We recently completed our annual oil analysis on seven transformers- 4 of them showed a significant rise in dissolved gases from last years results. We were somehat baffled because we've never seen a sudden increase like this. Someone then mentioned that several months ago the utility increased the incoming voltage by 2% ( 800v). Three of the transformers are fairly old but one was installed in 1997 and it too showed a sudden increase so my question is ( and I'm not an EE)- would the increased supply voltage ( to 44kV from 43.2)be one of the reasons for the sudden increase?
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Transformer Oil Analysis -worrisome results 3
- Thread starter ruble3
- Start date
QUESTION: …would the increased supply voltage ( to 44kV from 43.2)be one of the reasons for the sudden increase?
ANSWER: Probably not. Transformer designed per ANSI/IEEE C57.12 should be capable of operating continuously above rated voltage or below rated frequency, at maximum rated kVA for any tap, without exceeding the max. temp. rise when all of the following conditions prevail:
a) Voltage < 105%
b) Volt per Hz < 105%
c) Load PF > 80%
d) Frequency > 105%
At not load with same conditions as above, the allowable voltage and volt per Hertz <110%.
ANSWER: Probably not. Transformer designed per ANSI/IEEE C57.12 should be capable of operating continuously above rated voltage or below rated frequency, at maximum rated kVA for any tap, without exceeding the max. temp. rise when all of the following conditions prevail:
a) Voltage < 105%
b) Volt per Hz < 105%
c) Load PF > 80%
d) Frequency > 105%
At not load with same conditions as above, the allowable voltage and volt per Hertz <110%.
Did the firm that conducted the DGA have any recommendations? They usually have the benefit of a large database correlated to longer-term equipment performance. Needless to say, shorter re-test intervals may be in order. [Antiseptic sampling procedures are critical.]
ANSI C84 for “46kV” gear lists a maximum of 48,300V, but if the equipment is older a lower value might have been used for design.
electricpete
Electrical
Since you didn't tell us what was the nameplate voltage/tap to begin with, I would say there is a possibility the transformers were close to saturation and now pushed further into saturation. The increase in core losses from that may not cause any noticeable change in temperature (or pump/fan cycling if applicable) but stil the hot core can cause hot metal gasses.
4 changing at one time would seem to command some explanation other than all 4 developed a problem at the same time by coincidence. The only other explanation I can think of influencing multiple transformers is as mentioned above possible lab problem - check with a followup sample
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
4 changing at one time would seem to command some explanation other than all 4 developed a problem at the same time by coincidence. The only other explanation I can think of influencing multiple transformers is as mentioned above possible lab problem - check with a followup sample
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
- Thread starter
- #6
Thanks for all the replies- the same company has been doing the tests for quite a few years- three of the transformers are at the highest tap position & are at full rated capacity - the testing company has recommended shutting down each transformer, rotating the taps through full range;performing winding resistance tests & powering up again-they also recommend re-sampling within 6 months- one transformer is vintage 1953 but load has not changed much in 5 years; one is vintage unknown; one was built in 2000; and one is vintage 1965-three of them are 44kV/4160 and another 44kV/600v)
It's very strange that they would all suddenly show problems. Have the loads changed on all of the transformers? One thought - a lightning surge can cause momentary arcs which would produce some gasses. If this is the case, there should be no further increases. What gases specifically have changed?
Electrical tests might turn up something, and if it were just one unit, I might head that direction. But with 4, I suspect the samples were contaminated.
Electrical tests might turn up something, and if it were just one unit, I might head that direction. But with 4, I suspect the samples were contaminated.
As already highlighted by others, the test labs aren't infallible, even the good ones. It is normal to stay with one test house for the purposes of long-term trending as subtle differences in equipment and operator can cause variation between test houses. For results which show massive and inexplicable change from the previous test it is reasonable to either re-test a few days later with the same test house using a different instrument if they have one, or cross-check with a second test house. Provide a fresh sample in either case, because the sample bottle may be contaminated.
Have you had a re-test done, or tested a later sample which corroborates the first results?
