Little Gem Mods

[jishnudg]

Hey...so I've been thinking of building a Little Gem amp that can be used for bass guitar...the plan being

1) to replace the input and output caps on the stock Little Gem (here - http://www.runoffgroove.com/littlegem-perf.png ) to allow more bass into the signal

2) Adding a small buffer before the amp (like the Jack Orman buffer here http://www.beavisaudio.com/techpages/Buffers/ - but can a BJT like 3904 be used instead of the JFET shown in the schematic - how would the circuit topology change,then,if at all? JFETs are notoriously hard to find in my area)  - the idea being that I can stick a potentiometer at the output of the buffer that can attenuate the gain of the 386 stage.

3) replacing the little gems gain control with a fixed value resistor/cap for mild overdrive,not hard clipping.

So is this a valid means of making a bass headphone amp? ie input ->  buffer -> 386 -> out ....would I have to put in some simple speaker simulation at the tail end of the circuit,and if so,then are there any speaker sims that work well for bass? Also,what would be the values for

a) new input and output caps of the 386 stage
b) pot at the end of buffer to control gain of the next stage
c) fixed resistor/cap between 1&8 of 386 chip for mild OD - to prevent it from clipping.
d) any BJT that may be used for the buffer instead of the JFET shown in the Beavis Audio page
e) simple speaker sim design,if any is needed

Thanks,
J.

[jishnudg]

Also, another line of thought could be input -> bass preamp -> 386 as a power amp -> output...in this case, my main concern would be simplicity and form factor,as I'm trying to make this pedal as small as possible,preferably with low current draw...so would this approach work? And what preamp schematics should I be looking at?

[garcho]

1) try 470u on the output first. if it's not enough, try 100N for the input cap

2)

Adding a small buffer before the amp (like the Jack Orman buffer here http://www.beavisaudio.com/techpages/Buffers/ - but can a BJT like 3904 be used instead of the JFET shown in the schematic - how would the circuit topology change,then,if at all? ...)  -

FETs and BJTs are not interchangeable. What's wrong with the BJT buffer right below Jack Orman's on that Beavis Audio page, why not use that? If you want real simple, why not skip the buffer, especially if it will add another pot/knob? You're never gonna make the 386 sound like an Ampeg.

the idea being that I can stick a potentiometer at the output of the buffer that can attenuate the gain of the 386 stage.

There is a gain knob on the Little Gem already, but if you want an output control for the buffer, just wire a standard volume control, as if the buffer was a stand-alone pedal. Read this.

3) build a simple op amp overdrive pedal that you can use for any setup instead of wasting your time with the LM386 power amp

4) don't bother with the 386

The 386 is a utility IC. It's only useful when it's at its bare-bones minimum. Trying to make it sound 'good' is a waste of time, IMO.

JFETs are notoriously hard to find in my area

where? just curious, they're a common component.

[jishnudg]

Well, I just have a few 386s lying around from an old (misinformed) purchase that I wanted to get rid of by making headphone amps for some friends of mine.I'm presently based out of Bangalore, India,and the only JFETs available here are BFW10s and BFW11s...both have rather low gain when used in a fetzer valve type configuration (gain of around 1.6 for the BFW11 - much lesser than the average gain one can get out of a JFET like J201).
Thanks for the inputs.
Regards,
J.

[therizky]

Have tried myself stickin' bjt transistor in front of 386 when I was go wild building little amps for sustainer project, thing just buzzing and oscillating all the time I just gave it up.

You might get a better result by building a Ruby using a fet that you mentioned.

Change input and output cap as suggested. I would try 1uf for input.

[jishnudg]

Thanks for the inputs..here's an unverified layout using a JFET buffer (going to try using the BFW11s after all). Going to build this in a few days, so mods,corrections and edits of any sort are welcome

Regards,
J.

LAYOUT - Ruby with Bass mods and headphone output switch (Unverified) - https://docs.google.com/open?id=0BwMVia471c4SUUtkc2ZKT0lyaGM

[garcho]

Well, you just need high input impedance, not gain, so it could work just fine.

[Microtone]

I am considering using the little gem circuit as an overdrive circuit.

