Class B output power

[lt cdr data]

how do I work it out? for

2sc3858 with 70 volts psu?

thanks

[richie00boy]

What emitter resistors are you using and what load impedance?

Just work out total resistance in series with the load and then it's simple ohms law and potential divider theory assuming fully saturated transistors.

[lt cdr data]

Hi Riche, its 0.22 R, 8 ohms.

hope that helps.

Just wondering what I can get if I cool the output stage down, or then put it into class A.

[richie00boy]

Temperature or class is not a factor, that will just dictate how many output devices you need to cope with the heat and standing current.

[lt cdr data]

so for one device its .22r in series?

what about the series inductor etc.

amp has 4 parallel devices on each side....

so its 8/ 8.22 times 70volts minus drop across transistor and emitter resistor(10mv)?

[richie00boy]

it's simple ohms law...

work out parallel combination of emitter resistors
put this in series with load impedance
apply rail voltage (subtracting Vbe loss if you know it, otherwise guess at 5V)
what current flows

calculate power in load
rms power is half this

[lt cdr data]

ok assuminig 1 pair of devices, its

very close to 70-5 volts/ 8 ohms to give the maximum current for 8 ohms? about 8 amps

power =65 x that current? 512 watts ish...

And then half it for rms? 256?

why half it and not 0,707?

Where does the efficiency of the output stage come into it?

I am used to doing this with tubes, but transistors seem a little different.

If I turn the bias up, how do I tell when its all in class A? Power will reduce by about 2/3 I reckon.

std. Bias current is 45 milliamps.

thanks

[richie00boy]

Almost right but you have too much current because you have not included the other series resistances like I said.

RMS = peak*0.5 not 0.707 because it's power not voltage.

Output stage efficiency does not come into this calculation, that will be realised later when you add in the quiescent current and the total power drawn from the supply.

Turning up the bias to say 1 amp gives you 2 amps swing of class-a, assuming normal push-pull output stage.

[Christer]

The reason you get half the power is that for power you are multiplying voltage with current, and these are either both peak values or both RMS values. If you multiply two RMS values you get 0.707*0.707 = 0.5 as compared to the peak values. Hence, RMS power is half the peak power.

[pinkmouse]

Originally posted by richie00boy
Turning up the bias to say 1 amp gives you 2 amps swing of class-a, assuming normal push-pull output stage.

And, with 70V rails, a huge amount of heat!