----------------------------------
If we learn from our mistakes,
I'm getting a great education!
Have you had a re-test done, or tested a later sample which corroborates the first results?
----------------------------------
If we learn from our mistakes,
I'm getting a great education!
What are you considering to be a "significant" rise in dissolved gases over that past year and specifically what gases have increased.
As stated by others I would resample immediately and verify that the DGA results are the same. If they are not, as stated I would suspect contaiminated samples and improper test method in that order.
Just because their is a rise in gas isn't particularlly a cause for concern it depends on what gases and how much.
As stated by others I would resample immediately and verify that the DGA results are the same. If they are not, as stated I would suspect contaiminated samples and improper test method in that order.
Just because their is a rise in gas isn't particularlly a cause for concern it depends on what gases and how much.
- Thread starter
- #13
Sorry for the delay in responding(plant shutdown;golf holiday in Palm Springs!)
Here are sample results from one of the transformers(others are similar)
Year 2000 2001 2002 2003 2004
Hydrogen 35 50 25 30 40
Oxygen/Argon 22300 22220 18200 22800 29700
Nitrogen 63300 59600 50400 58700 69200
Co 299 325 112 266 336
Methane 15 25 <5 5 15
Co2 2250 3320 1200 1890 2440
Ethylene 32 68 6 14 46
Acetylene <2 <2 <2 2 2
Gas Content 8.83 8.57 7.00 7.88 10.18
Any comments?
Here are sample results from one of the transformers(others are similar)
Year 2000 2001 2002 2003 2004
Hydrogen 35 50 25 30 40
Oxygen/Argon 22300 22220 18200 22800 29700
Nitrogen 63300 59600 50400 58700 69200
Co 299 325 112 266 336
Methane 15 25 <5 5 15
Co2 2250 3320 1200 1890 2440
Ethylene 32 68 6 14 46
Acetylene <2 <2 <2 2 2
Gas Content 8.83 8.57 7.00 7.88 10.18
Any comments?
Ruble3,
Was the transformer oil processed through an Ilovac or similar in 2002? All the DGA results have taken a downward step at this time.
----------------------------------
If we learn from our mistakes,
I'm getting a great education!
Was the transformer oil processed through an Ilovac or similar in 2002? All the DGA results have taken a downward step at this time.
----------------------------------
If we learn from our mistakes,
I'm getting a great education!
Generally, DGA results alert to possible arcing/sparking or low/high temperature overheating. So typically, the problem is current related.
However, an increase in hydrogen in the oil may point to corona discharge which may be related to the higher voltage.
In this case the insulation of the transformer may have deteriorated to a level where the higher voltage is causing corona or we are seeing the affects of pd activity.
However, an increase in hydrogen in the oil may point to corona discharge which may be related to the higher voltage.
In this case the insulation of the transformer may have deteriorated to a level where the higher voltage is causing corona or we are seeing the affects of pd activity.
- Thread starter
- #16
Sorry the chart didn't tun out like I expected
ScottyK- not sure about that- I think this transformer , if I remember right, had a complete oil change in 2002 so I should have stated that;I'll have to check the records
This particular transformer was manufactured in 1953; of the four transformers in question this one has seen the least load increase but the results are just as bad- we will probably shut them down, rotate taps & resample & take it from there
Thanks for all the replies
ScottyK- not sure about that- I think this transformer , if I remember right, had a complete oil change in 2002 so I should have stated that;I'll have to check the records
This particular transformer was manufactured in 1953; of the four transformers in question this one has seen the least load increase but the results are just as bad- we will probably shut them down, rotate taps & resample & take it from there
Thanks for all the replies
-
3
- #17
Let me start my response by saying I'm not a PE, but I've years of substation and large transformer experience so I'd like to make a few comments here. Let's start with some basics about insulating oil and analysis of same as it pertains to power transfomers. When conducting DGA of transformer oil, there are four main catagories that I personally like to keep a close eye on. Please keep in mind that I'm making the following statements based on dealing with power transformers only. On-Load tap changers and other devices require adjusting the way we look at the numbers obtained from the dga. I've also taken into account your statement that you believe that the oil in the unit had been either processed or replaced in 2002.
