Help with an EQ pedal

[oliphaunt]

I decided to build an EQ pedal based on the Marshall tone stack from the Duncan tone stack calculator. I have been experimenting with various ways to implement it, and have decided that using a slightly tweaked SHO gain stage before the EQ section to boost the level going into the EQ works well.  I put a 100K volume pot at the end of the EQ section as I still have some gain left to work with even after the loss.  I have experimented with buffers before and after each section and in general I think that the simpler approach is working well.  I have looked at a number of EQ circuits and how EQ is incorporated after gain stages in OD pedals and there seem to be many ways to do this. 

I looked at the ROG Tone Mender where they use a buffer before the EQ then a boost after, but I got too much noise when I tried this method.  Boosting pre EQ seemed to give me the best result. 

Here is the circuit I have so far and it sounds pretty good.  I admit I don't know much about impedance issues, so there cold be all kinds of things I am missing that are causing problems, or may cause issues with some other pedals.  Any advice is appreciated!

[oliphaunt]

I updated the schematic as I had an error in it yesterday.  Any opinions or suggestions are appreciated!

[Colonel Angus]

I didn't check your schemo to carefully, but here is a useful app for designing tone controls:

http://www.duncanamps.com/tsc/

[Pollinator95]

I think he's already used that 

[Colonel Angus]

FACEPALM !!! Right I guess the part where he says "I used the Duncan Tonestack Generator" would have clued me in.

[frequencycentral]

I think it's more usually seen to buffer the input and have a gain stage to recover lost gain (and buffer) at the output.

[oliphaunt]

I agree that is a common arrangement.  I tried this using an IC and two different transistor buffers and each time I got a ton of added noise.   The schematic I came up with works, I guess my main concern is about the output impedance, and just generally if I can expect any issues down the road.

[Ben N]

Because of the high output impedance, you may get varying results depending on what you have following it, so you may want to put an output buffer on the end, after the controls. But other than that, if it works for you and it sounds good, enjoy.

[zambo]

I dont know much about buffers but it seems like the sho goes in front of a lot of pedals with no problems and really thats what you have here with an eq insterted. probably fine. A nice addition to this tone stack is a bright switch with a spdt center off switch and a 100pf on one side and a 500pf on the other from the top of the 100k pot to the middle pin. the center off is nice cause you can turn it off if its to bright. cool design! have you built it already?

[oliphaunt]

Zambo, thanks for the suggestion, I will add that to my breadboarded circuit and see how it works out.  I have not built the pedal yet, still experimenting with the circuit .

Ben N, you mentioned the circuit has a high output impedance.  That is something I am concerned about but I admit I don't know how to calculate the output impedance or change it without adding a buffer.  I know about the input impedance of this boost, so that shouldn't be an issue.

[Ben N]

Oli, I don't know how to calculate it either, but I do know that a tone stack and volume control will make it large. A buffer post will take care of that problem--that is, if it is a problem at all.

[Colonel Angus]

http://www.zen22142.zen.co.uk/Theory/inzoz.htm

Here some pretty thick info on calculating Input and Output Impedance 

[earthtonesaudio]

A buffer on the output would be good.  About the only downside to that arrangement is your total boost amount will be limited by the headroom of the buffer.  But that may be a non-issue depending on how you use it.

[oliphaunt]

Thanks for the additional input.  I will add a buffer.  I don't think that the buffer headroom would be an issue as I am not trying to create a huge boost situation, just enough having enough volume to work with to add a little bit of gain perhaps in some frequency ranges.  If I wanted huge gain I could use a boost at the output instead a buffer.

Here is another schematic.  I am basically cutting and pasting together various circuit blocks.  I added an AMZ transistor buffer to the end, and it left me with several questions.  



I did not use a coupling cap into the buffer since I figure there is no DC in the EQ section.  

In this schematic it is my understanding the input impedance of the buffer is half the value of the two matching bias resistors (in this case 1M so input impedance of 500K.)  I could raise this, and I will test it on the breadboard to see if I hear a difference at different values.

I am hoping to lower the output cap of the buffer to 1u so I can use film cap, as I generally don't like using electrolytics for signal caps.

I have bags of J201 and 2N5457 that "died" over the winter (not sure why, maybe a static issue), and will not work for a buffer, I have tried them.  My BS170s all seem fine which I used them for this project in the first place.  Can I use them here fore the buffer without additional modifications to the circuit?  I tried it on the breadboard and it seems fine, but I would appreciate a second opinion.

I moved the output volume to after the buffer.  I notice that some buffered overdrive designs have the volume before the output buffet (Tube Screamer) while others have it after (Big Muff).   I will try both, but wondering if anyone has a specific recommendation for which is better suited to this project?

[zambo]

I think the output buffer is going to put dc on your eq. If you use a 1uf input cap on your output buffer it will stop it and you should still get full spec frequency from eq. Dont know about the bs170 part. sorry.

[oliphaunt]

Thanks Zambo, you were right about the DC on the EQ.  I added a 1u coupling cap and all is well there.  I tried a lot of different coupling cap values to see how they would effect the EQ range and decided that the 10u output works well to give me more bass at the output.  I was surprised, but I still prefer the smaller .22 cap at the output of the first gain stage.  Much more than that and the bass became tubby but not noticeably deeper. I tried the volume control before and after the buffer and it made an impact on how the EQ worked if I used if before the buffer, so I have left it after.  I updated the schematic above to reflect these changes.

[earthtonesaudio]

The 100k volume control is probably fine but 10k will sound more consistent throughout its rotation (less treble bleed and less reactive to long cables).

The 1uF cap before the buffer, if polarized as shown, should be turned the other way (+ terminal goes to higher potential).  For that matter, if you're using two 10M resistors to bias you can still pass full audio with a much smaller (think: non-polarized) cap.  2nF here will pass 20Hz.

Finally, if you're noticing the bass is too "tubby" with a larger inter-stage cap you might actually be overdriving your buffer.  Remember your .22 cap is working with everything that comes after it to form a voltage divider.  You may be able to get equivalent results with lower noise by reducing the gain and shorting the .22 cap.  After that you might be able to acceptably lower the 10u output cap value, which will help reduce pops when used in conjunction with true bypass switches.

[Ben N]

The last stage in a Muff is not a buffer, strictly speaking, but a gain recovery stage. Since that stage has gain, I think it made sense to put the volume after in order to control the overall volume better, and also to cut the last stage's noise as you cut volume. The last stage of a TS really is a buffer, which is needed there because of the FET bypass switching network, so you wouldn't want anything that is part of the circuit, like the volume control, to come after. Neither of these considerations apply to your circuit. However, the volume control will have some effect on the effectiveness of your buffer as you turn it down--how much I don't know. As earthtonesaudio suggested, you should use a lower value pot for the volume control as it is now, post-buffer. If you run it pre-buffer, the 100k should be ok. You might want to try it both ways on the breadboard, in front of a variety of amp inputs and other effects, to see how it plays.