I finished a design with two variations, can I get some opinions?

[acehobojoe]

Here is the First design, utilizing a 3pdt to switch from a buffer to a boost, I have however found this design unsatisfactory and found a simpler way to do it. It produces pops because of the DC leakage, and I found it was easier and more compact in the second design below, though it has a larger part count.


This second design uses 2 transistors on a single supply where the user can switch between designs. It has plenty of caps and resistors to help with DC shift. I am kind of unsure about the last Pot though, as it is in parallel with a 51p cap. I thought I would try to get a sort of midrange boost or low cut with more volume. However it could backfire. I'm thinking I may have to get rid of the connection and just leave the cap. The idea is that the freqs allowed to pass in the capacitor will pass back into the signal as the lower ones go to ground with less resistance. Thinking about it now though, I may need to alter it even more.


So, let me ask, what are your thoughts? Is there any way I can improve upon this design?

[acehobojoe]

go ahead and right click "open image in new tab" because the pics are rather large...

[antonis]

LED should be connectend in reverse polarity...

JFETs are NOT biased..

C4 should be placed after the R9..
(if you use R9 for avoiding pops..)

A cap like C4 should be placed after R2 also..
(I don't understand why you put coupling capacitor only on buffer..??)

You can have variable  gain in boost mode by replacing the R8 with a 10k pot..

[acehobojoe]

Ok, thanks... I've always done the LED the other way, but something made me change it, I guess because I envisioned the current going the other way. anyway.. stay tuned.

[acehobojoe]

ok, fixed those coupling mistakes and added the variable gain.

Were you saying that the Jfet needs not be biased, or that I should bias it? If I need to bias it, I could simply add a 1m between the 9v and the other 1m.

The only other question would be, the tone cap at then end, perhaps I could combine it with the variable gain pot.



EDIT: lug 2 needs to be connected. sorry.

[samhay]

I don't understand the design remit - what do you want the buffer to do that you can't do with a variable gain amplifier?

[antonis]

ok, fixed those coupling mistakes and added the variable gain.
EDIT: lug 2 needs to be connected. sorry.

If I need to bias it, I could simply add a 1m between the 9v and the other 1m.

No, because you use the 1M resistors for avoiding popping...

You have to add a couple of 1M resistors (one between 9V and Gate and another between Gate and GND) for biasing and perhaps raise the value of R2 and R9 up to 10M...

The only other question would be, the tone cap at then end, perhaps I could combine it with the variable gain pot.

Theoreticaly, in a clear boost shouldn't matter where you put the Volume control - as long as you don't get any distortion from it...
(but the safer way is to place it at the end of the effect OR you have to carefully calculate the overall gain so it doesn't lead to any distorted signal..)

[antonis]

I don't understand the design remit - what do you want the buffer to do that you can't do with a variable gain amplifier?

Propably a relatively simple circuit in which he doesn't have to turn the gain to unity for buffering..
(just my opinion, Sam..)

But he could do this with a single JFET (or MOSFET) and a switch from the Source to Drain..

[acehobojoe]

Samhay, The only reason I want to switch is to allow the pedal to have a transparent and non-boosted buffer when the switch is disengaged. I guess there is a more efficient way to do this, but I'm not sure I want to get a 4pdt or anything, a 3pdt is enough to sort through.. haha

I'll add those biasing resistors..
http://i.imgur.com/msVnprr.png


The Idea with c7, is something like this :
http://www.muzique.com/lab/sat.htm
The third one down, but it might not work properly.

[acehobojoe]

Yes, Antonis, that was the idea.. perhaps I could switch back to the first design, add a 10k and fix the popping.

[antonis]

O.K. Joseph but I think that  you may build something more simple like thishttp://www.muzique.com/schem/mosfet.htm...

You can substitute the Mosfet with a Jfet for more simple biasing..

The idea is that you can have 2 switchable indepented outputs (Boost and Buf)

[acehobojoe]

Redid the original design, perhaps this one will be simpler, and won't pop.

http://i.imgur.com/3zGPqSs.png

[karbomusic]

Samhay, The only reason I want to switch is to allow the pedal to have a transparent and non-boosted buffer when the switch is disengaged.

Sounds like it should just be a buffered bypass. Or like I tend to do is just make a separate buffer in a 1590 that goes at the beginning of the pedal chain (or wherever) and throw my clean booster at the end of the chain.

[acehobojoe]

I'll just stick with the double for now, because it won't pop, it essentially is a buffered bypass setup. Then after that I shall try to get more knowledge on what might work better in the future.

[mth5044]

Why are you switching in and out the current limiting resistor on SW1C? In it's current state, you'd be sending power right to the LED and will probably cause it to fail.

[antonis]

Why are you switching in and out the current limiting resistor on SW1C? In it's current state, you'd be sending power right to the LED and will probably cause it to fail.

There isn't such instance because Joseph persists with connecting LEDs in REVERSE POLARITY..!!!!

edit:
Joseph, you have to take in mind that there is a differential drive between the two stages (Boosted singnal is phase inverted) in case you want to drive other phase shifting effects..

[samhay]

Samhay, The only reason I want to switch is to allow the pedal to have a transparent and non-boosted buffer when the switch is disengaged. I guess there is a more efficient way to do this, but I'm not sure I want to get a 4pdt or anything, a 3pdt is enough to sort through.. haha

Be careful with the word 'transparent' as it gets used in many bewildering ways.
I guess you have this on the breadboard? Can you hear the difference between your buffer and the amplifier with gain at 1(ish)?

As I see it, there will be 2 major differences if you have the FET biased sensibly:
1. The phase will be inverted. If you put this at the start of your signal chain before you split the signal, this doesn't matter.
2. The output impedance will be higher in the amplifier than the 'buffer'. Unless this is the last pedal before a long cable run, then it's unlikely to be a problem either.

Ultimately, this comes down to an arguement about what you think a buffer should do. This seems to be a topic of the moment.
http://www.diystompboxes.com/smfforum/index.php?topic=109288.0

Oh, and if 'I' wanted to build a 'transparent' buffer and/or booster, I would use op-amp(s). I think this would be lower parts-count than you current design.

Edit - not that there is anything wrong with using a JFET for your intended purpose(s). It is a popular and well-liked option.

[antonis]

Oh, and if 'I' wanted to build a 'transparent' buffer and/or booster, I would use op-amp(s).

BTW Sam
I've builded your circuit and it sounds realy good..!!! 

And it DOES cut all the way to 0...!!
(maybe my ears need cleaning but honestly there isn't any audible volume difference between Buffer and Bypass mods..)

Also, there isn't any noticeable difference for gain rate level with a linear pot..

[samhay]

antonis - cool, glad you like it.

[acehobojoe]

Yes, you got it.. Oh, and with the LED, it has a built in resistor haha.. And the switch changes the brightness of it basically. Yea, I'm sure the phase shifting won't mess with anything unless it is split and recombined somewhere else.