SOT: Naive question about a simple amp idea

[brett]

Hi
Looking at simple schematics for amps, I found this for a Class B amp:


I don't look at many amp schematics, but hadn't seen anything like this before.  To use an op-amp to switch the offset between +0.7V and -0.7V seems a cool idea.  But is there any problem with it?

My idea for a simple guitar amp is to use:
one (half) TL072 op-amp or similar as a pre-amp (voltage gain up to about 200)
a 24V, 1A wall-wart for a +/- 12V pwer supply,
the other (half TL072) op-amp as the offset switch,
a pair of 2N3055/2N2955 transistors on an appropriate heatsink.

I anticipate feeding the power section up to 8V RMS and drawing up to 1 amp, for 8W output.  That seems quite good for such a simple amp and power supply.

Any advice will be greatly appreciated.
cheers

[Joe]

It will work if the opamp can supply enough drive current for the output transistors. 3055/2955's are probably overkill (try TIP31/32). Crossover distortion in this thing is dependent on the slew-rate of the opamp. It should be more noticeable at lower volumes, where the distortion takes up a greater percentage of the output signal.

[brett]

Hi
I'm kinda surprised but  that this will work.
With a slew rate of 16 V/uS, the response of the TL072 should be quick enough to switch the two transistors (ie in less than 1 uS).  But it might overshoot a bit, I suppose.

Thanks for the hint about the 2N devices being overkill (and low hFE).  With a hFE of 100 at 1 amp, a TIP31 only needs about 10mA.  I'm figuring that a TL072 can produce that OK (10mA from about 10V is equivalent to about a 1k ohm load).

Any advice about how sensitive this will be to power supply ripple?  I was thinking of 2 x 2200uF caps on each rail, based on the rule-of-thumb that says you need a minimum of 2200uF per 1 A.
Thanks.

[gez]

A (suitable) resistor from the output of the opamp to each base of the tranistors should make it more stable/bias better.   I've seen what I just described on a schematic for a headphone amp that's on the net somewhere.  If I have some time later today I'll try and find it/link to it.

[DDD]

Adding single 100-500 kOhm resistor from bases to emitter will make the amp much more stable with "more headroom" from the point of view of symmetrical distortion.

[gez]

Couldn't find the schematic, but this page has some useful stuff under 'Class B and AB Symmetric Emitter Followers' (half way down page): 

http://www.headwize.com/projects/opamp_prj.htm