IC Buffer bias question

[chuckmoose]

I am making an IC buffer (TL071) to go at the end of my bedalboard chain (about 10 true-bypass pedals, there may be a TS with a buffer on there at times) using the following schematic:



I have seen this same schematic with R1 and R2 values ranging from 220K to 2.2 meg.  I understand that the input impedance is calculated as half of their value (500K if each resistor is 1m) but what value is best suited to my application, and what are the reasonable limits for an IC?  Thanks for any input.

[brett]

Hi
in my opinion input impedance can easily be too high (my opinion of too high is 1M or more).  That introduces noise, and encourages your lead to pick up radio signals, static, etc.

On the other hand, around 220k is considered low by modern standards.  So for this circuit, I would use 2 x 1M for a half meg load.

If you've like a little vintage tone and some milder highs, you can lose them by having about 50k of impedance.  That's about what boosters and distortion circuits had back in the 1960s.  The fuzzface typically has only about 20k of input impedance!!  A major reason why it is so mellow, and drags the signal down when playing honly to let the notes sustain and "bloom".

Sorry that there's no simple answer.
cheers

[chuckmoose]

That's very helpful, thank you.

While I'm asking questions does this layout have an error? 


I don't really understand the pins on ICs in a schematic, but when I wired up the buffer according to this diagram I got a lot of noise and low level somewhat distorted signal.  I removed the ground from pin 5 (? the lowest on the right)  it cleared up and seems to work fine, though I don't know how to measure things to find out for sure if it's doing what it should.  So should there be a trace break on the bottom row as well?

Also, can I run this at 18v as it is? 

[JKowalski]

The OFFSET pins (5 and 1) are typically not used, and should not be connected to anything. You seemed to have fixed that. What do you mean by cutting a trace in the bottom row? I assumed you already did this, since you said that you disconnected pin 5 from ground...?


As for the 18V, I think so. Take a look at the TL071 datasheet, it should tell you what the power supply voltages are.

[R O Tiree]

You did cut the track under R3? Hard to see...

From the datasheet, it will cope with a max supply of +18V/-18V, so +9/-9 will be fine.

[cab42]

Hi

I did that layout some years ago, and I didn't have any problems with noise and distortion.

Rereading the data sheet, I can see the I should have left pin 5 unconnected. However, I remember doing it because I had seen other circuits where pin 5 was connected to ground. I don't remember which, though.

I will change the layout.

Regards

Carsten

[JKowalski]

Hmm, it seems I have a question too. When the data sheet says +/- 18V, does this denote a maximum differential of 36V across the rails in any case, (meaning you could go from ground to +36V in a single supply set up) or does is it have to stay under 18V from ground no matter what? If so, why?

It's kind of funny that I never noticed or thought about this before - I have never gone anywhere above 12V for op-amps.

[composition4]

yep, +/- 18v = 36v rail to rail

[R O Tiree]

Yes - a total of 36V.

If you apply +18V and 0V, then those two 1M resistors will provide a fairly stiff "virtual ground" at +9V. The opamp doesn't care one jot whether it's truly receiving +9V and -9V or +18V and 0V - it's still seeing its rails 18V apart.

[cab42]

I just edited the layout. As far as can see it is looking OK from a quick glance at the schematic at AMZ.

I removed the LED resistor in order to avoid the track cut under R2. I really don't know why I added it in the first place.



Any comments or suggestions are welcome.


Regards

Carsten

[JKowalski]

Im curious to know whether you truly understood my question - I was asking whether since the Op-Amp will work at +- 18V, which is a 36V differential, will it work with one single polarity 36V supply, with the signal biased at 18V? For some reason, I seem to remember seeing something somewhere that said this was not the case, and that it can only go up to the data sheet max (in this case 18V) in either direction off 0V - meaning the maximum single polarity supply you could use would be a +18V supply.

I don't see why it would'nt accept a 0V/+36V supply... It doesn't make any sense to me - but I really feel like I read somewhere that it wouldnt. Am I just making stuff up?

[composition4]

Sorry, I did misunderstand.  As far as my understanding goes:

a single 36v supply with the signal biased at half way is the same as +/-18v to the opamp.  Just like running the opamp at 100v/136v, or -10v/+26v

It is the potential between the two rails - as long as the input is biased to halfway between the rails, all that matters is the voltage difference between the rails

EDIT: or another way to put it... 1 or 100 or 1000vDC means nothing until you reference it to another voltage... whether that be 0v or another voltage.  The op-amp doesn't "see" for example 100vDC - as far as I can reason, it only sees the total voltage across it

Jonathan

[brett]

Hi
indeed, voltage refers to the concentration of electrons.  If two places have equal concentrations of electrons (ie equal voltage), there will be no flow from one place to the other, and hence no current, and no energy produced.  For a given resistance, etc, the only thing that affects current, energy release, etc is the DIFFERENCE in concentrations of electrons (ie voltage difference) between the two places.

My mental analogy is that voltage is like the height at which water is stored.  Resistance (1/conductance) is the size of the pipe joining two water storages.
cheers

[JKowalski]

all that matters is the voltage difference between the rails

Okay, yes, that's what I thought. I suppose I didn't actually read anything like that, my mind's playing tricks on me I guess 

Thanks.

[chuckmoose]

The OFFSET pins (5 and 1) are typically not used, and should not be connected to anything. You seemed to have fixed that. What do you mean by cutting a trace in the bottom row? I assumed you already did this, since you said that you disconnected pin 5 from ground...?


As for the 18V, I think so. Take a look at the TL071 datasheet, it should tell you what the power supply voltages are.

I build layouts on my breadboard first, mocking up the vero layout so I can confirm it against the schematic before I go and solder things.  So I removed the ground on the breadboard.

I still have a lot to learn, and I did read the data sheet before I ask but I was confused as the sheet didn't spoon feed me a line that said "yes you can run it at 18v".  After reading this discussion about 36v I now understand what the data sheet is telling me, so this has been a big help.

Thanks folks.