Mixing grounds and switching in a parallel distortion box

[ianmgull]

I'm in the process of designing a parallel distortion box. My plan is to stuff 2 Tube Screamers and a Rangemaster (positive ground) in a box and run all three of them in parallel.

Below is a block diagram:





I'm not 100% sure how to ground certain portions of the circuit. There will be 2 DC Jacks (one for negative ground, one for positive ground). The DC jacks are plastic and therefore will be isolated from the enclosure (the enclosure will be meet negative ground at the input/output jacks). If I were to used shielded wire going to and from the Rangemaster PCB would the signal be sheilded in ground (-) or ground (+)

The switches will consist of 3 DPDTs. Half of each DPDT will be used for a status LED while the other half will ground the input of it's respective effect. I placed the switch before each effect but I'm not sure if it would make more sense to put this after each effect. (or better still use a 3PDT and ground the effect's input AND ouput)

The "Tri Buffer" listed in the block diagram is based off of the GGG IC Buffer. It will be used to "split" the signal three ways. Layout is below:





I know that the Rangemaster does not play well with buffers so immediately before the Rangemaster circuit I plan on putting the Orman Pickup simulator: http://www.muzique.com/lab/pickups.htm

I know this deals with problems relating to the input impedance of the Rangemaster. Considering the fact that the output of the Rangemaster will be RG's Mini-Mixer should I be worried about that? If so is there any easy way to make the input of the mini mixer to "look" more like my amp's input? (Fender Super Reverb). Mini Mixer @ GGG: http://generalguitargadgets.com/projects/15-boostersrouters/68-mini-mixer-project


Summary:

1) Does my power/grounding scheme look correct?
2) Which ground should the sleeve of any shielded cable going to/from the Rangemaster PCB connect to?
3) Does it make more sense to put the switching system before or after the effect PCBs? (Or should I ground both input AND output)
4) Is the input impedance of the mini mixer close enough to my amp's input impedance that it will sound "good".
5) Am I missing anything? How would you improve this idea? Comments??

You guys rock, thanks a million.

Ian

[cpm]

have you considered a charge pump to provide the negative power to the rangemaster, and thus having only one standard +9v power jack on the box?

[ianmgull]

I haven't. In fact I know very little about charge pumps. How would that work in this situation?

[CynicalMan]

I'd put the switching after the effects because then the noise produced in each of them wouldn't be mixed into the signal. For the charge pump, see: www.geofex.com/circuits/+9_to_-9.htm

[ianmgull]

Thanks guys. I've read the article before but just never realized that principle could apply here.

I have a question regarding the input buffers. The Tube Screamers already have input buffers. Do I really need the "tri buffer" board I was planning on putting at the start? Could I just as well add an identical "tube screamer buffer" right before the pickup simulator (into the rangemaster) and have the same effect? I'm doing the layout and it's getting kind of tight so if I could loose a redundant set of buffers it would help.

[CynicalMan]

I'd just take out the tube screamers' input buffers.

[ianmgull]

Is there any particular reason you would do it this way as opposed to removing the buffer board? I have two Tube Screamers built that sound perfect so I'm trying not to change them. I just figured adding a transistor and a few resistors I could have a third buffer for the rangemaster section and loose a fairly large pcb from a cramped enclosure.

[ianmgull]

I'd put the switching after the effects because then the noise produced in each of them wouldn't be mixed into the signal. For the charge pump, see: www.geofex.com/circuits/+9_to_-9.htm

I really like the idea of using a charge pump here. I'm looking at Mouser and can't figure out what the difference is between the three that are available here:

http://www.mouser.com/Search/Refine.aspx?Keyword=MAX1044


Also I think I read in the datasheet that the output current is only 10mA. I'm pretty sure that is too low for what I have in mind. Am I reading this right???

Data Sheet (pdf):

http://pdfserv.maxim-ic.com/en/ds/ICL7660-MAX1044.pdf

[Taylor]

I really like the idea of using a charge pump here. I'm looking at Mouser and can't figure out what the difference is between the three that are available here:

http://www.mouser.com/Search/Refine.aspx?Keyword=MAX1044

One is surface mount (SOIC-8 package). Of the other 2, 700-MAX1044CPA is fine. The more expensive one just has wider temperature range, useful if you're going to attach your pedal to a rocket or if you play in Death Valley.

Also I think I read in the datasheet that the output current is only 10mA. I'm pretty sure that is too low for what I have in mind. Am I reading this right???

Data Sheet (pdf):

http://pdfserv.maxim-ic.com/en/ds/ICL7660-MAX1044.pdf

You are reading it right, that chip only does 10ma. The LT1054 does up to around 100ma I think.

