[Prev][Next][Index][Thread]

Re: Self-built power transformer (some design parameters)



 Hi Reinhard, all:

First, as I said, the info I presented was taken from "The Radio Amateurs
Handbook", so your comments are directed at the authors, not me;-) 
The edition I grabbed first was a bit old (1950) so I looked at a later
version and their recommendations are somewhat less conservative but I
don't know if that's due to a change in philosophy or a change in the core
materials available, or... The later editions give less detailed info on
the core.

The 1983 edition gives a 'conservative' figure of 1000 cm/A and 700 cm/A
for 'amateur' service.

Here are my counter-comments/questions:

>   Suggested parameters for a 1KW transformer:
>    Primary wire size:  #10 AWG
> 
> What !!!?? Are you kidding ? AWG 10 for a 1KVA xformer?! AWG is 0.1019"
in
> diameter. I think this is waaaay toooooo large. I believe normal house
wiring
> is somewhere around AWG 10-12 (or even AWG 14?) in America. What gauge
would
> you use for..... say a 10kVA pig or even a 25kVA? That would be copper
bars!!
> (Something like AWG 000 or AWG 0000) I use AWG 14 for a 7.5KVA xformer,
so I
> think a minimum AWG gauge of somewhere around 17-19 would be quite big
enough
> for a 1KVA xformer. Take a look at some 1.65kVA MOTS. They donīt use AWG
10!!

The most common wire house wiring sizes here are #14 (15A), #12 (20A) and
#10 (30A), but those figures are only for 3 or fewer conductors in a
conduit. If you stuff more than that in a pipe, the ratings go down by 20%
to limit temperature rise. Wrapped up in the center of a transformer, even
with oil cooling, I suspect they would need to be derated even farther.

For a circuit 'feeding' a 10KVA pig (42A/240V) I'd use #6 or #8, depending
on the type of insulation and the length of the run. I'd feed a 25KVA pig
(105A/240V) with #1 or #2. As for what they use to wind a 25KVA pig, I
don't know for sure, but I suspect that it is more like a bar than a wire.
Anyone been inside one care to comment on the size and shape of the
conductors?

What voltage is your 7.5KVA primary running on - 240? That's 31+ amps, an
awful lot for #14 wire. I suspect that you can get away with it because of
the duty cycle, but how much are you wasting heating the core? And what is
the primary voltage of the 1.65KVA MOT? 

> 
>         Primary turns:  132
> 
> This sounds okay, (for AWG 17-19) could be more tho, if you really use
AWG 10.
> Otherwise your "little" xformer will pull substantial amounts of pure
> magnetizing current.

Isn't the magnetizing current strictly a function of core size, core
material and turns? Why would wire size make a difference here?

> 
>  Core cross section: 2.5" x 2.5"
> 
> The cross sectional area of the core is usually defined as the CSA of the
> largest leg. Is this what you mean? or do you mean the whole xformer
core?
> 
> The core I plan to use (for my 7.5kVA xformer) has a CSA of 2.75" x
2.36". It
> was originally used in a stick welder with a 8kVA output.

The CSA is the area of core that actually passes through the windings. In a
typical 'E' core the CSA is the area of the center leg. The outer legs are
each half of that. The required CSA depends on the core material too
though.


>  Quite a large chunk of iron. (Don't forget that five 8-turn windings in
> parallel is not equivalent to one 40-turn winding of larger wire.)
> 
> Yes, Mike this is true, very true. Your second comment is of utmost
importance
> to Doug. 
> But I think you hit yourself with this comment. Five parallel eight turn
> windings CANNOT be equal to a 40 turn winding. Ohms law would tell you
that it
> would only be a fifth of an 8 turn winding or 1/25 of a 40 turn. The AC
law is
> similar. In series they would be similar to a 40 turn winding, but not
exactly
> the same because of:
> 
> a.) The different wire size (you mentioned)
> b.) The losses involved, which shouldnīt be easy to calculate, BTW.

I'm not sure I follow you here. What would be a fifth of an 8 turn winding
(1/25 of a 40)? I was thinkng in terms of inductance and magnetic effects,
which should be (roughly) equal for one 8-turn winding of larger wire or
multiple 8-turn windings of smaller wire in parallel. 


> Sorry, Didnīt mean to hit you that hard.

That's quite alright. It's nice to know that someone is actually reading my
posts! Whenever I join a new list I try to add whatever I can, even if it
stirs up some controversy, before I start asking for help from others.

- former and (hopefully) future coiler greetings from DesPlaines,
Illinois...

- Mike