Bill:
I am still trying to come up with a feasible design for you, for a
toroidal replacement 80M coil. Meanwhile, you probably noticed the posting
RE: a coil advertised for sale by WC6W, that was advertised as a MM T400-2
core (with a winding of 20 turns in place.) I was interested in it also so I
did an analysis of this core, and fortunately I have all of the mfrs. data to
work with (what a luxury!) I think this issue and the results I obtained have
some general interest so I am going to post this.
The primary problem in this case if power loss and temperature rise.
The power loss directly affects amplifier efficiency, and temperature rise
due to this loss can and WILL heat up the core. Ferrites are not highly
stable materials, and it takes a very small temperature rise to knock a
high-Q tank circuit out of resonance. (It is a totally different situation in
output transformers of Solid-state amps, which are NOT resonant - a 25% drift
in mu could go unnoticed.)
I calculated the loss in this core under what, I believe, are the
typical operating parameters. Unfortunately, the mfrs. loss data only extends
up to 2.5MHZ and it is not wise to extrapolate. So, I calculated the scenario
for a 2.5MHZ amplifier, which is probably close to reality for 160, and a bit
over-optimistic for 80M. RF losses at high-power increase very rapidly with
frequency. Here are the results, which I am going to state in terms of RF
voltage into the tank circuit, to eliminate any questions about the actual
value of Rplate.
Assuming 25 turns on the T400-2 core, the inductance is 11.2uH, which
puts it in the right ballpark for a 2.5MHz amplifier and Rp of 2K. The core
losses are as follows:
500VRMS - 9Watts 1000VRMS - 41W. 1500VRMS - 91W 2000V - 180W
The temperature rise of this core will be approx. 5degrees(C.)/Watt, which
leads to a temperature rise of 200 degrees at medium power levels. At 3.5MHZ
I would estimate the loss to be almost double the values shown above.
Conclusion: This is a very large and expensive core. The OEM price range of
this core is $50-100. It will NOT be usable in a kW amplifier at 3.5MHZ and
above. It would take a stack of 3 or 4 of them. That is why you don't see
ferrite cores as the tank circuit in high power RF amplifiers.
I will send you data soon on a design which I am 'testing', which uses a more
modern HF material. It may allow building a workable inductor without having
to use a pound of ferrite costing hudreds of bux.
73
Eric von Valtier K8LV
|