What does single supply really mean, when referring to op-amps?

The difference from being able to power it from a car battery or a cheap 9v transistor battery and one that needs a complex power supply.

Jim

-- "If you think you can, or think you can't, you're right." --Henry Ford

Dual-supply opamps have three supply pins, so two separate supply voltages must be maintained. Single-supply opamps do away with the middle (voltage wise) pin.

For a single-supply opamp, if you have a +-2.5V supply then you could connect the +2.5V supply to one of the supply pins and hook to the -2.5V supply to the other supply pin.

With a dual-supply opamp, you can't just use a single 5V supply because you need that "middle" supply pin's voltage to remain in the middle of the 5V supply. You could always create a stable mid voltage for that middle pin, but then you've effectively got your two supplies.

Bob

From the opamps view point, there is no difference. Single supply opamps are so called, because they can function with lower minimum supply total voltage than traditional designs (that have something like nominal +-15 volt supply rails) and respond to input voltage and produce output voltage very close to the negative supply rail voltage, instead of wasting a few volts near that rail as the traditional designs do. These features make them more convenient to use with single supplies, especially when signals are referenced to the negative rail.

(snip)

I have never seen an integrated opamp with 3 supply pins. I have seen a few potted opamp modules that did have them. Do you have a data sheet link that shows an integrated dual supply opamp with 3 supply pins?

the biasing and output control is much easier to handle with a +/- rail supply. With 2 rails, the output of the op-amp can null each other to get a nice 0 volt with the exception of the offset factor that is spec's. the + input can be tied directly to the common. With single rail, the output will rest at 50% for example of the rail supply and you need to have a voltage divider at the + input to obtain this. Also, single rail operation causes a problem if one of the inputs fall to much below the common side which will cause a problem depending on which op-amp is used.

Also, with Dual rails, you can get +/- DC output from the op-amp where is, single rail, you get what ever the specs are on the op-amp 0..rail for an ideal op-amp. I suppose one could bridge a pair of single rail op-amp outputs to obtain the +/- DC output how ever, then you wouldn't have a common for this signal. this practice is used a lot in balanced line transmitters with a dual op-amp in place for example.

So, it's all in the design at hand as to what the application is.

The ones I remember with three supply pins may have been hybrid modules. Any I've used in the last 15 years have only two supply pins.

It's interesting, after doing some searching, that even some with two supply pins are referred to as "dual supply" and list their power supply requirements as '+ - xV'. This may be the source of the original poster's confusion -- and it IS confusing. In both cases, for these so-called dual-supply and single-supply opamps, the inputs must be appropriately biased to meet the input common-mode requirements.

Can you shed any light on this, John?

Bob

Hey, great thread. It has invoked 4-out-of-4 nonsense replies so far.

All opamps are pretty much alike, and all can be powered by a negative supply rail that is ground, or not ground. Industry convention seems to be that a "single supply" opamp has an input common-mode range that includes the negative rail, which means that if you power it from a single positive supply (V- = ground) it can amplify input signals that are close to ground.

It's a little more useful for a manufacturer to specify an opamp as being "rail-to-rail input" and/or "rail-to-rail output" instead of the vague "single-supply."

So the only diff between powering an opamp from +5/0 or from +-2.5 is the signal range it could handle in either situation. The opamp itself doesn't know the difference.

John

Yeh 5 out of 5 now

John Popelish snipped-for-privacy@rica.net posted to sci.electronics.design:

At least you brushed on it. It is the difference in what the reference is, and how it is handled. Rail referenced opamps almost always require DC blocking capacitors between stages, or extraordinarily careful design. With split supplies there is little difference in handling either / both AC and DC components of a signal.

I have not seen any integrated circuit opamp since the uA709 that had 3 supply pins.

That last point is what is important for any supply-signal situation. The input and output signals must all remain inside the functional operating range with a few exceptions. A few designs allow the inputs to exceed one of the supply rails by a considerable amount without any problem, except that the input is responded to as if it were at the rail.

Dual supply opamps usually just require that the inputs stay a few volts from either rail, and their output swings to only within a few volts of either rail, so it is usually convenient to power them with a pair of supply voltages that straddle the signal voltages (which are often conveniently centered on zero volts). Note that this does not require that the two supply rails have the same magnitude. I have done quite a few designs with unsymmetrical rail voltages.

They mean two different things to me. "Rail-to-rail" includes the positive supply rail (as well as the negative supply rail) in the input common mode range and the output can swing fairly close to either rail with a light load. Single supply op-amps usually can only go to the negative supply rail input common mode range the output swing only closely approaches the negative rail with a light load). There are also op-amps that can accept input that exceed the positive supply rail by a substantial amount. Of course the bias current has to come from somewhere (eg. the inputs). An LM324 or half an LT1013 is a single-supply op-amp. A uA741 is a dual supply op-amp. An OP184 is a rail-to-rail input/output op-amp. An LT1782 is an example of a R-R I/O op-amp that can also handle inputs well beyond the supply rail ("over the top").

Yup.

Best regards, Spehro Pefhany

From my perspective, the only difference is input biasing and maybe output coupling (i.e. common mode).

Cheers

PeteS

Wrong. John Larkin got it almost right - Spehro Pefhany did tidy up a few loose ends in John's post, but John's post certainly wasn't nonsense.

-- Bill Slloman, Nijmegen

Well, the last resort is to actually read the datasheet.

John

Apologies to JP; his stuff always makes sense.

John

What an EFFING kissass! You're right, of course, but still a kissass.

Bob

Got a link to an opamp with three supply pins? I mean, something less than 30 years old? Maybe an Apex hybrid, possibly.

John

I think you'll find the difference is in the output stage. In the early op amps dual supply amps were optimised to drive loads referenced to 0v whereas single supply amps were for driving loads referenced to the negative rail. Compare the datasheets for lm348 verses lm324. The input stage of single supply amps are optimised for signals down to the negative rail. For most modern op amp designs you can pretty much use them either way.

Look, John, I can see you're really upset about this. I honestly think you ought to sit down calmly, take a stress pill, and think things over.

Bob