? It's questionable whether such a system is safe in any case bearing in mind how primitive your protection is. How do you detect a downed conductor?
Regards
Marmite

 

[racobb]

Have you verified you arrester ratings on the Delta portion of the system?

Alan

 

[hd2000]

Yes a transient recorder sounds like the better option. The DC component sounds interesting, I will have to investigate fully the equipment connected in the network.

Marmite - Sorry for the confusion I must have caused. All the transformers have solidly earthed neutrals. I was somehow thinking of a specific special type earthing transformer. My bad...

Not sure what the surge arrestor rating are exactly, I guess we will have to take the whole feeder out and have a crew analyse all the surge arrestors. +/- 30 transformers.

I'm a bit new in the power systems control area so plz bear with us newbs

 

[GEORGE2311]

As far as i understand the 11kV network is ungrounded and delta connected (are you in an old cement factory?). The load flow is from the 33kv to 0,4kv and there is no power generation at low voltage side.

If the loads (motors or whatever) are connected in LV side then all the load asymmetries will not pass to 33kV side of the transformer as the triangle windings will equalize/trap them.

Any dc disturbances will stay at the network where the source is.

Any leakages to earth of the 11kv network are not able to burn any fuse in the 11kV network or the 33kV. How the fuse in this side were burned? Do you have any voltage transformers? (perhaps they are oscillating/use damping resistors)Do you have transformers not identical to each other in parallel permanently or for a small duration during any process?
The most possible situation is that you have over-voltages in the ungrounded network which causes unwanted operation of the surge arrestors. Check all the voltages of the system and take recordings.

G.F.

http://www.engineering-services.gr

 

[Skogsgurra]

Since Sympathetic Inrush has been mentioned here and in other threads, I made a test to see what that animal looks like. The recording was made using a couple of small (250 VA) tape-wound toroids. So, the circulating DC decays quite fast. But it shows what is happening in the two transformers very clearly:

http://www.gke.org/pub/files/Test with sympathetic inrush current.pdf

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

I had the thought that the sympathetic may be a negative step voltage rather than a positive step. (Reduction of voltage in the victim rather than an increase.) I never took the time to do the math. Thanks to prc's excellent description and your esting, Gunnar it all comes together for me.
Thanks again prc.
Gunnar; May we assume that the relatively higher reistance of the small transformers contributes to the rapid decay of the DC component? It is my understanding that the decay with power transformers is more rapid with greater separation as a result of the resistance of the interconnecting conductors.
Thanks for the test. lps
If you are still set up, it may be interesting to test with the two transformers fed from a small transformer. 500VA to about 1500 VA.
Yours
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[hd2000]

Thanks Skogsgurra that is a very nice explanation. Plz give me a couple of weeks to get transient recorder and set it up on site. Well get back to this thread then. Thanks alot guys!

 

[busbar]

Do a Google search on transformer ferroresonance. At least it may be something to rule out.

IEEE std C57.105 §8 has some basic information on transformer susceptibility to ferroresonance.

 

[GEORGE2311]

hd2000 do you have any news of the analysis?

G.F.

http://www.engineering-services.gr

 

[hd2000]

Nothing yet, waiting to install the power analyzer, hopefully this week.