Dan,
I have built approximately 50 RF transformers in the past several
years. There are 2 types of loss in these circuits: dissipative loss
(think heat) and mismatch loss (think uncontrolled leakage inductance,
wrong transmission line Zo in the transformer etc.). Dissipative loss in
the transformer consists of copper loss in the wires and heat in the
ferrite cores. Generally mismatch loss is the problem. In your case
the Zo of the 2 transformers needs to be 6.26 ohms and 25 ohms, leads
need to be short, and the ferrite choking impedance needs to be large
enough to balance the amp output currents. Are you sure that if you
loaded your amp with 3.125 + j0 ohms (50/16) you would be happy with the
efficiency? I generally use a 50 ohm (or 25 or 12.5 ohm) dummy loads
connected to a large L network (very low loss) and use a VNA to get the
desired impedance at the connection point to the amp. Then you can see
if your amp would work better with a load that is slightly reactive (it
probably will).
Approaching this problem in a logical manner will show you where the
problem is.
73,
Larry, W0QE
On 7/22/2014 11:49 AM, Dan Simmonds wrote:
Has anyone built an output transformer for a 50V LDMOS FET (Freescale
device) that could lend some help with this?
By my estimation a 16:1 constructed of (2) 4:1 Guanella’s should give
a satisfactory match to 50 ohms. The XFMR I've built currently only
gives about 33% efficiency.
Thanks
Dan KK3AN
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|