Transformer energisation restriction 2

[jimmy2times]

I’ve picked up an operating procedure implemented some years ago that recommends a period of 12hrs before a transformer is re-energised after in has been shut down. Nobody knows why this restriction is in place, I contacted manufacturer also they didn’t have any restrictions on how often and just referred to thermal red protection that will take transformer offline if temperature too hot in core/winding. Someone suggested it may have been frequently shutdown and energised previously, 7 or 8 times per day, and this was perhaps a way of limiting number transformer energisation to 2 /day. Anyone come across this before?

 

[itsmoked]

Maybe look beyond the transformer? Could be something else that gets quasi-abused during the switch ON. A breaker or generator?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Tell us a little about the transformer.
Tell us a little about the loads served.

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

 

[jimmy2times]

It’s a ship’s distribution transformer. 6.6/0.45kv, feeds general ships loads. Part of redundant pair, one feeds port bus, the other starboard bus. so typically each transformer carries less than 50% load. If one transformer were to fail, other could supply both port and starboard bus loads.

Typical ships loads, pumps, fans, domestic loads etc supplied.

I don’t think it is a power quality related thing. There is a pre-mag solution employed for energising transformer on weak grid. Transformer rated 3.5MVA. Generator plant on 6.6kV are four diesel, 2x 2.75MW and 2x4MVA sets. The pre-mag solution allows transformer to energise from single generator following a black out.

Interlocking on mv and lv sides, so transformer will be energised on no load. Then lv breaker closes (either on dead bus, or part of auto-transfer scheme to redundant transformer)

Ship staff find the time restriction very prohibitive as it stands, e.g they may transfer loads to one transformer. Then they have this 12hr operating restriction before they can swing loads back again on to the other transformer. This is not a daily occurrence, e.g. for maintenance on LV switchboard or transformer, or occasional switchboard training drills. They want to ditch the 12hr restriction that nobody seems to understand where it crept in.

 

[waross]

Thanks Jimmy.
Tell us about your pre-mag scheme.
Does it have any components that need a cool down between uses?

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

 

[jimmy2times]

Pre-mag transformer has S2 intermittent duty,
Specification states:-
On 10ecs Isec = 150A, then
On 15secs Isec = 10A
Off 215secs

Then follow by break at least 15mins

 

[argotier]

I'm curious, the procedure states clearly that this "waiting period" is needed after every shut-down?

Transformers manufacturers use to recommend a similar waiting period, but for the first energization, right after its assembly, to give enough time for possible air bubbles to reach the gas chamber/conservator and allow all submerged parts to be fully embebed with oil (This is more important when is not vacuum filled).

 

[jimmy2times]

Sorry, I should have said upfront. Transformer is dry type

 

[waross]

Transformers manufacturers use to recommend a similar waiting period, but for the first energization,

I wonder if a "Cut and paste" engineer found that spec, and not understanding it, pasted it into the procedures.
I have seen some silly specs from "Cut and paste" engineers.

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

 

[itsmoked]

That's such a looooong time. Only really massive things would take that long to cool down. Massive like only big transformers or reactors. Even an engine with water cooling has a relatively fast time constant.


A dry type 3.5MVA.

Is there any temperature monitoring in the transformer?

Got a nameplate photo?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[RRaghunath]

The energisation inrush current threatens the winding braces in transformer. Many a transformer fail during energisation due to the dynamic forces that the windings experience during energisation.
You mentioned premag scheme exists that reduces the transformer inrush currents (I suppose). If true, the 12-hour restriction can be done away with.

 

[jimmy2times]

Yes, winding RTDs fitted

Good thinking on using pre-mag system regardless what spinning capacity is on line. Surely better for transformer life anyway

 

[prc]

1) I believe the transformer is dry type (cast resin?) to be energized from 6.6 kV side using vacuum circuit breaker. For dry-VCB combination energization is always a dielectric stressing incidence of transformer insulation. So IEC 60076-11 ed2.0 -2018 on Dry Transformers states (clause 4.2)
Quote
i) Frequency of energization:
If frequency of energization is greater than 24 times a year, then purchaser shall specify
the number of energizations per year at the enquiry stage.
Unquote

2) But I don't think 12 hour gap between energization is to reduce the number of energizations. If you want energization, you will do it even after 12 hours. So the energization number may not come down. I believe this gap may be to reduce residual magnetism in pre-mag transformer. I am not sure how much it will be effective. This type of transformers, pre-mag solution is always seen , may to avoid over loading of diesel set by transformer inrush load.

3) Can you share the details of Pre-mag solution? Brochure, drawing or instruction manual?

