Transformer Rewind - learning the first steps

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Sorry to chime in was thinking about that since yesterday and it didnt let me go :-/

CJ in your math some posts above i saw that youre calculating with 4% Si. Is that some kind of standard number for calculating?
I do a lot of Corn-oriented steel at work and none of them had more than 3.4% Si... mostly its between 3.1 and 3.35%. And in 20+ years in the factory we never had more then the 3.4% mentioned above.
Other magic of that sort of steel is the amount of Cu,Sn and almost no Mn. Just sayin :-X

Greets
 
Hi Pusch! that 4% number is stuck in my head because that is what i used to chop up at the transformer factory,  might have been called Microsil, it has been a while, we made big three phase transformers so core loss was a factor in the type of steel used as you don't have to use as big an oil tank if there is less heat.

so you are right, for small transformers for radios and tv's they probably use less expensive steel.

Lets try the RDH4 method for transformer design and see what they say for core size,

Page 235:  Cross Section Area  A =  (VA)^.5 / 5.58  

VA = .250 x 17 V-ac = 4.25

A = 4.25^.5 / 5.58 = 2.06 / 5.58 = 0.3695  This is our cross sectional area in inches

take the root for a sq stack = 0.608"  so RDH4 says use 625 EI  (5/8" tongue)

they have an easy mehod for turns also>

Turns: N = E x 10^8 / 4.44 f B A      change 4.44 to 4 as stacking is 0.9 x 4.44

watch out! in RDH4 they use Lines per Cm^ for flux and Sq Inches for core area,

most folks swap those>  lines of force in Gauss and core area in centimeters, it all works out he same,

N =  220  x 100,000,000 / f B A    punch in frequency , core cross area (use 50 EI = and flux

N = 2.2^10 / 4 x 50 Hz x 12 ,000 Gauss x  1.61 cm^2

N = 2.2^10 / 3,864,000 = 5,693 turns  ,

so you are running that core at a pretty high flux level, if using a transformer in a preamp you might want to keep the flux down to like 7 KG to keep the magnetic power inside the core.

here is a real easy way to compute primary turns from RDH4, for area: Page 235  for 60 Hz and 240 VAC:

N = 1860 / Core Area    you could de-rate for voltage and freq,

for 50 Hz:    N = 2232 / A

for 220 VAC and 50 Hz:  N = 2046 / A

 
you can lay the coil on a table,  put a piece of wood over it and press down, this will make the circular turns more square, or just squeeze it with your hands,

don't worry about breaking wire, the winders at work use to hammer on their coils with base ball bats and that's no joke. of course they were using # 4 Rectangular which is a bit tougher than #37.

your meter might have been energizing he core, the lams were waking up so the resistance was dropping,

make sure you do not slice thru the coil with the razor lams, extra layer of something between the core and coil, lots of people put .015" pressboard in here,
 
> I do a lot of Corn-oriented steel

English is a funny language.

We would say "grain oriented steel".

Yes, corn is a grain, and outside the US "corn" often means "grain". In the US, "corn" means "indian corn", maize, a Mexican grain not known in Old Europe.

Almost every word in English has multiple meanings. "Grain" also means "the lines in wood". The beautiful patterns in fine wood. The way cheap wood breaks along the grain.

Many meanings of "grain" in English.

Iron/Steel workers say that the metal has "grain".

Part of steel-making is rolling it to smash the iron together and to work out the slag. In transformer iron they do another step which tends to align the crystals and magnetic domains. https://en.wikipedia.org/wiki/Norman_P._Goss

Wikipedia says electrical steel runs zero to 6.5% Si, but over 3.2% is brittle. Your company may not have wanted brittle steel. CJ beats his iron with baseball bats so maybe he likes his iron crunchy.

Carbon is apparently real bad, because it comes out of solution over time and degrades the properties. Because Steel is defined as an Iron-Carbon alloy, I wonder if what we call "Steel" really is steel.
 
PRR you got me  ;) i stumpled over my own excitement. Hope you dont mind and had a good laugh about that mistake  ;D ;D ;D

In German we call it "Korn orientierter Trafostahl" and to be honest, i work in the liquid part of the Factory not where the Heat is being processed to Coils but youre right.. they get a heat treatment to realign the Grain.
We use other names than the outside world is calling them.

And maybe Wikipedia is saying carbon can be up to 0,5% we keep that down to 0,045%. But N2 (nitrogen?) up to 0.0090% which is actually a lot for our normal day steel...and the melting point (we call it Liquidus) is actual lower than 1500K (Kelvin).
IIRC around 1460K to1470K. Normal Steel (for beer cans for example) has a melting point of 1532K. Maybe you think its just 40K but we work in a 0-4K window before we send the heat to our molding guys.

Just to give an impression of size and weight: we add around 20tons of high purity FeSi for 3.3% Si.
The finished Heat takes about to 4 Hours of work and weighs around 375 tons.

Puschl
 
CJ,
One roll weighs about 20tons. So we cannot say its very handy xD

If you give me some analytics from the 80 Ni then i am able to look it up... maybe its possible to recreate it.

We use different types of coating, mostly for the car manufacturers. The grain steel i believe is finished by our customers.

But as i told you some days ago.. nothing is impossible ;)

Puschl
 
> Normal Steel (for beer cans for example)

Huh. Here I have not seen a steel beer-can since I cleaned-up trash from 1979 left on my property. All Aluminum.

 
Back
Top