Will a true class B have thermal runaway?

[azira]

Lets define for this post that a class B is a a NPN back-to-back with a PNP in push-pull with no diodes or VBE multiplers so that it will have the 1.2V of deadzone and xover distortion.
Since the transistors aren't biased on with a fixed voltage, and say I use feedback instead to correct the xover, will it have thermal runaway?
--
Danny

[Tube_Dude]

Originally posted by azira
Lets define for this post that a class B is a a NPN back-to-back with a PNP in push-pull with no diodes or VBE multiplers so that it will have the 1.2V of deadzone and xover distortion.
Since the transistors aren't biased on with a fixed voltage, and say I use feedback instead to correct the xover, will it have thermal runaway?

No... in this case you don't have thermal runaway.


Regards

[sreten]

Originally posted by azira
and say I use feedback instead to correct the xover
Danny

Just note the xover cannot be "corrected" by feedback,
feedback will only reduce the crossover distortion, and
due to compensation as the amount of feedback reduces
the distortion will increase with frequency.

It will sound extremely unpleasant.

sreten.

[millwood]

thermal run-away is a characterisics of biplor transistors and have no regard for topology. so the answer in this case is a "yes".

It is that with the cross-over section unbiased, you reduce the risk of thermal run away.

But if you run the transistors too hot (driving heavy load for example), you will get thermal run-away.

[millwood]

Originally posted by sreten
Just note the xover cannot be "corrected" by feedback,
sreten.

what will the front stages do if they are driving an unbiased ops (class C) within a feedback loop? assume of course that all stages are perfect other than the cross-over region for the OPS.

if you apply a small voltage to the input that's within the cross-over at the OPS, there will be zero output. so the front stages will output more until and unless the ops is generating the right output.

This would suggest that feedback indeed corrected the cross-over distortion by working in a non-linear fashion.

Of course, how that works in real life with less-than-perfect front stages will be much more complicated.

[dhaen]

Yup, you'd need an infinitely fast amp working in a less than infininite passband.

Millwood, you surely don't mean class C..
By definition it only conducts for 120 deg.

[Tube_Dude]

Originally posted by millwood


But if you run the transistors too hot (driving heavy load for example), you will get thermal run-away.

Wrong!...in this case you will have high dissipation !And not thermal run away.

If you stop to drive the heavy load the transistors will return
to near zero standing current...and will cool down!
Thermal run away is when the current in the transistor increase...the transistor become hoter ...as the transistor become hoter the current increase...and the process ends with the self destruction of the device.

[sreten]

Of course, how that works in real life with less-than-perfect front stages will be much more complicated.

Its not complicated at all. Open loop gain is easily calculated
giving you the amount of closed loop feedback. This reduces
the distortion but does not eliminate it at high frequencies.

sreten.

[nemestra]

Originally posted by dhaen
Millwood, you surely don't mean class C..
By definition it only conducts for 120 deg.

John, can you give a reference for the 120 deg. figure you mentioned? I have read an article in which an audio designer who was classifying all the different types of audio amplifiers (Class A, AB, B, D, G etc.) has referred to a an output stage with no bias, as Danny mentioned in the first post, as class C. They defined Class B as an output stage in which each device conducts for 180 deg, i.e. optimally biased. Using that definition, any output stage in which each device conducts for less than 180 deg. would be considered as Class C. From a quick glance at my bookself Duncan's, Self's and Slone's books would not contradict this definition, though none of them state it outright in these terms.

James

[dhaen]

I was getting mixed up between the optimum conduction angle for some RF class C circuitry and a real definition.

You (and Millwood) are of course right that class C is defined as "Less than 180 degrees conduction angle".

Aplogies to all concerned

Maybe I should have bid harder on 7V's auction