OP-amp buffers vs. Transistor buffer

Started by Ethan, September 20, 2005, 01:18:59 PM

Previous topic - Next topic

Ethan

Why do some manufacturers use transistor buffers instead of op-amps.? Is it mostly a matter of cost?  For example: Since the tuber-screamer uses an input and output buffer, why doesn't it just use a cheap and simple dual op-amp for in and output buffers? Or one quad op-amp for everything. It takes less parts,  and doesn't even the cheap op-amps deliver better performance than transistor buffers?
-ethan

Seljer

layout problems maybe? with just one dual opamp you'd have to run everything back into the same place for the buffers

Johan

Quote from: EthanFor example: Since the tuber-screamer uses an input and output buffer, why doesn't it just use a cheap and simple dual op-amp for in and output buffers?
-ethan

..just a guess..perhaps op-amps where not cheap in the seventies when the ts was designed...today, an opamp cost little to no more than a transistor, but I'm guessing it wasnt always so..

johan
DON'T PANIC

Roberto

Quote from: EthanWhy do some manufacturers use transistor buffers instead of op-amps.? Is it mostly a matter of cost?  For example: Since the tuber-screamer uses an input and output buffer, why doesn't it just use a cheap and simple dual op-amp for in and output buffers? Or one quad op-amp for everything. It takes less parts,  and doesn't even the cheap op-amps deliver better performance than transistor buffers?
-ethan

A single transistor is cheaper, smaller and have less noise. Layout is very simple for a single transistor buffer. The original BOSS OD-1 was made with a quad opamp but they switched soon to a transistor buffer.
[

Sir H C


Ethan

For low current what about tl062?  A single transistor has less noise than an op-amp?  Is that really true?  Are these audible differences? Ya know what? I am going to wire up a an emitter follower and unity gain op-amp buffer and a/b them through my old Marshall (noise gauge).  I will then report back.
-ethan

davebungo

Any number of reasons:
- Not all op-amps make great buffers especially if for the final end-stage of an effect where it may have to drive a relatively large capacitive load in the outside world.  
- If a spare op-amp stage is handy and it is capable of the job then great use it otherwise it may be cheaper and require less board spave to use a tranny and a couple of r's and c's.
- power (already mentioned)
- remember that even op-amps have transistors in the output stage...
- add more here...

Eb7+9

... one big diff is transient response ... op-amp circuits loose lock when you hit 'em with a large or fast enough transient - and then recover through a ringing process ... depends on conditions: loop gain, phase margin of op-amp used ... combine this ringing with the dielectric absorption of signal caps and you can get midrange mush ...

transistor followers don't ring - they give better transient performance but worse THD ... usually with a predominant 3rd harmonic ...

other diff is drive impedance - although not constant in op-amp feedback ckts it's theoretically a lot lower than what you can get from tranny buffer ... high drive (low Zout) helps immunize line against noise pickup (competing external fields) ... if you don't need super high drive then transistor followers are fine if not better sounding for music signals ...

in some cases, op-amp buffers can be quieter than single transistor stages that are biased hot to produce a lower Zout ... it varies in each design ...

~jc

moosapotamus

It's been discussed before... an opamp is the best way to go for a "clean" buffer. So, maybe it just depends on the design goals for the specific circuit in question. For a fuzz/OD circuit, cheap and dirty may be just fine. For something that is intended to preserve more of the instrument's natural tone, there are opamps out there that will do that more simply and cost effectively than discrete circuitry.

~ Charlie
moosapotamus.net
"I tend to like anything that I think sounds good."