33kv/231v Transformers Are Burned Out

100 km is about 62 miles. Thats a long way. How is the lightning in this area?

Is the 33 kv system a wye or delta system? sounds like a wye system.

Check the name plate on the transformer to verify the voltage 33 kv/230 volts.

Have you checked the load on the transformer?

Is the 100 kva transformer connected at the end of the line? Are the other transformers connected at the end of the line?

If this area has a high frequency of lightning, transformers at the end of the power line will experience a higher failure rate.

All these transformers are stepdown transformers and having voltage 33000 at primery side 231 volts at lowvoltage side. thease transformers are delta /wye,

and this is dessert area in saudi arabia, here you can find rarely lightning,as wel as rain, at presently there is no rain or lightning( all this year)it is the completly dry land

1- Are the transformers protected by properly dimensioned fuses?

2- Are there any power factor compensation capacitors installed in the villages or onto the line feeding the villages? These can increase the line voltage dramatically at low load if not switched off.

3- Are there any very large source of harmonics? Could produce resonances with capacitive loads.

4- Most likely: Is there a local "inventor" that uses a single diode DC rectifier to draw DC current from the line? (DC current is not measured by the meter and is "free"). But even low DC current level saturates the transformer magnetic core and multiplies its core losses. You can "burn" a large transformer just by drawing a few amps of DC current. Note that a "failed" rectifier in an AC or DC drive can also draw DC current.

Hi

How are you, hope u doing great.

1- Transformers protected by properly dimensioned fuses.

2.No capacitors are installed on the line at all.

3- ??????Are there any very large source of harmonics? Could produce resonances with capacitive loads. ?????? Please define me the Sourceofhormonics,

4. There is no any local inventors.

All these villages are very small villages, few houses are scattered here and there, with little load consumtion,at each transfomers. please check out this eartyhing system we installed. and their resistance measured 13 ohm, is there any problem?

Check the load on each leg of the low voltage side. It is possible, with a small village, all of the load is being drawn from a single phase...

Cosumers are using both valtages 231volts and 110volts, for lighting(110volts) and air conditioners 231volt, each phase of the transfomer have 40-30-55 amps of load, I mean R-40, Y-30, B-55, Amps

If as the man said the transformers are protected by properly dimensioned fuses pulling too much load on one leg will simply blow a fuse.

j.

Sir, harmonics are caused by non linear loads such as motor drives, battery chargers, UPS, computers, power controllers. It takes a lot of harmonics before damaging a transformer. I would estimate 50% THD at minimum. What is more likely is that one of the above mentioned load has one diode blown and is drawing DC current. This can be checked using a DC clamp meter at the transformer secondary. DC current also makes a transformer very noisy.

I am not an expert in high voltage distribution. I am however a power electronics engineer used to deal with 0 to 1KV. As far as I know, the grounding is only useful to protect the low voltage side during ground lightening strikes. In normal time we could do without as it is done in high resistance grounding systems. We do however need it for human safety.

Viewed from the low voltage side, the ground grids are not likely to have any effect on the transformer reliability. My experience on high voltage transformer failures has been from core heating from DC current and insulation failure caused by eddy current heating of the windings. Each failure mode has a very distinct visible signs on the transformers. Have you inspected the failed devices? Did you ask the manufacturer for an evaluation? They are the most likely competent people that will be interested in figuring out why their product is failing.

Marcot,

We are here talking about transformers taking a 33kv alternating current line and dropping it to 220/110.

You talk about someone with a blown diode circuit "drawing DC current."

The alternators at the generation station do not produce any D.C. to draw.

What you are saying is not comprehensible. Would you please explain. What is the source of the D.C. that you see someone might be drawing from that line?

Nonetheless, if indeed D.C. were being fed to those transformers, as in theatre lighting systems where reactance dimmer transformers were used as a means of lamp dimming, bleeding D.C. into the transformer simply saturated the core and shut down A.C. flow.

How could that damage or short out and burn transformer windings?

j.

j.

It doesn't matter if DC is injected at the primary or secondary. It doesn't matter either what is the size of the transformer. As soon as a sizable amount of DC current flows into a transformer (usually from the load), the core saturates and the losses can increase up to destruction.

