How bias is developed for the driver/final... (AM/Class C only at this time!)

[guitar_199]

I can not COUNT the times I have been told.... "the only stupid question is the one you don't ask"!!! I am here to test that theory.

I am trying to strengthen my knowledge in the particular area mentioned in the subject line. It is one of those things that, at one point, I THINK I understand it, but then I look at a schematic and just can't get my head around it. I know it is probably simple... but when I go out and search around on the net for "training/educational materials" on it.... I never find anything that really seems to describe it.

I know that there is generally no "fixed bias" voltage applied... rather it is a bias that is developed when the mic is keyed and the oscillators/mixers kick in and RF is sent down the transmit chain. So I would imagine that, if the mic is UNkeyed and I measure the base of the driver and final... they will likely be zero. Now, on the other hand, if I check service manuals, when I key I would expect to find a slight negative value on the base but NOT enough to turn it on such that it is actually getting biased on by the RF signal coming down the transmit strip... which gives us our Class C operation. The driver/final are NOT ON for the whole wave form but, rather, they come on as the rf signal is climbing on it's sinusoidal cycle. SO this means that what actually gets amplified and pushed out are 'upper peaks' of the incoming signal... and it goes out more like pulses.... that feed into the tank/LPF. Sounds good.

But WHERE was that slight negative bias developed? Where did it come from? And possibly... what can go wrong that could cause bias to fail to develop???

To be dead honest.. my big paragraph description ... may be totally wrong... but that is the way I currently understand the whole "here comes the RF, now a LITTLE BIT of bias gets developed, and the signal pushes through the PARTIALLY BIASED transistor and creates pulses" thing.

I am going out on a limb in suspecting that others here might benefit from this same knowledge and just haven't said anything. So I am GOING TO SAY SOMETHING!!!!!

Can any one refer me to any reading material that describes step by step ... how this works!!! ( I will be GLAD to go do my own study if I can find it... but so far I haven't found anything that really describes what is going on in the driver/final area say.... in a Cobra 29!!!) . Failing that, I would welcome any comments, suggestions....... even a swift kick..... that puts me in the direction of understanding this whole mystery!!!!

I am sure that, if everything is working... there is no component failure.... the transistor is healthy.... then the bias WILL DEVELOP. I just don't understand how. And this makes me feel so inadequate!!!!!

Again, I welcome comments, suggestions, cat-calls, name calling..... ANYTHING.... if it help!!!!!

(Serious here: thanks in advance for anything anybody might throw my way!)
Bob

 

[brandon7861]

This is my understanding of it... (cobra 29 LTD ST schematic)

The base-emitter junction is a diode. Although forward current works to turn on the transistor, forward (positive) voltage is clamped at around .7v. But only the positive peaks are clamped, the negative peaks remain untouched.

Imagine an RF signal centered on 0v, but instead of a sine wave, the positive peaks are clipped. The waveform's average is no longer at 0v, its going to be lower because those clipped peaks shift the average down below 0v. That's your negative DC bias and the 22ohm resistor controls how much of it you get. The ferrite bead above the 22ohm resistor prevents the RF from seeing ground.

I believe this is called self-bias.

Edit to add: Lou's book specifically talks about the cobra 29 final section and uses the term "base-leak". In that schematic (which must be the plain old cobra 29), he calls R86 a Q spoiler. That schematic shows an inductor-style RFC at the base instead of a low-Q bead, so using a resistor to spoil the Q makes sense. I would say that resistor is dual purpose in that it controls negative bias and also reduces the Q of whatever impedance keeps the base above RF ground.

 

[brandon7861]

I really wish Lou was active on the forum because I believe there to be a head scratcher in this section of his book.

In the cobra 29 final section (page 152 paragraph 6), he says lowering the value of the Q-spoiler resistor will increase stability. That seems backwards to me. Less resistance means less negative bias developed, and that means more power output. Also, less resistance in series with an inductor increases Q. It would make more sense that raising this resistance would increase stability because raising it not only increases the amount of negative bias developed (which reduces duty cycle and power output and the chance of feedback), but it also adds resistance in series with a potentially self-resonant inductor.

Maybe its just too late in the evening to think clearly, IDK. I hate to question a legend.

Edit: I'm gonna assume this was one of those 3am not enough coffee paragraphs because in the same one he mentions R86 also serving as a load for the preceding driver stage (which isn't possible with the choke in the way, no RF gets to R86). Those 10 or 12 pages are absolute gold on final sections though. Great book!