Annealing Hi-B DR distribution transformers wound cores

[F.Yermanos]

Hello, i work on design and R&D in a transformers enterprise.
I've seen that many of the CRGO steel mills, are switching their production lately from standard CRGO steel to hi grades like HiB or HiB DR silicon steels.
This is also a countermeasure from the regular silicon steel market to compete against amorphous materials , wich in the last 15 years have won market due the grow in regulations of energy efficiency all around the globe.
Many small distribution transformers manufacturers which use wound core technology (Tranco or AEM), found that work with amorphous materials isn't and option yet, that's why they still designing using CRGO or/and switching to higher steel HiB grades, which have lower losses than standard CRGO to achieve the efficiency levels of the market.
When you design and manufacture a wounded core using regular CRGO steel, its mandatory (or the common market practice) to relief mechanical stress (to reduce losses, due domain dis-alignment) of the steel by annealing the core.
Now if you switch to a HI-B DR steel with better permeability, after you "shape" the core and measure the power losses ,looks like there's no need to annealing the core and also because if you anneal HiB domain refined steel , the redefinition of steel domains will be loss thereby losses will increase after annealing, which has no sense.

We know by theory that steel destruction factor in wound cores, is related to core mean perimeter to folds number ratio and as the core is smaller the higher the DF.
So on smaller wound cores with shorter perimeters the DF is so high after forming, that using DR steels seems not an option.

That's why i wondering how good will be not anneal the HiB DR cores if the specific losses won't increase above the design values after core forming?
Is there another factor besides power losses, excitation current and sound pressure level to take in count, whether or not to decide to anneal?
Does the core after shaping process will keep some harmful remanent flux due the areas where the grain where no re-oriented and stress did not get relieved over folded corners?.
Is there any influence over the core gaps and flux density distribution, by no annealing?

In general avoid the annealing process in wound cores, implies huge savings in costs (nitrogen, spare parts, energy, maintenance), customer lead time and overall manufacture time, at first sight it's an improvement. However i'm not quite sure if it is a safe practice or in short time this will be harmful to transformer behavior or performance.

I want to hear your thoughts about it. thanks

 

[prc]

Wound cores are to be annealed to reduce the building factor or destruction factor as you say. Of course if you are getting designed losses, excitation current and sound level without annealing, you can go for that. But it means you have still potential for savings. You are right, DR grades should not be annealed as it will affect the domain structure. But Japanese manufacturers have special DR grades that can be annealed.
Some history.More than 50 years back when making power transformers, we used to anneal each lamination in a moving annealing oven and then varnish and bake it in another oven. Later we found out that with wider sheets effect of annealing is negligible and stopped annealing and varnishing.Improvement in cutters ( silicon carbide cutters instead of carbon steel cutters) reduced the burr and strain in laminations. Better inorganic insulation coating on laminations by steel makers, helped in removing varnishing.

 

[F.Yermanos]

Hi. Thanks for your kind reply.

Yes, we made several tests and transformers using Japanese steel, which has mechanical scribed domain refined instead of regular standard laser-scribed.

The first one is immune to annealed because mechanical scribed improve the grain's bloch walls inter distance.
The issue of Mech-DR , is the cost and factory lead time (similar to amorphous materials). We run many optimizations using this grade of steel after annealing and even when losses aren't as low as you'll get with amorphous materials , the final TOC is quite low even adding the annealing cost.
My concern is about remanent flux issues this non annealed cores will may get, as many pole-type distribution transformers has large and continued un-controlled TOV (if not accurately protected), which will trip HV fuses or disconnectors due transformer overexcitation under this conditions.