[Sneap] High Voltage Isolated Power Supply (correction)

[ThunderHooves]

Sorry.

2.4 volts peak per turn below should read 0.24 volts peak per turn. The 0.17 volts rms is correct. I think all the other values are approximately correct.

Gregory C. Harper               Phone:(206) 543-4084 (Office)
Research Engineer/Scientist IV        (206) 543-4062 (Lab)      
University of Washington        FAX:  (206) 685-4634
CENPA NPL Bldg.
Box 354290                      email: gharper at u.washington.edu
Seattle, WA 98195-4290

On Tue, 18 Sep 2007, ThunderHooves wrote:


> You want what to be portable? The transformer itself, the power supply and transformer, the entire deck? None of these are really specified.
>
> Looking at your request, I think you are asking for a 120VAC primary, some lower secondary VAC, and 50kVDC isolation primary to secondary. You plan to make a low voltage DC power supply of order 1W using the isolated secondary. The deck is at some DC elevation of order 50kV.
>
> This is not too tough to make if you can't buy it. If there is a local transformer shop contact them about what is required to make the transformer without isolation. Ask for a rectangular core and circular windings. The windings can be made of small gauge wire because the power requirement is so small. Then tell them the core you want has to have legs long enough to provide 5" of track length on each side of the winding to the top and bottom of the core. That's 10kV per inch track length on polyethylene sheet. You also need to stuff at least .050" of polyethylene between the primary and secondary so the gap between the windings is important. That's 1kV per mil dielectric strength. A wrap of .050" of polyethylene around the secondary ensures that the secondary doesn't arc to the opposite leg. In short, the secondary must look along 10kV per inch before it sees the top and bottom of the core.
> It must also look through 1kV per mil of polyethylene before it sees the primary or the opposite leg of the core.
>
> The core can be made of Tempel 29 gauge (0.014") sheared sheets of silicon steel or equivalent. Stagger the strips at the corners to overlap completely. I make core clamps out of aluminum angle. Remember, no conductors other than the windings encircle the core.
>
> Tempel saturates at about 1.6 Tesla so designing it to operate at 1.0 Tesla is way safe. If you want a 1" cross section, that is 6.4e-4m^2. Phi max is area times B max or 6.4e^-4 times 1 tesla or 6.4e^-4 webers. Omega is at 60Hz so dphi/dt=omega phi max or 2.4 volts peak per turn which is 0.17 Vrms per turn. For 120 VAC this is 700 turns for the primary. A small gauge wire will constitute some significant resistance. The secondary turns are determined by the desired voltage.
>
> The exciting current depends on the magnetic field path in the core. This is a long core and it will consequently take some exciting current. If the legs are 12" long and the top and bottom are 4" long the path length is 32" or 0.8m.
> The exciting current ie=BL/uNp where u for Tempel is 0.004. Then ie is 0.28Ap.
> Play around with the dimensions and wire gauges until you are satisfied.
>
> -Greg
>
> Gregory C. Harper               Phone:(206) 543-4084 (Office)
> Research Engineer/Scientist IV        (206) 543-4062 (Lab)
> University of Washington        FAX:  (206) 685-4634
> CENPA NPL Bldg.
> Box 354290                      email: gharper at u.washington.edu
> Seattle, WA 98195-4290
>
> On Tue, 18 Sep 2007, Chris Westerfeldt wrote:
>