Re: [TCML] How linear does a ballast need to be

[David Dean <deano@xxxxxxxxxxxxxxxx>]

Hi Jim

On Thursday 09 July 2009 01:02:56 am Jim Mora wrote:
> It is well known that MOT's saturate at ~100 volts. NOT a good thing for
> current control! It has been stated here that there should be 1.5volts per
> turn and 3amps per square inch of cross section of the core.

I believe you mean turns/volt. What size core is that for? As for the 3A/square 
inch, I believe that is a conservative figure based on ideal transformer 
design. In real life, based on cores I have measured, 6A/square inch is 
typical.

>My MOTs have 81 turns when it should be closer 180t/120v=1.5v/t so (2)

81 + 81 = 162t/120v = 1.35t/v
My 3.5in^2 core requires 1.25t/v to stay linear. With 10 turns starts to 
saturate after 8V, 10t/8v = 1.25t/v * 120V = 150 turns to stay linear.
See the graph at:
http://deanostoybox.com/projects/ballast/saturation-curve3.5.png

>primaries in mutual series is perfect for 120volts if they are similar to
>mine. 

Definitely. See the comparison at 120V of 15/30 NST shorted, and  MOT core with 
1 primary also with 2 primaries:
http://deanostoybox.com/projects/ballast/vi2.png

>4 in series on inversely stacked "E" cores should be perfect for 
>240volt operation! 

Unfortunately I only have the two primaries. Have some 11AWG magnet wire though.

later
deano
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla