How old can generator cores get?

[Jensgisla]

We are at the moment looking at a couple of 14MVA ASEA hydro generators from late 1950's. Both stator windings and rotor poles will be replaced but is it completely ridiculous to consider keeping the cores? Of course given that they turn out ok in flux tests. How old can cores really get? All your comments are really appreciated.

Thanks,

jens

 

[electricpete]

I don't have a lot of experience but if it's that old and still tests good I would feel prettty good about it having passed the test of time.

Don't forget to test again AFTER the winding has been burned out of the slots for rewind.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

I have a friend that works for a rather large workshop that does hydro-electric generator repairs. He has done som work where the stator plate was changed to new plate and where the air gap and plate profiles were changed after some FEM simulation. He raised efficiency by a couple of percents - and that alone may be worth the trouble.

 

[Sargardani]

In old stator cores one should look for any deformations, localized heating, damage due to shorted winding, fretting corrosion etc. Core loop test is a good way of knowing if there are any hot spots. Normally minor hot spots can be treated by weeping epoxy. However bigger hot spots with considerable damage to the core requires replacement of that particular section of the core. El-Cid test may be done in place of a core loop test, but is not as effective as the core loop test.

Also look for movement of the stator frames. In certain instances the core laminations may be in a good shape but over the years the frame might have moved out of tolerances, which may lead to restacking of the core.

Just a few thoughts.

Sarg

 

[QCE]

I agree with Sargardani:

I have seen many hydro rewind projects were a new winding has been placed in a much older core.

A good inspection is necessary and minor repairs may take place but in general it is acceptable to keep the old iron. Even a partial restack is very common.

 

[edison123]

Replacing cores is a costly affair because of cost of tooling, cutting etc. As others have pointed out, condition assessment is vital for the replacement decision.

If your m/c has hot rolled core laminations, then replacing them with cold rolled (CRNO) laminations, which have lower iron losses, may sometimes justify the cost.

In a case of a 40 pole, 50 W generator (which had 4 sections), the customer decided to replace one section of the core which was badly damaged due to winding fault. But when we opened up the other sections just to check, we found entire laminations were rusted right from the teeth to the back iron. So, we replaced about 50 tons of laminations, which was a costly and time consuming affair.

Of course, one advantage was, new core was CRNO (as against original hot rolled) and we eliminated all the section joints / vertical splits.

Generally, core initiated faults are rare and a healthy core is reused since the core plate varnish has a higher temperature capability and hence longer life as compared to the windings.

 

[mc5w]

Thinner laminations that are made out of better steel would be worth investigating. There have been some advances in magnetic steels even though amorphous steel is effectively dead.

I was also reading in a book on magnetism that somebody was able to reduce core losses in a wound core transformer for power frequencies using 2 techniques. The first one was very thin laminations on the order of 0.003 inches thick. The other was to etch very tiny grooves just a few microns deep in the direction of magnatic fields on 5 inch centers. Mechanically scratching the surface did not work because annealing tends to make scratches go away. The groove etching cut the hysterises loss by a large amount and the total losses in half.

The groove etching technique would not work on generator cores because the magnetism needs to turn 90 degrees at the teeth. Hysterises loss will always be higher than for a wound core transformer because in a generator or stacked core transformer some of the magnetism is 90 degrees to the crystal grain direction.

 

[mc5w]

I forgot to add that there have been advances in lamination varnish. All told, today's transformers are about half the size of transformers from 50 years ago and they are a lot more efficient.

I do not know of any advances in water turbine design but I do know that there is a lot of room for improvement. I would not be surprised that some hydrodymanics Real Smart Guy has figured out away to double the efficiency of water turbines.

 

[Jensgisla]

Thanks to all of you guys.

We will look at the possible efficiency increasement, we are only considering keeping the core if it measures good and the efficiency increasement can not justify the cost of a new core.

mc5w: This surface etching thing is very interesting. The turbine wheels will be replaced.