unused single opamp pins

[tempus]

Hey all:

I'm using a TL071 for a buffer, and (having never used a single opamp before) am unsure what to do with the 3 remaining pins (balance, comp, etc.) Can they just be left unterminated?

Thanks

[Rob Strand]

what to do with the 3 remaining pins (balance, comp, etc.) Can they just be left unterminated?

Yes.

In general for unused pins
Offset Null (Balance)/Comp:        Open
+Input/-Input:                             Gnd
Output:                                      Open

Good idea to read the datasheet for weird stuff and any special notes (eg. about bypass caps).

[amptramp]

The TL071 input stage has a pair of P-channel JFET's as the input stage with a PNP source feeding their sources from the positive rail.  There is an NPN current mirror stage connected through nominal 1080 ohm emitter resistors to the negative rail.  The offset null pins are connected to the emitter - 1080 ohm resistor junctions and it is intended that a 100K linear pot be used as a balance control where one is needed.  The slider of the pot is connected to the negative rail and the ends of the pot are connected to the null inputs.  It seems like there should be some protections against going too close to the ends of the pot and shorting out the 1080 ohm resistors but this is the factory recommendation.  A resistor in series with the slider of the pot would be adequate.

If you are not using the offset null feature, leave these pins open.

There is no explicit compensation pin on a TL071 but it is conceivable that the offset pins could be used for that purpose and they are in some other op amps.

[ElectricDruid]

Worrying about nulling out a little bit of offset seems a bit odd on an audio op-amp, where mostly you're going to have signals AC-coupled with a cap. I suppose that's not the only use for it, and they don't make 5 pin IC packages anyway, so they might as well do something with them!

[Rob Strand]

Worrying about nulling out a little bit of offset seems a bit odd on an audio op-amp, where mostly you're going to have signals AC-coupled with a cap. I suppose that's not the only use for it, and they don't make 5 pin IC packages anyway, so they might as well do something with them!

If you build test equipment offset null is pretty important for old school analog designs.  For the older bipolar opamps nulling the DC offset usually did a good job of nulling the temperature coefficient - something you don't get when you inject a DC offset into one of the input pins.

I find it interesting how the audio domain (AC coupled) and the precision domain (handling small DC voltages and concern with temperature effects) worry about different aspects of opamps.

[tempus]

OK got it up and running - thanks guys

[Aph]

...and they don't make 5 pin IC packages anyway...

Oh, I can't let that go! 

[ElectricDruid]

Worrying about nulling out a little bit of offset seems a bit odd on an audio op-amp, where mostly you're going to have signals AC-coupled with a cap. I suppose that's not the only use for it, and they don't make 5 pin IC packages anyway, so they might as well do something with them!

If you build test equipment offset null is pretty important for old school analog designs.  For the older bipolar opamps nulling the DC offset usually did a good job of nulling the temperature coefficient - something you don't get when you inject a DC offset into one of the input pins.

If I was building test equipment, I doubt that'd be my first choice op-amp...but point taken.

I find it interesting how the audio domain (AC coupled) and the precision domain (handling small DC voltages and concern with temperature effects) worry about different aspects of opamps.

I agree. That's something that comes up in analog synth design a lot, since you've got the audio path, but you've also got a sensitive CV path. 83.33mV is a semitone at 1V/Oct, so a few mV out gets to be noticeably. And temperature effects too - the bane of VCO stability. Different op-amps for different jobs.

[Rob Strand]

If I was building test equipment, I doubt that'd be my first choice op-amp...but point taken.

Not mine either really, offset nulling is a pain for production.  Modern designs just use better opamps.   Things have changed as well.  For many modern opamps the tempco doesn't zero when you null using the offset pins.   The only thing the offset pins offer is a way to null without worrying about injecting noise.   If you trade that against more pins and more PCB area the null-less duals are a better deal.   Even with modern opamps the worst-case offsets can still be too high for some applications.

That's something that comes up in analog synth design a lot, since you've got the audio path, but you've also got a sensitive CV path. 83.33mV is a semitone at 1V/Oct, so a few mV out gets to be noticeably. And temperature effects too - the bane of VCO stability. Different op-amps for different jobs

You will see the problems there for sure.   Analog synths have a lot of cool analog stuff.