If you want a more technical explanation here it is: DC bias pushes the excitation curve (B/H curve) to one side or the other depending on the polarity. As you move sideway on the B/H curve the slope of the magnetization curve is reduced (saturation). This also reduces the magnetizing inductance seen at the primary. A lower magnetizing inductance means a higher magnetizing current and higher losses. The B/H curve also becomes wider which increases the losses each line cycle.

Now, I have identified the potential sources of DC current from the load side. If you cannot measure any DC current then it is something else and forget about this suggestion. I am simply trying to help you consider potential causes for these "strange" failures. Check them and if it is not that just move on.

By the way you said "where reactance dimmer transformers were used as a means of lamp dimming, bleeding D.C. into the transformer simply saturated the core and shut down A.C. flow." It is actually the other way around. Injecting the DC current saturated the core and allowed the AC flow. This is exactly what I was explaining.

Marcot,

You posited a device drawing D.C. from an A.C. circuit itself devoid of a D.C. source. Moreover you posited this device as a rectifier that held a blown diode.

You still have not told us where the D.C., under the circumstances you set out, comes from especially so since you posited that the device DRAWING D.C.. was a rectifier with a blown diode, i.e., no longer capable of rectifying the A.C. source so as to have D.C. to feed back into the circuit, as opposed to drawing such.

So where does the D.C. you say is causing the problem come from?

Further, D.C. saturation of transformer coils is, as I explained, an old technique for controlling A.C. current flow through a reactance transformer. Years ago it was used, and in some old houses probably is used, to control theatrical lighting. We, at Brooklyn Tech with its four thousand seat auditorium, used to have such a lighting system and we could take a console and plug it into the system in many places in that house and with tiny potentiometers control the A.C. flowing through those large, probably five thousand watts, reactance transformers.

That D.C. did not cause the burnout of those transformers. There were no losses to burn out the transformers, there simply was no current flow since like any choke in a circuit, a rectifier circuit for instance where the choke serves to smooth out the peaks remnant of the rectified A.C. precisely because the D.C. saturates the choke core presenting an impedance to the peaks or vice versa if you please we may have a language difference.

I would suggest that these are really very simple functions that you should be able to explain without what to my unschooled mind, I'm just a simple electrician with wide industrial and some theatrical experience, sounds like mumbo jumbo.

All you need do is explain the principles involved and why the result is burned out transformers and why you assert, although practice in industry is to use the core saturation effect of D.C. flowing through a winding around an iron core (Saturable reactor) for control of A.C. circuits, in this case it is burning out the transformers.

Further, why if what you assert is so, i.e., D.C. from somebodies connection of a device causing a D.C. saturation in one transformer is in fact burning out numerous transformers. If it is D.C. it is going to be limited to that one transformer secondary.

I am not the originator of the query. I signed onto this exchange because I found it an interesting problem. Nonetheless, I don't see your point since transformer like devices, saturable reactors, are in general use and feeding D.C. into the D.C. winding in order to control A.C. current does not burn out those devices.

I don't see how what you have contributed is in any way related to the problem as stated especially insofar as you posit a D.C. source that seems to not have a source in the circuits under consideration nor that if D.C. did get into those circuits it would cause the catastrophic failure of the trannies.

Again! Please explain!

j.

OK Jack, this is the last time. Here is a specific answer to each of your previous questions. If you need more details, look for magnetic amplifiers on Wikipedia or read a book.

Your questions are in Italic and my answers in normal text.

• You posited a device drawing D.C. from an A.C. circuit itself devoid of a D.C. source. Moreover you posited this device as a rectifier that held a blown diode.
• You still have not told us where the D.C., under the circumstances you set out, comes from especially so since you posited that the device DRAWING D.C.. was a rectifier with a blown diode, i.e., no longer capable of rectifying the A.C. source so as to have D.C. to feed back into the circuit, as opposed to drawing such.
• So where does the D.C. you say is causing the problem come from?

A DC rectifier for single phase or three phases AC line has an even number of diodes or thyristors... They are arranged in such a way that the same current is drawn from the two half of the AC cycle. This produces very little DC current in the AC line as all the diodes are very similar. Now, if one of the diodes fails open, a different amount of current will be drawn from the two half cycles of the AC line. This difference is a DC current mixed with some AC. Depending on the size of the rectifier and its load with respect to the transformer, this DC current can be negligible < 0.1% or become dangerous > 1%.