 

[itsmoked]

And.. You have RTDs! Do you have any before and immediately after (minutes) temperature recordings? That would certainly inform us as to what the transformer thinks thermally.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jimmy2times]

Itsmoked, that’s what I was planning to do next. Take transformer offline, let it cool down to ambient and then energise (without pre-mag transformer). Then shut down after 1min. And see what the temperature profile looks like. Then I was going to try and load transformer to as high as possible load then switch off and see the temperature cooling down profile as it is taken off load.

From that I was wondering if it could be seen that if transformer had for example being running at high load, then taken offline, then within short time frame reenergised whether just the inrush current profile alone would cause winding temperature to overshoot. I wouldn’t have thought so but just trying to build some reassurance for case to removed this restriction.

 

[itsmoked]

That will be a conclusive test! Meanwhile learn about the RTD readings. Do they seem to be truthful? Temperature seem reasonable? Does it go up with load. They fail sometimes and so it's good to get some knowledge and a bit of trust with them before you depend on them for test results.

Hopefully you have some history on them already and can just examine it. If not get the operators to start logging every hour or some workable period; Ambient, RTD(s), Loading. You should see rational results which will confirm you have healthy temp sensing.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[prc]

Jimmy, please refrain from the experiment to see overheating in transformer. It may simply kill your transformer !

1) You did not confirm whether you are using VCB for switching in. Most probably it will be so.

2) Inrush current flow will be there for maximum 1-2 seconds and it will be fast coming down. Your transformer is not going to be over heated and no need to check it by trials. It is not RMS current and the duration of peak of inrush current is negligible to cause heating.

3) Now dangers of your experiment- when you remove pre-mag and energise without resetting relays, it will cause a tripping due to higher inrush current.( 10 times full load current. Tripping by VCB during inrush is a sure ticket for dry transformer failure ! I had to handle failure of many brand new dry transformers due to tripping during energization. Relay engineers have a tendency to trip unit to check their setting. This will result much higher TRV from restrike in VCB. Please see C57.142 Guide for occurrence and mitigation of switching transients.Engineers found out this phenomenon during early days of VCB (early 1970s).If energization is too frequent manufacturers like ABB have special surge resistant dry transformer designs to offer.

4) IEC on dry transformer cautioned on the no of energizations not from thermal angle but due to dielectric reasons.

 

[waross]

A test suggestion.
ALWAYS USE YOUR PRE MAG DEVICE TO ENERGIZE.
Note the transformer indicated temperature when under load and the ambient temperature at the same time.
For example:
Ambient temp = 30 degrees
Transformer temp = 70 degrees.
Transformer heating = 40 degrees.
40 degrees x 0.62 = 24.8 degrees.
The time to reach 54.8 degrees (30 degrees + 24.8 degrees) is one time constant.
These numbers are an example only.
Please use your own numbers, Jimmy.
This is based on energization with the same ambient and with the same load as before the transformer was shut down.
It is generally accepted that the transformer temperature will stabilize in five time constants.
For the purists, the error will be (1/e)[sup]5[/sup]
Or roughly 0.37[sup]5[/sup] = 0.69%
Using e, we get (1/2.718281828459)[sup]5[/sup] = 0.00673794699908546709663604842315 or 0.67%
Please be guided by prc's advice and always use the Pre Mag.

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

 

[jimmy2times]

Thanks gents

Yes switchgear is VCB

The transformer is insulation AC20kV and LI 60kV

Prc, I looked at part 11 of IEC60076 and could not find anything further than what you already quoted in the standard. What would be the transformer manufacturers response to a purchase order inquiry that requested more often energisations that 24times per year? Would it to be increase insulation level or otherwise recommend surge arresters? I need to check if later is fitted either at switchboard or transformer. I always thought this issue was only a problem with VCB and low rated induction motors, it is basically same issue with transformers though right, restrike?

 

[waross]

When you de-energize, note the time to drop 63% of the way to ambient.
At 5 times the time to drop 63% it should be within 1% of ambient.
Beware, recently, maybe in this thread, maybe in another, I saw a link that wrongly described the time constant.
The time constant IS NOT THE time to reach steady temperature, the time constant is 1/5th the time to reach steady temperature, or the time to reach 63% of the final temperature.
And when using the 63% it is the time to reach 63% of a block change in temperature.
Heat transfer in a transformer is complex and may not follow simple formula exactly. Heating and cooling may not have the same time constant as the heat is transferred away from the windings through the insuation and through the oil by both conduction and convection.
Do you have any information on this aspect, prc?

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