1. Internal arcing in a power transformer causes primarily the formation of Acetylene gas. As a transformer ages it is normal to see slight elevations in the acetylene numbers. Sudden jumps in acetylene content indicate that internal arcing has taken place and close monitoring is a good idea. If a follow-up(3 to 6 months) test indicates significantly higher numbers than that of the first test that revealed a relatively large jump in acetylene numbers, its time to think really hard on removing the transformer from service and seeking its rebuild or replacement. Especially if it is in a load critical area. As with all dga analysis, trending of results is essential. At least to me, your Acetylene numbers look fine.
2. Ethane, methane, and ethylene are called "heating gases". Their detection and rate of increase in a power transformer are directly linked to the heating(or more importantly, the OVER heating) cycle of a transformer. A significant rise in these numbers indicates that there is heating to the point of degradation of components internal to the transfomer. The most important of these components is the paper insulation which is present in almost all large power transformers. If you see(ppm) numbers in the hundreds and certainly in the thousands, your transformer needs immediate attention. Are you operating the unit withing its temp limits? i.e. 55deg.C @ XX MVA or 65deg. C @XX MVA? Accurate temp gauges for winding temp and liquid temp are mandatory here as are people who know how to read them and understand the significance of those readings. Maintaining the transformer within its temp limits is absolutely necessary if you expect normal life from your trans. More heat, less life.
3. I like to watch the acid content in oil. The acid numbers indicate a general condition of the oil itself. Age, oxidation of the oil itself and overall condtion of the internals of the transformer are also reflected in the ppm of acidic content in transfomer oil. Acid also attacks the cellulose insulation in the transformer. Generally, acid precedes the formation of sludge and is a very bad thing to have in your transfomer.
4. There are 2 numbers most significant to the the above post. Oxygen content and Carbon Dioxide content. There are two main avenues for oxygen to appear in the insulating oil of power transformers.
(1.) Oxygen forms due the overly rapid breakdown of the paper insulation in the transformer. This a a certain death senario. It's just a matter of time. There are a number of different ways to get the paper to deteriorate more rapidly than normal. A likely way in this situation is to overheat the transformer on a regular basis. Paper insulation oxidation rates are much higher when subjected to heat outside of the norm. The by-products of "cooking" the paper are significant levels of Oxygen, and Carbon Dioxide. The latter is most prevalent when over-heating of a transformer is a regular occurence. When these numbers begin to climb, you need to due some serious thinking and investigation if you want your transfomer to live a long and heathly service life.
(2.) Another prevalent way for oxygen to form in a transformer is through contamination. Water entering from "outside" will most surely lead to elevated O2 numbers. This condition is most serious! Very rapid deteriorization of paper insulation, internal corrosion of tank and workings, and in sufficient quantities, catastrophic failure of the unit can be attributed to water infiltration. In short, if you've got water, you'd better uplug the unit before it unplugs itself..........permanently!
So, ruble3, In my opinion, shutting down and changing taps will do nothing for you at all. If your units have the tanks nitrogen pressurized, I'd be checking to see if the n2 charging systems are working properly. A tank with positive pressure inside will slow(but not stop) contamination from the outside. If an n2 system is not installed on your transformers, I would definately be looking at the possiblilty of water infiltration. Are these transformers installed in typically wet areas or are they subjected to high moisture levels on a regular basis? (Like maybe a water truck spraying down a dusty mine road and spraying down your transformer several times a day as well?)
Look at the temps and make sure your temp gauges for liquid "top oil" temp and winding temp are working properly. Are the operating temps exceeding the ratings on a regular basis? (Your ethelene content numbers indicate at least some overheating, at least to me)
Lastly, I'd say the most likely situation here is you've got transformers that are running at their max rating(electrical) and you are regularly exceeding the thermal rating. Thus the steady climb in Ethylene, and a very high o2 content which comes from degradation of cellulose insulation due to excessive heat and possibly from contamination from outside the tank.
