[Prev][Next][Index][Thread]
Re: Variacs in parallel ?
Tesla List wrote:
>
> Original Poster: "Paul Eugene Kidwell" <tmb-at-ieee-dot-org>
>
> Hi Again!
>
> I've been thinking about the 3 phase diagram that was posted a few
> days ago. Here's the original...
>
> > > For a 3-phase variac you'd need chokes between pairs in a tree
> > > configuration:
> > >
> > >
> > > W1 ------------
> > > O
> > > o-----------
> > > 0 |
> > > W2 -------o---- |
> > > | |
> > > | o----------- Vout
> > > | |
> > > | |
> > > ----- |
> > > O |
> > > o-----------
> > > 0
> > > W3 ------------
> > >
>
> I'm planning on paralleling 3, 30 Amp Variacs, and I'm trying to
> figure out exactly how to wind them. So (assuming a perfect world :)
> I started plugging in some current numbers...
>
> Chokes
> V
> W1 -30--------30----
> O
> o----45-----
> 0 |
> |---15---- |
> | |
> | |
> W2 -30----o o---90------ Vout
> | |
> | |
> |---15---- |
> O |
> o----45-----
> 0
> W3 -30--------30----
>
> You'll notice that the the chokes have 30 amps comming in on one
> leg, and 15 comming in on the other. To me, this doesn't seem quite
> right. Someone else suggested an arangment that looked like this...
>
> Chokes
> V
> W1 -30-o-------15----
> | O
> | o----30-----
> | 0 |
> | |---15---- |
> | | |
> | | |
> W2 -30-----o |
> | | |
> | | |
> | |---15---- |
> | O |
> | o----30----o---90------ Vout
> | 0 |
> | |---15---- |
> | | |
> | | |
> W3 -30-----o |
> | | |
> | | |
> | |---15---- |
> | O |
> | o----30-----
> | 0
> |-------15----
>
> Here we're using 3 chokes, but the ends of the chokes are balanced,
> and the current numbers are quite a bit lower. Low enough that I
> don't have to use 10 gage wire to wind them. I could get away with
> using 14 Gage wire, but I'll probably go with 12.
>
> I've also managed to dig up a matched set of 3 solid steel cores.
> (a couple of days ago, someone suggested solid cores rather than
> laminated ones. This seems right since there's little magnetic
> field generated in the core) So I'm all set to start winding.
>
> I'd *still* like some idea of how I can check to see if everything
> is working the way it should.
>
> Later!
> Paul
Paul,
Using three chokes in the above configuration would be preferred for
better load sharing, and it indeed be possible to use a smaller wire
gauge. However, if you are going to "wind your own", I'd still stick
with laminated core material to keep the cores from overheating due to
eddy currents.
Following is another trick that variac manufacturers use to efficently
balance currents across large numbers of variacs. Instead of winding
chokes with a single, center-tapped winding, make them with two
independent N-turn windings to make 1:1 transformers. Simply wind "N"
turns and bring out both ends, then wind another N turns in the same
direction. The starting ends of each winding will be those marked with
the polarity dots below.
By connecting them as independent 1:1 transformers, you can then
cleverly interconnect them to do true current balancing between wipers
with NO circulating currents. Current balancing acoss all the variacs is
done elegantly via transformer action. This method requires "N"
transformers for "N" paralleled variacs, the transformers must be
magnetically independent in order to eliminate circulating currents, and
the wire gauge must be large enough to handle full load current. BTW, I
didn't figure this out on my own... I got it from the clever folks who
make Powerstats - this is taken right out of their brochure on Model
1256D Powerstats. :^)
W1 W2 W3
^ ^ ^
| | |
| | |
| |---------------o--------------o-------> OUT
|. | |. | |. |
OOOOOOOO OOOOOOOO OOOOOOOO
-------- -------- --------
-------- -------- -------- .... etc
-------- -------- --------
OOOOOOOO OOOOOOOO OOOOOOOO
|. | |. | |. |
| ---------- --------- |
| |
|-------------------------------------|
Safe coilin' to you!
-- Bert --
--
Bert Hickman
Stoneridge Engineering
http://www.teslamania-dot-com