Re: paralleling variacs

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Bert Hickman by way of Terry Fritz <twftesla-at-qwest-dot-net>" <bert.hickman-at-aquila-dot-net>

Hi Steve,

This is a bit different problem since most coilers have tended to use 
matched pairs of variacs. If you use a pair of variacs from different 
manufacturers, there are a couple of new "wrinkles": Differences in 
Volts/Turn and different angular travel from CCW to CW limit. The good news 
is that these should be surmountable. The angular travel (once the variacs 
are mechanically linked) will be limited by the variac having the smallest 
0 to Vmax rotational angle.

Here's how to solve the problem:

1. Mechanically link the variacs and insure that both are simultaneously at 
their zero volt (max CCW) mechanical limits.

2. Carefully power up both variacs leaving the wiper terminals of each 
unconnected. Now adjust the linked pair to the maximum voltage (max CW 
setting) so that at one variac is at it's maximum CW limit. The other 
variac may or may not be at its CW limit.

3. Carefully measure the voltage between the wipers of each variac at the 
above setting. Any significant difference between the two voltages must be 
balanced so that the two variacs can be safely connected in parallel. 
Otherwise the variacs will fight each other, and the resulting circulating 
current between variacs may cause needless heating and potential damage.

4. If the measured voltage difference (Vd) at the maximum CW position is 
small (say 1-2 volts) you can likely directly tie the two wipers together 
and tolerate a little bit of circulating current. And, if the voltage 
difference is small, you can temporarily connect an ammeter between the two 
(with no other output loading) to measure the amount circulating current.

5. A larger voltage difference will likely require using a balancing 
choke/transformer to neutralize it. You can make a center tapped choke by 
wrapping #14 or #16 magnet wire or #14 AWG THHN or equivalent house wire 
around a junk laminated steel core or a relatively large ferrite or iron 
powder core.

The total number of turns you'll require will be a function of the core 
size and voltage difference you need to neutralize. You can estimate the 
number of turns if you know the core's cross sectional area using a form of 
the transformer equation:

     a = core area (square inches)
     f = operating frequency (Hertz)
     B = saturation flux density (use a ballpark number of 1.5 Tesla
         for silicon steel laminations and 0.3 for a ferrite core)
    Vx = 4.44*f*B*(a*0.0254)^2  (Volts/Turn)
   and:
    Turns = Vd/Vx (minimum number of turns - more is no problem)

   Example:
     Let a = 1 square inch core, f = 60 Hz, laminated steel core
     and lest Vd = 5 volts
   Then:
     Vx = 0.22 volts/turn
and:
     Turns = 5/0.22 ~ 22 turns (at least 11 turns on each side of CT)

BTW, you can also use the secondary of a center tapped filament transformer 
whose rated output voltage is Vd or above and whose current rating is at 
least 7.5 - 10 A. In the above example, a 5 volt (or above) 10 amp center 
tapped filament transformer could be used. Leave the primary side of the 
transformer open circuited.

           -----------------
           -----------------
           -----------------
     ------OOOOOOOOoOOOOOOOO-------
     To            |              To
   Wiper 1         |            Wiper 2
                   |
                   |--- Output

Hope this helped.

Happy holidays to you and to the Tesla List!

-- Bert --
-- 
Bert Hickman
Teslamania, from Stoneridge Engineering
"Electromagically" (TM) Shrunken Coins
http://www.teslamania-dot-com

Tesla list wrote:
>Original poster: "Steven Ward by way of Terry Fritz <twftesla-at-qwest-dot-net>" 
><srward16-at-hotmail-dot-com>
>Ok,
>Im sure there are plenty of you guys who have done this!  I would like to 
>parallel 2 of my (well my only) 7.5A variacs, so that i have 1 15A variac.
>Im kinda doubting that this will work, becuase one is a powerstat, and the 
>other adjust-a-volt (or something like that).  But they are nearly 
>identical, and look to have the same wire guage.  They are physically the 
>same.  The one difference is that the powerstat is rated for 135V out, and 
>the adjustavolt 140v.  But the powerstat says 115V in, so im guessing 
>thats why its only rated 135v out.  So, is there any feasability of 
>paralleling these smaller variacs?  I suppose i dont really NEED a 15A 
>variac.  But i was swayed when my new tube coil blew the 15A fuse after a 
>short period of a lower power run!  My other tube coils run at about 
>14-16A, and i always use the little 7.5A variacs (not together).  But, i 
>feel the new dual 833A coil will need something a little more robust.  So 
>am i dreaming, or it is possible to parallel these slightly different 
>variacs?  If so, what do i need to do?
>Thanks,
>Steve Ward.
>