ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

4.5V-multi Op amp question

8/25/1999 10:19 PM
Ed Rembold	4.5V-multi Op amp question O.K. whats the rule of thumb here? - NON bipolar power supply, multiple dual Op amps. do we place 4.5v "bias" (through a high value resistor) to each input, of each Op amp? or only at the input of the effect? I can see getting away with "single input bias" if the Op amps were not R/C coupled. Another example- one Quad Op amp, does it need 4.5v at each input , if the signal just "loops" in and out? I'm not an EE, and don't know how to word this just right, But maybe one of Vetrans will help. Thankyou, Ed Rembold

8/25/1999 10:19 PM

Ed Rembold

4.5V-multi Op amp question

O.K.

whats the rule of thumb here? - NON bipolar power supply,

multiple dual Op amps. do we place 4.5v "bias" (through a high value resistor) to each input, of each Op amp? or only

at the input of the effect?

I can see getting away with "single input bias" if the Op amps were not R/C coupled.

Another example- one Quad Op amp, does it need 4.5v at

each input , if the signal just "loops" in and out?

I'm not an EE, and don't know how to word this just right,

But maybe one of Vetrans will help.

Thankyou, Ed Rembold

8/27/1999 12:57 AM
Mark Hammer	I'm not an EE either, but the simple version of the correct reply runs something like this....(assming I've got it right) The "4.5V bias" you speak of is often indicated on schematics as VRef, meaning that it is a reference voltage; a fake "middle" that any signal changes are in "reference" to. It is not 0 volts, but IS the midpoint of the power supply voltage that is available, the same way that ground is the midpoint of a bipolar power supply. Any time an op-amp stage is about to do something relative to a reference point, you generally need to take out the DC from the previous stage to, in effect, "reset" the bias. The reference voltage is then introduced to the next stage all over again. Op-amps vary in terms of offset errors introduced. I'm sure there is a precise technical term for it, but when looking through more detailed catalogues or manufacturers applications books, what you'll see is that the really pricey op-amps are often those that introduce very little offset error. If the offset is a half a millivolt, and the op-amp stage has a gain of 100, the next stage is not seeing something that hovers above and below exactly 4.5V. Unless you block DC with a cap, this error is likely to be compounded by the next stage, hence the RC coupling and reintroduction of a "true" VRef for each relevant stage. Alternatively, you can stick in a trimpot for each op-amp and compensate for offset error. Most folks prefer to just use a DC-blocking cap and rebias because it requires no adjustments, and is simply much more expedient if you lack a scope or good meter. An anecdote: About 14 years ago, I was helping a prof buddy build a precision electrophysiological recording pre-amp for some work he was doing. He was using DC coupling and a bipolar supply, but he was also using cheesy 741's (I take it from his behaviour that Horowitz & Hill is a terrific book to learn about design, but lacks much info on op-amp selection). Needless to say "precision" was not the word that popped into his head when he used it. I suggested popping in some CA3140 MosFet op-amps (because I was fond of using them at the time, rather than because I knew about the more exotic properties, such as offset error). The difference was like night and day. He was so thrilled by the change in performance that he ran out and bought me the very soldering iron I use to this day, just to say thank you. DC couple at your own risk, my friend.

8/27/1999 12:57 AM

Mark Hammer

I'm not an EE either, but the simple version of the correct reply runs something like this....(assming I've got it right)

The "4.5V bias" you speak of is often indicated on schematics as VRef, meaning that it is a reference voltage; a fake "middle" that any signal changes are in "reference" to. It is not 0 volts, but IS the midpoint of the power supply voltage that is available, the same way that ground is the midpoint of a bipolar power supply.

Any time an op-amp stage is about to do something relative to a reference point, you generally need to take out the DC from the previous stage to, in effect, "reset" the bias. The reference voltage is then introduced to the next stage all over again.

Op-amps vary in terms of offset errors introduced. I'm sure there is a precise technical term for it, but when looking through more detailed catalogues or manufacturers applications books, what you'll see is that the really pricey op-amps are often those that introduce very little offset error. If the offset is a half a millivolt, and the op-amp stage has a gain of 100, the next stage is not seeing something that hovers above and below exactly 4.5V. Unless you block DC with a cap, this error is likely to be compounded by the next stage, hence the RC coupling and reintroduction of a "true" VRef for each relevant stage. Alternatively, you can stick in a trimpot for each op-amp and compensate for offset error. Most folks prefer to just use a DC-blocking cap and rebias because it requires no adjustments, and is simply much more expedient if you lack a scope or good meter.

An anecdote: About 14 years ago, I was helping a prof buddy build a precision electrophysiological recording pre-amp for some work he was doing. He was using DC coupling and a bipolar supply, but he was also using cheesy 741's (I take it from his behaviour that Horowitz & Hill is a terrific book to learn about design, but lacks much info on op-amp selection). Needless to say "precision" was not the word that popped into his head when he used it. I suggested popping in some CA3140 MosFet op-amps (because I was fond of using them at the time, rather than because I knew about the more exotic properties, such as offset error). The difference was like night and day. He was so thrilled by the change in performance that he ran out and bought me the very soldering iron I use to this day, just to say thank you.

DC couple at your own risk, my friend.

8/27/1999 2:38 PM
paul perry	Mark Hammer said "He was using DC coupling and a bipolar supply, but he was also using cheesy 741's (I take it from his behaviour that Horowitz & Hill is a terrific book to learn about design, but lacks much info on op-amp selection)." This, Sir, is a base libel. There are YARDS of good stuff on op amp selection. I guess like many academics, your prof had the arrogance that comes from knowing everything about almost nothing.. Certainly, he hadn't READ H&H.

8/27/1999 2:38 PM

paul perry

Mark Hammer said "He was using DC coupling and a bipolar supply, but he was also using cheesy 741's (I take it from his behaviour that Horowitz & Hill is a terrific book to learn about design, but lacks much info on op-amp selection)."

This, Sir, is a base libel. There are *YARDS* of good stuff on op amp selection. I guess like many academics, your prof had the arrogance that comes from knowing everything about almost nothing.. Certainly, he hadn't *READ* H&H.

8/27/1999 4:07 PM
Mark Hammer	Well, there you go. Yet one more reason why one of the few remaining blank spaces on my bookcase cries out for this tome. I stand corrected.

Page 1 of 1