EMRFD Message Archive 1726

1726 2008-06-19 01:57:24 Stan Link coupled output question REF Solid State Design page 25 Message Date From Subject Ref Orginial Solid State Design book Page 25 figure 16 In the calculations there are two items which are assumed. One the load Q which was chose to be 6, I understand that one. The other value assumed (chosen) was the number of turns on the resonant coil. The value of 15 turns was chosen which resulted in an impedance of 99 ohms. I understand all the calculations and other items for the circuit. My question is the 15 turns. Was the value of 100 ohms chosen and that resulted in the 15 turns. Or for some reason the 15 turns was chosen (frequency was 14 MHz). Anyway I am looking for what the rule of thumb might be on which ever was chosen the turns or impedance. The RL for the circuit (1/2 watt at 12 vdc) is 144 ohms. Comment: I do not see this link coupled output circuit used today, but still feel it is the better design for a small qrp transmitter. Thanks, Stan ak0b Hi Stan, and group, I doubt that there was any specific reason for picking the 15 turn winding. We obviously can't go back and ask Doug. I think it was his piece, abstracted from a QST piece. If I did this one, it may just have been that I looked in the junk box, found a 15 turn toroid, and ran through numbers to indicate that "yea, that will work." A link coupled tuned circuit like this will offer poor harmonic suppression when compared with a low pass type matching network such as pi or L network. It is usually a reasonable assumption that impedances transform as the square of the number of turns, but it is still just an approximation, especially with powdered ir I agree with Wes. Point number 1, the output impedance of a transmitter stage set by the usual Vcc^2/2*Pout. This too is an approximation. Knowing your impedance then you must select your loaded Q. If you want a loaded Q of 6, then multiply the Output Imedance by 6. This value should be used for the XL value and you would then determine how many turns it would take to get the inductance that would yield this value. Finally, you would figure the impedance ratio between the transmitter output load and the antenna of 50. The turns ratio would SQR(Impedance Ratio) and thus the number of turns. Ken - AE1X Can't resist ranting a bit: Kenneth Stringham clearly states right at the top that Vcc^2/(2*Po) is an approximation. Good job. This equation is an effort to put a linear spin onto a very non-linear part of a power amplifier. I think it is not stressed enough that our often-used and powerful network math doesn't apply here. We make the mistake of assuming that the filter between power amp and antenna operates as it would driven from a linear class-A amplifier having output Z of Vcc^2/(2*Po). A similar caution applies to any non-linear circuit like a mixer and many oscillators. We should not be content to build power amps as we have for many decades, and assume the result is optimum. That Class-C circuits can be pushed toward Class-E efficiency is evidence that there's room for improvement over conventional methods, where an explicit, optimum solution is lacking. There is ample opportunity for experimental methods espoused in EMRFD to make real advances in this area. -Glen