Is there any reason why this might be a bad idea?

I simply want to put the circuit in a pedal, without changing any values, and use it between the guitar and an amp.

Will it maybe be too loud?

[antonis]

I am considering using the little gem circuit as an overdrive circuit.
I simply want to put the circuit in a pedal, without changing any values, and use it between the guitar and an amp.
Will it maybe be too loud?

Loudness can easily be set by Master Volume pot setting..

Take into account that LM386 minimun gain allowed is about 9 (not compensated for gain of lower values..) so it should be a good idea to place a pot between IN and input capacitor to be flexible enough for signals of high amplitude and low gain combination..

[Microtone]

I am considering using the little gem circuit as an overdrive circuit.
I simply want to put the circuit in a pedal, without changing any values, and use it between the guitar and an amp.
Will it maybe be too loud?

Loudness can easily be set by Master Volume pot setting..

Take into account that LM386 minimun gain allowed is about 9 (not compensated for gain of lower values..) so it should be a good idea to place a pot between IN and input capacitor to be flexible enough for signals of high amplitude and low gain combination..

Thank you for your reply. I fear I may have already waded in out of my depth! What exactly does minimum gain mean? And the same question to high amplitude low gain combination! Is that to do with the input signal?
I went ahead and built the circuit into a pedal and it seems to work nicely with a medium output single coil guitar as well as a higher output humbucker guitar.
But I'd love to understand the implications of the caveats you mentioned, to ensure it's versatile.
At the moment I don't have a gain control in the circuit, so it's always at full gain with just an output volume control. But if I were to include a gain control, what would I need to consider for the high amplitude low gain caveat? And how does a variable resistor on the input help that? Would that not just reduce the voltage of the input signal?

[Matthew Sanford]

The input potential meter lowers the signal so then it's not as loud because it has to amplify it a lot due to how an LM 386 is. For the gain you might wanna look up the Ruby amp schematic and follow how they handle pins one and eight, or look at the LM 386 data sheet for the info.

[Microtone]

The input potential meter lowers the signal so then it's not as loud because it has to amplify it a lot due to how an LM 386 is. For the gain you might wanna look up the Ruby amp schematic and follow how they handle pins one and eight, or look at the LM 386 data sheet for the info.

Thanks for the advice. So, I could potenti(ometer)ally put a variable resistor at the start and find a good resistance that works for most guitar-based situations and then just add a single input resistor of that value at the start of the circuit?

[antonis]

I'd love to understand the implications of the caveats you mentioned, to ensure it's versatile.

At the moment I don't have a gain control in the circuit, so it's always at full gain with just an output volume control. But if I were to include a gain control, what would I need to consider for the high amplitude low gain caveat?

P.S.
It's late night here so plz someone else to check my writing..

[antonis]

Just to make things (and schematic) more clear: 

[Mark Hammer]

A construction-project article in Nuts & Volts some years back was for a circuit to provide more audible television for someone whose hearing had declined.  I forget whether it was for earphones or for small speakers situated beside the viewer (e.g., behind their head on a couch).  I have the issue somewhere in the basement, and could look it up, but the principle described here remains valid, whatever the final implementation. 

Progressive hearing loss generally starts at the high end of the spectrum, and works its way down.  Since it is the higher-frequency components of speech and other sounds that facilitates identification, definition, and localization, the circuit enhanced higher frequencies in an LM386-based amplifier, so as to restore those qualities.  It was done via the pin 1-to-8 loop.  Normally full-bandwidth gain is increased by bridging the 1.35k resistance with something parallel to reduce that effective resistance.  HOWEVER, one can bridge it selectively.  So, if one used a small-value capacitor to do the bridging, that would provide a low-resistance path for higher frequencies, and the usual 1.35k path for bandwidth below that.  The principle is not at all different from that used for compensated guitar volume controls and bright switches on amplifiers.  The degree and area of enhancement could be achieved in a more articulated manner to achieve different slopes.  So, there might be one smallest-value cap doing the bridging, and a second higher-value cap in series with a 1k resistor and even a third larger value in series with 470R.  Etc.