help with buffers!

[njkmonty]

I was breadboarding a simple jfet buffer circuit from http://www.muzique.com
through some of my older vintage pedals, (seeing how much of a difference with highs coming through while on.



then i tried a different buffer (mi audio boost n buff V1 clone in bypass (on/off switch parallel with r2 open))
and notice that it was more pleasing and noticeable to my ears.
since im not needing the boosting / treble part of this schematic, what can i get rid of to simplify it?
components coming off emitter?

[Elijah-Baley]

If I understand, you want the second schematic, the MI Audio Buff'nBoost, but with less gain. You could use a trimmer, and adjust it how you wish.
Or, If you want the minimum gain setting I guess it's enough a use a 10k resistor between the emitter of the transistor, and a link between the 1uF and the ground.

Then if you want the switch opon that means you need the 220k resistor, so you have just omit the switch.

[Gus]

2nd schematic is not a buffer it is a boost and it interacts more with the pickup, note the collector to base resistor(R4) this is part of the bias and also feedback and IMO not the best way to build a circuit like this, one more resistor from base to ground makes it more repeatable.
and it roll off highs note C5

So maybe you don't want a buffer like the first schematic. Maybe you want a boost with EQ

Buffer often means high input resistance, low output resistance with a gain of just under 1 or closer to 1 with an opamp buffer.

[GGBB]

You should be able to get rid of C1 and R2, and change R5 to a fixed 10k resistor. That would effectively make it permanently in buffered bypass mode. None of the other features - F/T/M or even volume - are shown in that schematic so it is unclear just how those might be removed. If volume is the only issue and it's too loud, change R3 22k to a 25k pot and take the output from the wiper.

[njkmonty]

i got rid of c1,r2 and spst switch and replaced 10k pot with a 10k resistor.
still sounded the same to me which is what i was hoping for .
my only other question is, i tried pulling out the 220pf c5 cap
to try and reduce the number of components.
i didnt appear to notice any change.  do you think this cap is essential? or for my application fine in removing?

[Gus]

What do you think you want?
Maybe you don't want the 2nd circuit.
If you post what you want that will make it easier to suggest a circuit.

Do you want a lot more gain?
Do you want just a little more than a gain of 1?
Do you want EQ? Maybe a little rolled off highs less or more bass?
Have you tried a RangeMaster(bass and mid cut and boost) like circuit or a LPB or NPN boost Ver 2 or any of the other circuits you can find.
Are you using a solid state combo or tube combo or something else?

[njkmonty]

What do you think you want?

as i said earlier in my post, I m happy with the results of Boost n buff version 1 in bypass modes (spst switch left open)
what i am wanting to do is simplify the boost n buff as it is by removing the components not needed ie treble boosting etc.
i have removed c1 , r2 the switch and replaced the pot with a 10k resistor.
I am happy with the result, however what i was hoping to get some advice on was the c5 220p cap. when i removed it from breadboard i couldn't detect any audible difference
and i was hoping to get some assistance in what that components purpose may be  , to understand if it needs to stay or whether it isn't a critical component?

[GibsonGM]

C5 can go.   Other than that, you are probably at as simple as you can make it.

[njkmonty]

thanks

[dschwartz]

8.2 meg... sounds like it may introduce a lot of thermal noise.
Maybe C5 is there to reduce the hissing..but i have always had this question:
C5 acts as a low pass filter by shunting high freqs to V+..but what would happen if the voltage supply is a switching style? It looks like the high frequency whine will be directly coupled to the output..is that the case?

[Peter Snowberg]

If you listen closely you might hear a tiny change in the highs without C5 but in any case you've reached maximum simple for this circuit or even slightly more/less. Einstein said make everything as simple as possible but not simpler and you're up against a little too much simplicity here.

C5 is probably there mainly for rejection of RF energy that might cause high frequency oscillation via feedback so it might not be needed in most situations but then there are the times when it is and that will be mainly up to construction and the transistor used. A less simplified version of this stage breaks R4 into two resistors in series and puts a larger version of C5 (like 100n) at the junction between them. That way it allows the base to see the collector voltage but all the AC can be removed from the bias voltage without really changing the signal at either the base or the collector. Keep in mind that high frequency oscillations won't be audible themselves, but they'll manifest as intermodulation distortion and poor tone.

If I were building this, for R4 I would use a pair of 1.5 to 4.3Meg 1 or 2 watt metal film resistors with a 0.1uF to ground at the junction. The physically larger high value resistors seem to have lower thermal noise which is a good thing here given the super high impedance. Lowering the overall value of R4 a hair will also reduce thermal noise and that value is a trade off between noise and highs.

I'm glad you found a good tone so quickly. Some people obsess over boosts without ever finding the tone they want. It's stunning how much variety you can find in such a "simple" building block.

[Peter Snowberg]

C5 acts as a low pass filter by shunting high freqs to V+..but what would happen if the voltage supply is a switching style? It looks like the high frequency whine will be directly coupled to the output..is that the case?

Ideally the power rail and ground are well bypassed with capacitors so that when it comes to high frequencies there is effectively a short between power and ground and all that high frequency noise is filtered out. Filtering out switcher noise isn't really that hard but it gets much easier as the frequency gets higher. I like >1MHz switching supplies for pretty much everything but for delicate audio I always use linear regulation regardless, even following a switching supply.

[PRR]

> 8.2 meg... sounds like it may introduce a lot of thermal noise.

Which is shunted by a much lower source (guitar) impedance. More R here would actually be less hiss (because more shunting loss). However the value has to be in a zone for happy transistor bias. Some math in the dirt suggests 3Meg-10Meg. 8Meg is in the zone. 10Meg would not be audibly less hiss. 33Meg would starve the transistor.