Output transformer matching to active device driving it

[macrohenry]

I'd like to understand the relationship between the impedance of the primary of an output transformer and the active device driving it.

Say you have an NPN transistor with the collector connected to Vcc through the primary of a 1000:8 ohm output transformer and the emitter connected to ground through a current limiting 1 ohm resistor. Must the impedance of the transistor match the impedance of the transformer primary? Or is impedance matching irrelevant in this direct configuration?

Alternatively, let's say you have the collector straight to Vcc and the emitter connected to GND through a 1K resistor, with 1K section of the transformer connected to emitter and ground. In this case would the 1K:8 ohm transformer be matched to the 1K impedance it's in parallel with?

Macrohenry

[darkfenriz]

If you want to go retro and use an output tranformer, then primary impedance with Vcc should fit to transistor's safe operation area (SOA) ratings, which you can find in datasheets.

[BudP]

darkfenz.

Regardless of the reflected impedance from load into the primary, you must also provide enough inductance in the primary to match the total reactive impedance. Common formula is (Ra + Rl) + DCR . The Ra is the transconductance of the driver device and Rl is the reflected impedance to the device from the load. Then, since this is the inductance that will give you half power at your frequency of interest, you need to multiply the inductance by a factor of 3.63 to get 0.5 dB down from i K.

You can use a transformer designed for tubes, by the way, and run your SS devices at the 150 vac drive voltage common to most tube OPT's, SE or PP. You can also use an interstage designed for tubes and SE to PP or PP to PP can be a much better interstage device than a capacitor, especially if you are isolating all of your stages.

If you get a high performance audio transformer and use it's capabilities ,you will find the difference between tubes and transistors disappears sonically, and the SS circuit, because of it's lower distortion will have an edge.

Bud

[macrohenry]

Thank you. If I understand you correctly, this would be more along the lines of the first case I presented, right?

Therefore I take it that, in this case, that it's unnecessary to try to match the 1k impedance of the transformer to the impedance presented by the transistor and current limiting resistor.

And in this case, to obtain any desired output level, I can select any transformer impedance with the proper turns ratio to match my speaker. And to get maximum output, I pick a transformer primary that will allow maximum safe current, then pick the turns ratio to match the speaker.

Is it really this simple?

Mac