Tony's latest traffo DIY build

[Tony]

have a look at these videos:

Manufacturing Of Voltage Transformer part(1) Iqra University .flv - YouTube

Manufacturing Of Voltage Transformer part(2) Iqra University .flv - YouTube

Manufacturing Of Voltage Transformer part(3) Iqra University .flv - YouTube

[pieter t]

Originally Posted by johansen

I've always assumed that actual core area was 98% of theoretical, and that 2% reduction doesn't really matter. (I included that 2% in my spreadsheet btw.)

Checking data of Waasner EI cores (and also Vacuumschmelze for c-cores) the 98% is too optimistic.
I see about 92-93% for Waasner EI cores, which is a reliable source.

[pieter t]

Originally Posted by Tony

^the width of the lamination is 6 inches in the drawing therefore the tounge width is 2 inches....

stacked to 4 inches and you get a core with 8 square inches, stacking factor of 95% gives you a core area of 7.6

as per RDH3 you can get a core VA of 1796.......

Waasner catalogue: EI150Nc (stacking height 9.1cm which is a bit less than 3.6 inches) gives 940 watts of secondary power using the best quality 0.35mm grain oriented silicon steel.
I don't see how you would get 1796 watts out of a very slightly bigger core.
Are you sure you saw a 50/60 Hz specification instead of 400 Hz?
I'd rather trust "modern" specifications a bit more than RDH3 data of maybe tens of years ago.

[Tony]

we don't really know the type of core he is using, therefore i tend to be deliberately conservative in my estimations......

the RDH is maybe 50 years old, but i trust the knowledge found in its pages....

[pieter t]

Originally Posted by Tony

we don't really know the type of core he is using, therefore i tend to be deliberately conservative in my estimations......

I'd say you are very very optimistic by expecting almost double the power from that core.... , or was the 17 hundred odd watts a typo?

[Tony]

I don't see how you would get 1796 watts out of a very slightly bigger core.

the volt ampere figure of 1796 is at best estimate, it is rarely achieved in practice, you have your copper losses and the window area that limits wire size and therefore power rating, in my experience, it is the copper limiting the power.....

look at the picture below, it gives a core power rating even greater:

[pieter t]

IMO that transformer calculator is not correct, the max power is much too high for that core. 220VAC with 16.7A primary (apart from winding space)? Forget it.

I just checked RDH3.
For core area in inch² the RDH3 formula is A = root of W / 5.58.
W = voltage/ampere (secondary power).
For a 940 watts core (the EI150Nc) with this RDH3 formula A should be 5.5 inch²; in fact the core is 6.5 inch², but this is for 1.6 T (Waasner specification). RDH3 formula is for 1.24 T.
Number of primary windings for 220V/50Hz is 224 with 1.24T according to RDH3.
Please note that RDH3 indicates 1.24T (80000 maxwell per inch²) for (at that time) good quality core material to be used for radio equipment!
That already indicates that core materials were lower grade at that time.
We discussed this already time ago: you can't use these pre-historic formulas for nowadays core materials. After all manufacturers supply the formulas which apply for todays materials, so use these.

[johansen]

what is missing here is power density follows flux squared. there's no way in hell you can get 3600 watts from a 51 by 119 mm core, at .9T!! At 1.4T, sure, in fact it would be about 95% efficient assuming the core loss is under 100 watts and you can stuff it 75% full of copper, which means #10 wire for the primary.

[pieter t]

To get that power, you need an EI 231c, M6-0.35mm quality (this core weighs 23.2 kg, which is about two times the weight of the EI150 with 119mm stacking height):

According to Waasner (EI 231c):
- secondary power 3890 watts;
- induction 1.7 T;
- iron loss 69 watts;
- copper loss 58 watts;
- efficiency 97 %.

[Tony]

Originally Posted by johansen

what is missing here is power density follows flux squared. there's no way in hell you can get 3600 watts from a 51 by 119 mm core, at .9T!! At 1.4T, sure, in fact it would be about 95% efficient assuming the core loss is under 100 watts and you can stuff it 75% full of copper, which means #10 wire for the primary.

yes if you stuff all those coppers into a 1 inch x 3 inch window....primary and secondaries.....

i am using cores which i do not know the characteristics let alone the manufacturer they are 0.5mm thick....