Stacking op-amps - pros and cons.

[Fast Eddie D]

If I were to parallel opamps, whether stacked or alongside, I would incorporate an output resistor on EVERY output pin.
Much the same as adding an emitter/source resistor for paralleled output devices.

I may also look to inserting an input resistor to EVERY -IN pin.

Then they're not really "stacked", are they? You have described a textbook parallel circuit. The big problem with this appriooach is that it'll probably work correctly the first time. 😱

 

[ginetto61]

Gutiar players are always looking for new distortions, and apparently some of them like this one.
It seems like a very poor implementation to me. but evidently it must work because people are doing it. I thought it was crazy when I first saw it, but guitar players are always devising clipping circuits and exploiting diode nonlinearities to come up with a new "sound."
The classic "sound" is an overdriven triode. It's the "sound" us older folks knew 30 years ago. The "sounds" in some new music sounds like something is burning up in your amplifier. 😀

Thanks a lot for the explanation.
Actually the context is completely different
Guitar players do not look for fidelity but more for "effects", maybe also nice but hardly faithful to the original signal
Anyway I have understood that the solution is not "sane" for hi-fi
Kind regards,
gino

 

[Fast Eddie D]

Guitar players do not look for fidelity but more for "effects", maybe also nice but hardly faithful to the original signal

I'm well aware of this. But understand that I just don't "get" a lot of the new "music." It sounds like random electronic sounds (like when I stick my DVOM probe in the wrong place 😀 ) set to a very monotonous beat.

Anyway I have understood that the solution is not "sane" for hi-fi

I'm not convinced that it's "sane" at all. There are many distortion circuits I could come up with off the top of my head that do not indulge in such quackery. It must be my training ("brainwashing?).

 

[Dan Moos]

I build tube amps for guitar, so I can't imagine this trick doing anything that I would like. That said....

Just about every sound we like about electric guitar has its roots in non-technically minded players doing things with their equipment that is "not intended". So if some shlub tries stacking some opamps, and gets a unique sound that for whatever wacky reason, he likes, then none of us are really able to say they are "doing it wrong".

This "quackery" has the advantage that anyone with a screw driver and the abilty to determine which end of the opamp matches the existing part can do this. The dc offset problem seems like it would be a roadblock to me, but whatever.

I suspect we will have the last laugh anyway because I doubt this kinda sound will have any staying power, as evidenced that its pretty rare already.

 

[5th element]

This was delivered to my inbox today as part of the TI news letter and I thought of this thread.

Paralleling Op Amps?is it possible? - The Signal - Blogs - TI E2E Community

 

[iaxxaxxai]

It's funny how every single one of these articles about paralleling opamps says almost exactly the same thing.

I guess this thread at least serves the purpose of enlightening anyone who might have entertained the thought of doing the foolishness shown in the original post.

I think the fact this is an idea born from the bootweak guitar pedal scene should be evidence enough that it's not worthy of serious consideration for anything related to hi-fi. So, stacking a bunch of opamps on top of each other makes your Tube Screamer have a more interesting distortion... the word "distortion" pretty much says it all right there.

 

[memofer]

Call me a fool, but then again, that's what DIYAudio is all about, to have fun, right? it sure was fun when I tried it yesterday on my Marantz cd48. I liked what I heard with 2 stacked up opamps an lm4562 and a ne5532.

On my system the LM4565 gives me a nice lows and good highs, but sucky midrange, and the opposite with the 5532, it has a good midrange. so both combined I get both worlds. Hope I can keep it this way! cause I like what I hear 😀😀
The worse it can happen is I burn my d@mm cd player, in that case I'll just go to GW to pick another for cheep!

 

[Elvee]

Make a simple sanity check: touch them (if you dare).
How hot are they?

 

[craigtone]

I remember doing this with FETs on RC cars to increase current carrying capacity with larger battery packs. 😀

 

[spladski]

I imagine that stacking should work out OK and be easier with opamps that have resistors in series with the transistor outputs as part of their design. Most opamps have current limiting so trial shouldn't be deadly. An example that might work is the TLE2142 which I actually intend to stack for a headphone output. While this chip won't win awards for distortion, it has other attributes that make it more suitable for noisy mixed digital and analog environments.

 

[memofer]

Make a simple sanity check: touch them (if you dare).
How hot are they?

I hear you. I just played them for 3 hours and they do get very hot I can't hold my finger on top of them for more than a second, but it seems once they reach a certain temperature, it stays there. also the voltage regulator and transformer get warmer, which to me it seems normal. Most electronics should get at least warm.
I always wondered why my cd players never get at least warm at all.
it just sounds too good to my ears to go back the way it was, I am really surprised.

 

[memofer]

Spladski, give it a shot and let us know how it goes. I am also biasing those opamps into class a with a resistor on each output channel. I really wasn't expecting to hear a difference but I do.. just trying to have some fun here..

 

[spladski]

I have run simulations which show that it is a bad idea to stack opamps in general. If the opamps do not have matched DC offset they will fight each other and have a large static current of opposite polarity. If the current rise is too great it will trip the current limiters. A few opamps will probably be OK like the TLE2142 because it has resistors in the emitters of the output and a typically low DC offset. In this case the static current might rise to around 5mA.

Dissimilar opamps will have a large static current and be pushed into the class B region to deliver it. Each opamp will be in the opposite class B state. A pseudo class A? A 1 ohm resistor in the outputs can be used to determine the static current which can be removed if all is well. So stacking can be made to work with a bit of planning. Randomly stacking different opamps is not recommended.

 

[mchambin]

I doubt you ran any simulation.

 

[spladski]

I doubt you ran any simulation.

The simulations were in TINA. Texas Instruments spice prog. They have the TLE2142 model. I wanted to know what to do if the opamps started fighting each other before actually trying it.

 

[Monte McGuire]

In 'simulation world' every example of a particular op amp model will have exactly the same DC offset, so what you observe will be completely unlike reality. You could of course hand edit the models to randomize the offset values for each amplifier instance, but then you have to decide how to alter the offsets and what a reasonable 'spread' would be.

Regardless, when you construct the circuit, you'll get yet another distribution of offsets, one that may also change over time and temperature, making these stacked circuits unpredictable in practice.

 

[RJM1]

You could always cap couple the outputs.

 

[spladski]

The models will be identical of the same type. However it is fairly easy to synthesize a possible variation without having to adjust the models themselves. Data sheets give you the spread. You can also hand select devices. It is not a done deal that stacking will be unstable. The OPA192 might work straight off the bat. Elvee's main point is that if the static current rises to 30mA then 30mA x 15V = 0.45W. So dissipation is a key factor. If you relax the requirement for a true stack and implemented input offset adjustment then the stack may become stable at a lower current.