Microamp pot range

[il_mix]

Hi everyone!

I build (on breadboad) a Microamp clone, using ToneAmp layout. I used a TL071 instead of the 4558, so I changed all the connection for the single channel OpAmp.
I've read this really useful thread
http://www.diystompboxes.com/smfforum/index.php?topic=95120.0
Here are many interesting info, and a couple of example of grond-leg resistance values VS output gain. I've tried to insert a couple of fixed resistors (the ones suggested in the above thread) instead of the potentiometer, and I get the expected gain, so the circuit seems ok.

The problem is that, with the 500K reverse log (C500K) potentiometer, the gain is really hard to adjust. The output volume doesn't change from 7:00 to 3:00, than all the gain range is compressed between 3:00 and 5:00.
I've measured the impedance values between the center pin and the side pin of the potentiometer on board, and these are the results: 
7:00 -> 460k Ohm
12:00 -> 400k Ohm
3:00 -> 110k Ohm
5:00 -> 0 Ohm
So, the circuit behavior is indeed correct (no audible gain since impedance is below 100k Ohm). The problem here is hat the potentiometer changes too "slowly".
I've tried a 500K linear, but the results are similar.

So, do I have a faulty potentiometer (actually, 2 folty potentiometers, since I've testes 2 C500K), or am I missing something?
Is there a better pot to use?


Thanks!
MIX

[Mark Hammer]

Seems to me you may have the pot wired up wrong.  Normally, one uses a C-taper pot to "get through" a big change in resistance very quickly.So, it should be the case that your pot takes you through over 2/3 of the total resistance by the time you hit 10:00.

I'm looking at a regular A-taper (log) 500k pot right now, and at exactly 12:00, I have 73k on one side of the wiper, and 430k on the other.  In principle, it would be exactly the same (with sides swapped) for an anti-log C-taper pot.  Your measurements are roughly in the ballpark, with 400k of the the 500k "eaten up" by the time you hit 12:00.

HOWEVER, what you want is for the pot resistance to get smaller as youo turn clockwise, since in a non-inverting op-amp, gain goes up as the ground-leg resistance goes down.  Looking at the pot from the shaft side, with the lugs pointing down, you want to use the middle and rightmost lugs, such that the wiper moves towards the other lug as you go clockwise.

With 100k of resistance at the midpoint, that will produce a gain of (56k+100k+2k7)/(100k+2k7) = 1.55x, which is a boost, but not all that much.  If you were looking for something more noticeable, by the time you hit, say, 10:00, then what you probably want to do is to raise both the value of the 56k feedback resistor, and the 2k7 ground-leg resistor.

Let's do a thought experiment.  The max gain of a stock Microamp is just under 22x.  Lets make the feedback resistor 82k, and the ground leg resistor 3k9.  When the gain pot is at 0 ohms, that also gives us a max gain of 22x.  The difference is, however, that when your pot resistance hits roughly 100k (around 9:00, by your comments, and you are using the middle and rightmost lugs), your gain is already at around 1.8x.

So try those component changes, and see if you like the diability of the gain a little better.

[il_mix]

Hi Mark,

many thanks for the info.
Sooo... I've cheated a little in the previous post....
I've connected the pot with random orientation, thinking that, if connected inverted, I just have to turn it counterclockwise (what I was really doing, so the hour->impedance measures are indeed inverted) to have the same behavior of a correctly connected pot.
Connecting the pot inverted, the range limits are the same, but since the pot change impedance in a Log way the "fast change" ranges will be inverted! Connecting it the wrong way (like I did) means that the "fast range" will be in the wrong place, i.e. the impedance will decrese faster when it is already small (100K to 0) instead of decreasing fast when it is high (500K to 100K).
Will invert the port orientation and check if everything is ok.

Then I will probably change some resistor values to have a smaller maximum gain. Like using a 3k3 or 4k7 instead of the 2k7 series resistor, so I will have a max gain of 18dB (with 3k3 resistor) or 13dB (with 4k7 resistor).
Is this a good approach to change just the 2k7 resistor, without touching the 56k feedback one?

Thanks
MIX

[Mark Hammer]

I suppose it depends on what you want to use the Microamp for.  If the purpose is simply to have a little more level on tap for solos, without changing the tone of anything (or very much), then confining yourself to 12-15db gain is okay.  If your goals include pushing anything after the Microamp into overdrive, or conceivably using the pedal to bring a voice mic up to level, then that will not likely be enough.

The fundamental problem you posed in this thread, however, was an issue with the pot taper - and corresponding changes in gain - offering poor "dialability", with the range of most interest being squished into a few meagre degrees of rotation.

Maybe it is worth taking a step back from my earlier suggestions and giving a read to RG's timeless classic "The Secret Life of Pots", where some insight is provided into making pots behave the way you want.  It may be that you only need to stick a single tapering resistor on the pot itself, and don't have to change anything else.  http://www.geofex.com/article_folders/potsecrets/potscret.htm

[induction]

My guess is that un-reversing the pot rotation will fix the dialability problem right up.

[Mark Hammer]

That's my guess too, but pot tolerances have no obligation to comply with our expectations. 

[il_mix]

Didn't have the time to try inverting the pot connection yet. Will do it soon.

About changing the 2k7 resistor to limit the microamp gain, my aim is to have a small booster for solos, not to overdrive other pedals. So yes, I think a 15dB limit or so will be enough. Will made some test.

I will read the document you liked, that looks quite interesting (I love to destroy things to understand how they works ). About tricking the pot to have a custom behavior, well, it's quite too much... The rev log pot connected right will do what I need, I suppose.
You said "It may be that you only need to stick a single tapering resistor on the pot itself, and don't have to change anything else."
Did you mean that adding a taper resistor I can use the pot connected the wrong way? (original pot range problem)
Or that adding a taper resistor to the correct connected pot I can limit the gain range without working on the 2k7 resistor? (gain range question)

[il_mix]

Ok, connecting the pot the right way have resolved the main issue. It seems that there's still a fast gap in the final range, but I've tested it with an audio card, where the input range is probably limited.
One thing that I've noticed is that, then the gain is max (pot turned fully clockwise), with the effect inserted there is saturation (and this is ok), without inserting the effect, so with clean sound, there is a background hiss.
What can it be?
I have a footswitch connected on the breadboard to switch between clean and effected sound. There is some long wires to connect the breadboard strips, and of course no metal enclosure. Maybe some amplified electromagnetic noise? It's quite loud...

[il_mix]

I tried using a battery to power the circuit, instead of an AC adapter, but the problem is still there.
I noticed that the "hiss" disappears when touching some circuit metal part. I thought about a failing ground. But it disappears when thouching a coated wire, too. Dunno...
BTW, I will re-check for the n-th time the circuit.

[il_mix]

Got it.
I'm using a 2PDT switch on the breadboard, to route the in/out, without any LED control. This way, when the effect is bypassed, the input pin is floating. The IC is amplifying a random signal, probably generating some electromagnetic noise that affects the clean signal. That's what I suppose is happening.
Grounding the input pin when the effect is bypassed (can be done with a 3PDT; on breadboard I do it manually), the hiss disappears.

Time to move the project on perfboard.

[Hatredman]

It smells oscillation.