Old Radio Shack mini amp?

[Paul Marossy]

the 5k pot will connect across (load) the guitar, if that is what you inplug. the guitar will see 5k when it would really like 500k for unrestricting highs. as it is connected as a volume control with respect to the guit, you should be able to replace it with pretty much any value you like. start with 100k, like many a fuzz pedal pregain pot

According to something I read by Craig Anderton, anything with an input impedance of <200K will adversely affect the highs. I think I read it in the impedance tester section in his DIY Projects For Guitarists book. 100K works OK for my taste, but I don't like overly bright guitar anyway.

Check out page 72 here: https://doclegend.com/download/93411742-diy-projects-for-guitarists-craig-anderton_59ddd77cd64ab2e01dcbe461_pdf

[GibsonGM]

Yeah, tho I think that 'adversely' depends what you expect to get out of the pedal, just in my way of thinking.  The 5k pot is clearly terrible.  Transistor effects tend to be on the low Z side...LPB is like 42k input impedance!  But still good-sounding.  There can be a lot of junk up that high, to my ears, and losing some of it is not always bad, esp. if you have overdriven first. 

Since we're dealing with an LM386 here anyway...probably not very critical.   Its input impedance is only 50k, yet it does ok for what it is...usually driving a tiny speaker that is 'tinny' anyway.  You might want to tweak some things if you re-amped and were running a 386-based circuit thru a 'real rig', I suspect...

http://www.muzique.com/lab/imp.htm

[Paul Marossy]

Yeah, tho I think that 'adversely' depends what you expect to get out of the pedal, just in my way of thinking.  The 5k pot is clearly terrible.  Transistor effects tend to be on the low Z side...LPB is like 42k input impedance!  But still good-sounding.  There can be a lot of junk up that high, to my ears, and losing some of it is not always bad, esp. if you have overdriven first. 

Since we're dealing with an LM386 here anyway...probably not very critical.   Its input impedance is only 50k, yet it does ok for what it is...usually driving a tiny speaker that is 'tinny' anyway. 

Depends. I think of something like the Little Gem MKII which has an FET buffer, it provides a high enough impedance for guitar and a relatively clean sounding output (I guess partly because it's bridged LM386s). Anyway, it all depends on what you are going for. The OP mentioned the mini-amp being used on recordings, and having one myself, I could see that becoming quite a fuzz machine at high volume.

[anotherjim]

I say Depends too.
I've got away with running passive bass straight into 10k line inputs many times. Sounds more like Jamerson than Claypool, but for bass-bass it's ok.
For fuzz tone from clipping an amp, if the guitar was like a Tele  single coils with heavyish strings, I can see it working for some once-in-a-while purpose. 5k load would cut most of the harmonics out so the fuzz tone should be clearer, but probably somewhat splatty.

But as test amp for audio probe, the 5k load would be very unfair on many places in a pedal circuit, except right on the output of op-amps.

[Paul Marossy]

But as test amp for audio probe, the 5k load would be very unfair on many places in a pedal circuit, except right on the output of op-amps.

That's why I made a box that goes between the mini amp and D.U.T. to which I connect an o-scope probe. Doesn't load the circuit down that way.

[ljudsystem]

Gibson, how did you calculate the 50k impedance?

Yeah, it sounds like very low impedance. But then again tele bridge pickup + fuzz + 2 inch speaker could benefit from a little treble rolled of perhaps... 

Started breadboarding it yesterday then I found out I was fresh out of lm386 ICs... sigh

[diffeq]

Gibson, how did you calculate the 50k impedance?

Yeah, it sounds like very low impedance. But then again tele bridge pickup + fuzz + 2 inch speaker could benefit from a little treble rolled of perhaps... 

Started breadboarding it yesterday then I found out I was fresh out of lm386 ICs... sigh

Right from the datasheet. Look at these 50k pull-down resistors on the input pins:

[GibsonGM]

Yep.  LM386 has a 50k input impedance...it even says it, literally, in the data sheet.  Most (all?) devices will give this, tho it often depends on what is 'connected' to the input.   Now, any other resistive element loading the input will be in parallel with that 50k the device comes with.

So, if you had a 100k input level pot (rather than 5k) and had it turned fully up, that would leave 100k to ground in parallel with the 50k.
100k x 50k / 100k + 50k = 33k     With different levels of input impedance, "Z", depending on how you rotate the pot.   

This is a lot better than what you get with a 5k pot!  I don't think you'll be much bothered by that, but I could be wrong. If so, use 250k pot...    Higher input impedances preserve more of the signal's high end.    High frequencies bleed off to ground faster than low thru a (medium or) low impedance, resulting in a 'dulling' in tone like when you turn the guitar volume knob down.   

Sometimes you will see this written as " 100k || 50k" (100k in parallel with 50k).

[anotherjim]

On that, a bit of a puzzle (maybe) is the bright input of many Ampegs...

...clearly the normal input gets a stonkingly high impedance 5.6M + 47k...
...but the bright input gets 147k and 100k of that is bypassed by 470pF... and the treble passing  via the cap gets a lower 47k load instead. Huh!
So how come lower impedance is used to make a "bright" input? I've a feeling that it benefits treble in this case by raising the turnover frequency of the tube Millar capacitance, but I'm guessing that quite a bit.

Anyway, since the 386 can have bags of gain, it may to be possible (for minimum parts and no pre-amp) to fit series resistance in the input with a treble bleed capacitance fixed across that resistor, or if using a pot on the input, add a treble bleed cap just like many tube amp volume controls (including the Ampeg) do. Guitars fitted with treble bleed volume controls might be a completely different match direct in to a 386 than those without the cap.

[PRR]

> how come lower impedance is used to make a "bright" input?[/u]

The 470pFd in 100K+47K tends to a 10dB treble boost. Full analysis is not that simple. Dumb analysis suggests the boost isn't all there until way-high frequency. But I accept it is "brighter".