• Further, D.C. saturation of transformer coils is, as I explained, an old technique for controlling A.C. current flow through a reactance transformer. Years ago it was used, and in some old houses probably is used, to control theatrical lighting. We, at Brooklyn Tech with its four thousand seat auditorium, used to have such a lighting system and we could take a console and plug it into the system in many places in that house and with tiny potentiometers control the A.C. flowing through those large, probably five thousand watts, reactance transformers.

I am fully aware about the operation of magnetic amplifiers. I designed one many years ago. As I said in the previous post, your understanding of their operation is incomplete. I will try to fill in some gaps.

• That D.C. did not cause the burnout of those transformers. There were no losses to burn out the transformers, there simply was no current flow since like any choke in a circuit, a rectifier circuit for instance where the choke serves to smooth out the peaks remnant of the rectified A.C. precisely because the D.C. saturates the choke core presenting an impedance to the peaks or vice versa if you please we may have a language difference.

To begin with, a magnetic amplifier uses a core with characteristics very different from your standard distribution transformer core. The square B/H curve of a magnetic amplifier was optimised for the switching applications. The level of AC magnetizing currents were relatively low in the OFF position and extremely high in the ON position. They operated in switching mode like a modern semiconductor (thyristor). This minimized the losses. If you try to do the same thing with a normal transformer, you will not obtain a clean switching and it will operate in the linear mode. If we refer back to the semiconductor example, the losses in a linear amplifier often reach 50% of the total power flow.

• I would suggest that these are really very simple functions that you should be able to explain without what to my unschooled mind, I'm just a simple electrician with wide industrial and some theatrical experience, sounds like mumbo jumbo.

If you really want to understand those fairly intricate devices, you will need to read some electronics and magnetics design books. There are too many details to cover them here. Sorry for the "mumbo jumbo" but it will become clear once you understand the basic theory of power electronics and magnetics. You cannot go around it.

• All you need do is explain the principles involved and why the result is burned out transformers and why you assert, although practice in industry is to use the core saturation effect of D.C. flowing through a winding around an iron core (Saturable reactor) for control of A.C. circuits, in this case it is burning out the transformers.

A magnetic amplifier (not used anymore in modern applications except for some very specific devices) was usually a saturable inductor placed in series with a load. It acted as a fast switch with phase control ability. This was similar to a modern thyristor power controller. When the switch was on, the AC current flow was mostly limited by the load. The load was "protecting" the magnetic amplifier by limiting the AC current flow. Of course, there was a fuse for ultimate protection.

In the case of a transformer saturating. there isn't any load in series with the primary to limit the current flow. As soon as the core begins to saturate, the magnetizing current increases and so do the core losses. The primary fuses are dimensioned to protect a transformer with about 1% core losses. They are ineffective when the core losses reach 10%+. The only protection against this is a temperature switch that will disconnect the transformer primary.

• Further, why if what you assert is so, i.e., D.C. from somebodies connection of a device causing a D.C. saturation in one transformer is in fact burning out numerous transformers. If it is D.C. it is going to be limited to that one transformer secondary.

As I said before, this is just a suggestion. I have seen some people fooling power meter before by drawing DC current that is not metered and slows down the mechanical meter dial. This is easy to do and you only need a "clever" guy in one village telling is brother in the other village...

The real cause for the failures can be something totally different but this has happened before. That is why it is illegal to draw DC current from the AC line.

• I am not the originator of the query. I signed onto this exchange because I found it an interesting problem. Nonetheless, I don't see your point since transformer like devices, saturable reactors, are in general use and feeding D.C. into the D.C. winding in order to control A.C. current does not burn out those devices.

You do not know enough on this subject to argue against a power electronics engineers with more than twenty years of experience. The fact that you don't understand something doesn't make the other side wrong. Try to learn the basics and one day you will see the light.

• I don't see how what you have contributed is in any way related to the problem as stated especially insofar as you posit a D.C. source that seems to not have a source in the circuits under consideration nor that if D.C. did get into those circuits it would cause the catastrophic failure of the trannies.

You are the only one who can cure your blindness. Read a good book.

Regards,

Marco

What is burning in the transformer? What fails, the primary or secondary?

Some transformers are burned with primery coils and others are secondry coils,

How did you figure this out? We need more details, pictures etc.. if you want help.