1. Internal arcing in a power transformer causes primarily the formation of Acetylene gas. As a transformer ages it is normal to see slight elevations in the acetylene numbers. Sudden jumps in acetylene content indicate that internal arcing has taken place and close monitoring is a good idea. If a follow-up(3 to 6 months) test indicates significantly higher numbers than that of the first test that revealed a relatively large jump in acetylene numbers, its time to think really hard on removing the transformer from service and seeking its rebuild or replacement. Especially if it is in a load critical area. As with all dga analysis, trending of results is essential. At least to me, your Acetylene numbers look fine.
2. Ethane, methane, and ethylene are called "heating gases". Their detection and rate of increase in a power transformer are directly linked to the heating(or more importantly, the OVER heating) cycle of a transformer. A significant rise in these numbers indicates that there is heating to the point of degradation of components internal to the transfomer. The most important of these components is the paper insulation which is present in almost all large power transformers. If you see(ppm) numbers in the hundreds and certainly in the thousands, your transformer needs immediate attention. Are you operating the unit withing its temp limits? i.e. 55deg.C @ XX MVA or 65deg. C @XX MVA? Accurate temp gauges for winding temp and liquid temp are mandatory here as are people who know how to read them and understand the significance of those readings. Maintaining the transformer within its temp limits is absolutely necessary if you expect normal life from your trans. More heat, less life.
3. I like to watch the acid content in oil. The acid numbers indicate a general condition of the oil itself. Age, oxidation of the oil itself and overall condtion of the internals of the transformer are also reflected in the ppm of acidic content in transfomer oil. Acid also attacks the cellulose insulation in the transformer. Generally, acid precedes the formation of sludge and is a very bad thing to have in your transfomer.
4. There are 2 numbers most significant to the the above post. Oxygen content and Carbon Dioxide content. There are two main avenues for oxygen to appear in the insulating oil of power transformers.
(1.) Oxygen forms due the overly rapid breakdown of the paper insulation in the transformer. This a a certain death senario. It's just a matter of time. There are a number of different ways to get the paper to deteriorate more rapidly than normal. A likely way in this situation is to overheat the transformer on a regular basis. Paper insulation oxidation rates are much higher when subjected to heat outside of the norm. The by-products of "cooking" the paper are significant levels of Oxygen, and Carbon Dioxide. The latter is most prevalent when over-heating of a transformer is a regular occurence. When these numbers begin to climb, you need to due some serious thinking and investigation if you want your transfomer to live a long and heathly service life.
(2.) Another prevalent way for oxygen to form in a transformer is through contamination. Water entering from "outside" will most surely lead to elevated O2 numbers. This condition is most serious! Very rapid deteriorization of paper insulation, internal corrosion of tank and workings, and in sufficient quantities, catastrophic failure of the unit can be attributed to water infiltration. In short, if you've got water, you'd better uplug the unit before it unplugs itself..........permanently!
So, ruble3, In my opinion, shutting down and changing taps will do nothing for you at all. If your units have the tanks nitrogen pressurized, I'd be checking to see if the n2 charging systems are working properly. A tank with positive pressure inside will slow(but not stop) contamination from the outside. If an n2 system is not installed on your transformers, I would definately be looking at the possiblilty of water infiltration. Are these transformers installed in typically wet areas or are they subjected to high moisture levels on a regular basis? (Like maybe a water truck spraying down a dusty mine road and spraying down your transformer several times a day as well?)
Look at the temps and make sure your temp gauges for liquid "top oil" temp and winding temp are working properly. Are the operating temps exceeding the ratings on a regular basis? (Your ethelene content numbers indicate at least some overheating, at least to me)
Lastly, I'd say the most likely situation here is you've got transformers that are running at their max rating(electrical) and you are regularly exceeding the thermal rating. Thus the steady climb in Ethylene, and a very high o2 content which comes from degradation of cellulose insulation due to excessive heat and possibly from contamination from outside the tank.
Documents 3-30 and 3-31 on the following page may also be worth reviewing.
- Status
Similar threads
- Question
- Locked
- Question