Original poster: Terry Fritz <vardin@xxxxxxxxxxxxxxxxxxxxxxx> Hi, At 05:45 PM 12/15/2005, you wrote:
How close does a good simulation program like FANTC allow you to get? Would it be possible to do the "prototype" via simulation rather than winding a primary twice?
Most of the programs can figure things out to within about 1%. "Theoretically" they are really really good. However, the data you give them has to be equally as perfect. Every measurement and every value has to be very accurate. I have made coils were the programs worked super well at figuring out everything.
However, it is not easy measuring everything and double checking it all well enough to "challenge" the program's accuracy... For example, the average secondary diameter is actually taken from the center of the wire. You have to use paper to wrap around it and mark the distance and then find the diameter and subtract the wire diameter to get that measurement "perfect". The measurements were done with a 0.001 inch caliper and a magnifying glass... The primary capacitance needs to be known to better than 1%. I use a digital camera to photograph the secondary and then blow the pictures up on the computer screen to count the "exact" number of turns... So basically, it is a matter of "perfection in, perfection out"....
Of course, "normal" folks don't need super high accuracy, 5% or 10% is fine and the measurements can be fairly lax to get that. It is far easier to simply move the primary tap around a little rather than spend hours going over the coil with calipers.... The primary tap point is also influenced by what size room the coil is in and all the primary wires to the cap and gap also "count" as inductance. So even if a coil is perfect in the garage, it may need a little primary adjustment out in the driveway.
So the computer programs are for all practical purposes "more perfect" than the "real world"... 99% of the folks here will not be able to verify any errors in the programs because the programs are better at calculating than they are at measuring.
One certainly could make a coil and tap it so it runs just right the first time (I have ;-)) But it is one heck of a lot of work!!! the programs are really meant to get darn close so once the coil is built the final fine adjustment is really easy. They also spot problems like Malcolm Tesla had where is primary was not going to be big enough. It is sure nice to find that out when he did, rather than when everything was all ready built!!
The only "loose end" I can think of is we don't know, for a given coil, when the coupling will be too high so that racing arcs start to occur. Thus, we always suggest to be ready to raise the secondary a little if needed.
Cheers,
Terry