Regards,

recently one more transformer burnt out, I check this transformer by the meggar for resistance and also check out the turn ratio of the windings,these are the readings,

KVA-1oo, Voltage HT 33kv, voltage LV-231,

Meggar readings,

LV to HV ground-250,200,230 Ohm

HV to LV ground - 230,220,250,ohm

HV to LV -0.1,0.1,0.1, ohm

Ratio test

142.66,149.13 and 256.

Aamersyed,

You have not told us your location. That may be of value.

As somebody with experience in industrial electrical work, maintenance, the fact that some burnouts were primaries and the others secondaries and without the kind of speculation such as Marcot's that folks up and down that distribution line are teaching each other to rip off the power company suggests the transformers themselves are at fault.

As I understand it they were installed by a contracting company.

Were they new when installed? Was there documentation as to their having been appropriately and properly tested?

Were they supplied by the contractor? Have you seen the paperwork as to their purchase?

If contracted was the work and the equipment warranted? In other words who is bearing the cost of their replacement?

Who manufactured those trannies and where were they manufactured?

The possibility is that somebody has dumped a bunch of bad trannies on you. That is far more likely than the esoteric speculations of an engineer about folks stealing power by drawing from the line piddling amounts of non-existent D.C.

What I am suggesting is that you do some careful backtracking and forensic investigation as to those transformers so that you can be sure as to their lineage and quality and that they themselves are not the source of the problem.

A diagram and perhaps a map of that distribution system, including the source of the 33kv might also be of value.

j.

Give us a single line drawing, and pictures, showing all connections to both the primaries and secondaries of the transformers. Don't leave out a thing no matter how trivial you think it is. Is the primary neutral connected to the secondary neutral? We also need to see more then one transformer on your diagram. Show the source and the provisions for grounding at the source as well. Show all of the wiring between the source and each of the transformers. As well.... we need to know they type and ratings of each of the transformers. Any of them connected in parallel on both the primary and secondary? What about any large consumers of power? What about an municipal utility type stuff

You're answer is there, you have not given us enough info to help you find it.

1. primary neutral is connected to the secondry neutral, I edited the upper attached diagram just check out that and tell me if there is any problem, Thank you

This is the Earthing system they made exactly

I can not make out the drawing. Getting old... the eyes are going.

Aamersyed,

You said that you are getting a reading of resistance to ground of 13 ohms. I have said previously that is to high. Here they look for 1 ohm but the literature says certainly not more than 5 ohms.

I suspect, as I said recently that you are getting bad transformers. That's one possibility given the burnout pattern, i.e., secondaries on some and primaries on others.

But, in addition, I would look at the resistance to ground in each case. The earthing layout, such as I can see from this diagram, seems to be appropriate.

It cannot be that in the instance of each transformer the resistance is 13 ohms. You might want to look at that and see if the burnouts take place where there is higher resistance.

You also ought to make the contractors do whatever it takes to bring the resistance at each transformer down to 5 ohms or less.

I am sure we would all appreciate knowing the resistance at each installation and how that correlates with the burnout pattern.

I did ask you for a map of the entire system. I think a specific wiring diagram for the system, i.e., H.T. lines, connections to the primaries, internal transformer diagram, etc., would be useful to all of us.

For instance you mentioned a primary center-tap???

j.

Aamersyed,

Reading back I see you did tell us where this installation is.

I also note that you say that earthing is at 13 ohms.

I am not expert in that area because here you drive an eight foot rod, often almost into the water table (They do hand dug wells in this region, Atlanta, Georgia, USA) and you don't even bother to measure, but my understanding is that Resistance to ground should be down around 1 ohm although research says as much as 5 ohms is acceptable.

j.

I recently had a very strange transformer failure, anecdotally related to a lightning strike in the vicinity of the transformer (I say anecdotally because I was not on site when the failure occurred, so have no idea what really happened, other than that the generator dropped off line due to overload condition immediately after the lightning strike. No obvious burning anywhere. I identified the failed transformer by isolating the various sections of the distribution system). Opening the transformer, there was absolutely no visible evidence of damage, no discoloring, nothing. When power was applied with no load however, I had no voltage on the low voltage side. Finally, a megger told me that the high side (delta) was shorted to ground, which most likely means there was a short in one of the high-side windings (delta connected) to one of the low-side windings (wye connected). No visible evidence whatsoever. I suspect faulty manufacturing, because if it was a lightning fault, I would expect significant evidence, and other transformers in parallel on the system to evidence problems as well...This was a different brand transformer than the one I normally use, although all brands to which I have access are generally fabricated in either Mexico or Costa Rica.

Oh, yes, this is a 3 phase 480 V Delta primary/208/120 V Wye secondary, 75 kVA transformer connected in parallel with 5 other transformers of different sizes, most smaller than the 75 kVA transformer. Shipping the 500+ pound unit back to the factory for analysis is not an option, since the installation is on an isolated Island without facilities for handling heavy equipment, and the cost of the freight is more than the cost of a replacement. I simply specified a different brand for the replacement...

It's great to see the passion of those with history in this area of failure. I'm not so familiar with the specifics of transformers, but from what I understand so far

The transformers are "burning out" (not exploding, not shorting)

There is no talk of fuses failing (or breakers tripping, or other fault protection devices)

The area involved is very dry and sunny.

The failure keeps hinting to "heat in the transformer" as the failure symptom.

From these I presume the failure is not a short circuit or incorrect connection as the failure in that circumstance would be "spectacular".

Now I ask "How could I get heat into a transformer?"

Yes, a DC bias on either side could cause that. Others have already talked this through.

High resistance windings could cause extra heat. (Measure the resistance of the windings and compare good to not good.) Catalogue them all (before failure) so you get some predictive information.

Is it possible that "The heat is not getting out of the transformer?"

Given the environment, what are the ambient temperatures, what is the capability of the heat exchangers for the transformer, are they properly ventilated, are they properly spaced, is there additional heat coming to them from other sources (like reflections off other surfaces, is the peak load at the hottest time of the day, is the coolant fluid in the transformers correcly functioning?

Like I said earlier, not exactly my field, so hopefully the ideas are not ridiculous but looking in from "outside the box".

The

Engineer,

Sometimes observers from out of the box pick up on stuff others hadn't.

Your point about heat is well taken. I would think that transformer installations in an area in Saudi would be set up for hot conditions, for instance, oil filled, and fan cooled, and of course with radiators to assist dumping heat.

But then again...?

I would point out that there could be high resistance shorts that cause failure over time.

Nonetheless, as I have pointed out before, there are two things that strike me, i.e., the apparent random nature of the shorts, and as well the fact that the grounding system, on at least one unit since one value is all we have been given, is way over the 5 ohms the literature says should be max.

You mentioned breakers and fuses but we have been given no data thereto. It would seem to me that as a unit fails shorted, because that would be the reality in the process of burning out, breakers or fuses would get involved in order to protect the line and the generating source.

j.

Regarding the breakers and fuses, I understand your comment completely. They will activate when the transformer has failed to protect the rest of the system.

My observation is that the OP hasn't referenced any occasions of unexpected blown fuses/breakers that indicates a severe overload condition. (Other than the final failure of the unit.

Earth line resistance is noted, but would that heat the transformer or just the earth stakes/cables and such? (Not a rhetorical question, I would like to know the answer.)

I would have thought (again not my specialty) that a high resistance earth would only lead to a floating earth potential rather than heating in the transformer.

Again, the OP has commented about failures of primary and secondary coils. Is that significant? That's partly why I suggested the cooling of the transformer be visited.

We sometimes use the "5 Why" system to identify root cause of problems.

Why do transformers fail? Answer: They are overheating

Why are they overheating? Answer: The coils are getting too hot

Why are the coils getting too hot? Possible answer: There is too much heat getting into them (high resistance coils, eddy currents in laminate, hysteresis in laminate etc) OR they are not able to eliminate the heat that is developed.

Why are there high eddy currents? And so on.

As each possible is eliminated with data, we eventually know more about the fault.

The reason the grounding is important is that lightning strikes could damage the transformers but not necessarily immediately wipe them.

As well, given the reason for the transformer failing, I could see a short to the casing of the transformer insufficient to blow the unit but in the absence of a good ground causing heating and eventual failure.

The other thing I had thought about is what kind of wire is the line composed of and as well the secondary feeds and in those conditions are there any special considerations to be given to the connecting points. High resistance and/or arcing could cause problems.

Sitting here in the states, or in any of the western industrialized countries we take a lot for granted. For instance does Saudi Arabia have any manufacturing capacity for any of that electrical gear.

Our friend could be going ape just getting his hands on replacement transformers. Here, in Atlanta, I could pick up a phone and have one in an hour.

There really is a lot more to know.

The OP has not given us a lot of information that would help. My questions about the sourcing of those transformers, for instance. I have read other places of brand new transformers failing or, for that matter, of transformers failing after lightning strikes.

The other thing that would be useful is a system map and wiring diagram with data like grounding resistance specific to each location and installation. That kind of stuff lets us see if there is any pattern to the failures.

Still, given where he is, my bet is the OP has his hands full.

j.

hi

H T fuse links were blown out( 2 N0's), as wel as temp noted at the time of fault reaches at 70 c

Aamersyed,

How about the rest of the info.

What is the resistance to earth at EACH of those 40 transformers.

What is the resistance at EACH of the units that burned out?

Were each of those transformers checked before turning HT into them? If so how were they checked?

Are they each of the same manufacture?

Where were they made and who made them?

These are the kinds of forensic investigation issues that may tell us why those transformers failed.

Here is another question. How long has that generation source and HT line been in operation? The way you put it initially gave me the impression that the whole kit and kaboodle were part of a recent project. Is that true or is the line long standing and installations made on it over a long period.

In that vein, are the villages long existing and the generation and distribution system part of an electrification program?

If so there might be comparative technical data here in the states where in the last century major electrification programs were conducted in this country.

j.

Hi

Let's try to refocus on the transformer's failure and the available data and information provided by all your mails.

1- The HV distribution network is a radial type, feed by a central substation and 4 overhead lines 60 to 100 km long.

2- Measured earth resistance of neutral is, at faulty transformers, around 13 Ohm

3- This is a new rural electrification project

4- All power transformers are of the same manufacturer and built according the same standards

5- All transformers are of sealed type ( I'm not quiet sure about that, I could not identify the presence of a oil reservoir on the top of the transformers).

6- All delta/star transformers are protected by fuses at the HV side and the fuses are ceramic encapsulated type, the LV side as no protection by fuses or breaker and the transformer itself as no temperature protection or minimum oil level protection.

7- Tee-of from LV transformer cabinet also are individually protected by fuses.

8- Daylight temperature in all transformer sites can reach 70º C

9- In 2 cases of seven burned out transformers fuse links at HV side also have fused

10- % of load charge to the nominal charge is not the same for all transformers, same of them are more charged than others.

I presume that in all cases faulty transformers (HV and LV faulty transformers) have been cleared from the network by fusing the fuses at the transformers in feed, other wise transformers do not seam to be well protected, especially if they are almost "cooked"

1 - These lines are long lines but here in Angola we also have very sparse population and very long lines operating at 30 kV level (lines with more than 60 but less than 100 km). When these lines have a low load, tension at the far end (transformers end) is a little higher but inside the withstand capability of 36 kV material. So ok during low load hours isolation materials are more stressed. According your data faults do not happened during low load.

2- Measured earth resistance is high but the consequence only can be related with long fuse times to clear earth faults on the LV side of transformers and a low human hazard protection. So, if we take in consideration the low lightning probability at transformers sites, at a first glance this high earth resistance can not be related with faulty transformers in a quite new electrification project.

3 – Transformers are manufactured according same international or national standards and nominal power is stated based on air temperature and pressure. This is also true to all other electrical material including the fuses that protect the HV side of the transformers. Most standards stated 45º C as the normal temperature and when the electrical equipment operates above this temperature limit the nominal power must be derated. How to calculate the derated power it depends on the maximum environmental temperature, how many hours this temperature is maintained and also it depends on the materials used to manufacture the equipment (fuse links encapsulated with ceramic material and silicon sand probably are less derated than transformers submitted to the same environmental conditions.

So, quite probably, at the base of your problem you are facing a temperature issue. If it is only a matter of air temperature and air density (lower air density also translates as lower power capability) you can expect a random transformer failure, that random behavior disappears if failure can be related with transformer load, if transformers are more close to their rated power when they operate at a high environmental temperature they will fail in a shorter time.

All faults (or most of them) occurred during daylight on a hot day? At what hour is peak load in the villages with faulty transformers?

