EMRFD Message Archive 3536

Message Date From Subject
3536 2009-09-05 09:24:22 Kenneth Stringham Re: Twisted Pair Transformers
There is one aspect of this design that seems to be missing here and that is that you are dealing with a transmission line device. The design requirement is that the inductive reactance needs to be 10x the impedance of the load at the lowest frequency for which the device is to be used. Since the specified impedance is 200 ohms, then the inductive reactance must be at least 2000 ohms.

Your design indicates that the reactance of the transformer primary is 1460 ohms @ 14MHz. This means that your transformer will work properly at all frequencies above 20m.

I have made transformers using the techniques you are indicating that work from 1.8MHz. to 500MHz. and then others that extend from 50MHz. to the low microwave frequencies. You must use ferrite cores to achieve high inductance per turn to achieve the lowest frequency of operation. The transmission line behavior extends the performance to the highest possible frequency.

Ken - AE1X
3537 2009-09-06 15:01:29 w1kilofoxtrot Re: Twisted Pair Transformers
Actually, implied, not missing. I was using the 4x rule of thumb, however, which gets me closer to the 3-30 MHz target. The ferrite cores I am using are 0.5 x 0.5 type 43 'beads', if my math is correct.

And, part of the characterizati
3538 2009-09-06 15:24:32 Graham / KE9H Re: Twisted Pair Transformers
All:

Actually, I have found that you can push down to about 3x, at the
bottom of your lowest band, and suffer a little extra loss there, but
you can still get reasonably usable performance at 3x.

You usually having to make some compromises if you are going
to make a transformer work from 1.8 to 54 MHz. (5 octaves.)

And don't be afraid to add some parallel capacitance to resonate
or peak the leakage inductance at the very top frequency of
interest. You end up with a steeper roll off above that frequency,
but you can reduce the loss at, and up to, that corner.