Sealed type transformers are fare more stringent about temperature and a common problem is some tank deformation when operating at high temperature, these can result in small oil leaks at the weddings at tank bottom, then oil level inside transformer drops a bit and the off-load tension selector at HV side can be in open air, next step will be a arc between the terminals of the tension selector. Sealed transformers are not normally used on open air in Angola.

Faulty sealed distribution transformer (15/0,4 kV 630 kVA) interior type – good aspect at the top, oil leak at bottom – tank deformation caused by poor design – long distance between U shaped supports – fault at HV tension selector

Your last information about insulation test carried out on one of the faulty transformers seems to show a "well cooked" transformer (with no more than 250 Ohm as earth insulation resistance, is this correct?) but don't you think that ratio values are quite strange?

If you can take one or two faulty transformers from their tanks you can confirm or dismiss the temperature issue.

Try to monitor load conditions of operating transformers, as well as air temperature until you can decide to open one or two faulty units and have a clear idea of faulty cause (load and temperature recorders of operating units also can help to confirm or not a temperature issue).

regards

Japcs,

Either you presume a lot or you have had other contact than on this web site.

j.

Hello

Jack Jersawitz

You are right, I presume a lot from all small bits of all given information (even if some is also doubtful, as a so high ambient temperature as 70º C – when 50 or 55º C seems to be more common – or what seems to be a strange ratio for these delta/star transformers):

1- a radial network is presumed from 4 different distances all related with a common starting point.

2- Earth resistance at faulty transformer neutral point was stated as 13 Ohm

3- A new rural electrification project because was initially stated that same transformers (5+2) and now more one had burned out within one week past installation. Also was stated that all installation work was done by the made contractor. I presume that is a rural electrification not only based on long overhead lines but also on rated power of transformers (100 and 200 kVA)

4- All transformers of same manufacturer and standard that is a tiny supported presumption (this presumption is related with the information that all work have been done by only one contractor).

5- If transformers are or are not sealed type transformers that can only be of marginal importance.

6- HV fuses have been referred as transformer's protection device, also have been stated that in one occasion HV fuse links had blown when one transformer burned out. All other considerations, including type of fuses of LV side and transformer protection are presumptions.

7- Is stated that ambient temperature can reach 70 ºC, the assumptions about temperature are: that is maximum ambient temperature and all electrification area has similar maximum temperature.

8- Transformers load is stated about one unit and can be computed as 20 kVA (a fair low load), my assumption is that not all transformers have a so low load.

"Just an Engineer" point that faulty transformers are not described as a blown transformer or a shorted transformer and asks about ambient temperature and peak load time. When "Just an Engineer" point to possible heat problems and asks for same more data about transformers coil resistance it was a good point (as you also stated). Then it comes the 70º C ambient temperature information (it is a quite high ambient temperature, not far from maximum permissible oil temperature) and heat transfer problem now seems to be much more than a good point. To push data collection in that direction (and a better definition of previous collected information about system characteristics and behavior) using faulty transformers' inner inspection, load recorders, ambient temperature data recorders, and tank temperature monitor of operating transformers seems to me to be a straightforward approach, regarding all available data till this moment, to uncover possible fault causes.

regards

High temperatures in that area are a given and have been in everybody's mind without doubt.

Beyond that, as a materialist, a technician, and scientific in the manner of my thinking, I make no assumptions without specific data.

For instance your assumption about a radial layout. Perhaps if the villages were laid out that way. If it is an electrification project of old villages it would not be likely they fit a radial pattern. Further affecting such notions would be road ways, geography, etc.

In fact the first descriptions led me to believe the layout was linear with the generating system somewhere along that line.

I queried you because I don't think speculation or assumptions are a good way to solve problems.

Here in the states they have a saying about assumptions based on the first three letters of that word which is what assuming makes of all of us.

The problem is material. There are material factors involved. The problem can no doubt be solved by reference to material data, all of which we do not yet have. It is an interesting problem and it is to be hoped that Aamersyaad will give us all that data so we can provide whatever help he needs.

No doubt he is working under pressure. Once folks get electrified they are very unhappy when the electrics go down.

j.

hey guys i lived in sichel ave and all the suddend we heard like boom one time.all my family where talking about that.that was creepy.also that all of us where talking about that all night.