[ianmgull]

Well I did hear about this really awesome show in the desert...  

To add further confusion to things the 700-MAX1044CPA lists it's output at 20mA here:

http://www.mouser.com/ProductDetail/Maxim-IC/MAX1044CPA+/?qs=sGAEpiMZZMsWkX3fPoxIPbeqF3yEFNJr


Better, but still not enough. Here is what I'm wondering:

1) Could I run JUST the rangemaster from the charge pump. That is supply it from the -9v and ground it to the DC jacks ground (where everything else in the box grounds). This way the ground that the rangemaster sees is 9v "more positive" than it's supply. This is all that matters right?

2) This being the case I would supply everything else (2 tube screamers, mini-mixer, LEDs) from the +9v directly from the DC jack (before the charge pump) and ground them to the DC jack as well. Will this cause any problems considering I'm not using the +9v rail of the charge pump? Does a charge pump need to draw equal current from it's negative rail as it's positive rail to operate properly? Or can I just use the charge pump's -9v rail and ground the rangemaster to the DC Jack's ground?

[ianmgull]

Ok after doing some more research I THINK all I really need is a negative voltage regulator such as a 79L09. This would eliminate the need for a charge pump when I'm really only using half of it anyway. Does this sound plausible? Power the rangemaster with a 79L09 while the rest of the box is powered with no regulator?

[Taylor]

A negative voltage regulator will not turn 9v into -9v. You do need a charge pump if you want to power everything from the same power supply.

Yes, you should only power the Rangemaster from the charge pump. So 20ma should be plenty.

[Quackzed]

why not just do an npn rangemaster?? seems like alot of fuss to mix the 2 polarities when you could just do a negative ground rm?
is there a big sonic difference btw a pnp  rm and an npn  one? maybee there is ? .

[ianmgull]

The real reason is that I have two PNPs on hand (oc44 and oc77) that I got from smallbear a while ago and never used. I breadboarded the RM and love the way the oc77 sounds. I'm thinking it won't be that much of a hassle if I just use the MAX1044 for the rangemaster and nothing more. As it stands I'm still going to have to make an auxiliary board for the rangemaster. Since I've abandoned the idea of having a dedicated buffer board (the tubescreamers are already buffered) I'll just build a third TS style buffer on this aux board, along with the pickup simulator and the charge pump (hopefully there are no problems with the charge pump being in such close physical proximity to the signal path here).

[Quackzed]

sure, that makes sense. after all 'the sound' is the most important thing... i just figured ,with all else being equal, it might be easier...
actually i seem to remember reading that there is indeed a sonic difference btw an npn and a pnp RM... though thats a whole 'nother discussion...

[ianmgull]

OK. So I just tested out the mixer section with 2 of the Tubescreamers.... Success!! I didn't even have to put a buffer before the Tubescreamers; their internal buffers worked fine. I've got a question though:

The Mini-Mixer has a 10k volume pot at the input of each of the channels. The Tubescreamers have volume pots immediately before the output buffer. In the interest of simplicity one of these volume controls must go. My question is which one should I remove?

I think the obvious answer would be the Tubescreamers' volume pots. I could put a fixed resistor on both of the Tubescreamers to simulate their volume pots being turned all the way up all the time.

The other thing to consider is that the rangemaster will have two "gain" pots. The first being the normal 1 knob that is on the Rangemaster (which will act more like a gain control in this case). The second one being the input to the mixer.

Any feedback would be greatly appreciated!

[wavley]

I think this is a great idea, I run parallel dirt myself.  The issue that I haven't seen addressed here is phase.  I don't own a tube screamer so I don't know if the output is 180 degrees out of phase with the input, but I do believe that the rangemaster by the nature of being a common emitter amplifier is going to be out of phase with the input, which isn't a problem until you mix it with a signal that IS in phase with the input, now you end up with weird phase cancellations.  In my case, the real lows went away completely, the low mids were boosted and phasey, and my highs were hollow and weird.  I went through my rig with a scope looking at the phase between my two signal paths and finally ended up flipping the phase of an isolation transformer I had in my splitter box and everything is golden now.

[ianmgull]

Wow. Good point. I can't believe I forgot about that. Looking at an older explanation I dug up this:

ian:
is not that hard..look:
- if the output of a transistor or fet stage comes from the drain (or the collector), the stage is inverting (typically gain stages)
- if the signal in an opamp goes into the - input, the stage is inverting.. (some gain stages)

- if the output of a transistor or fet stage comes from the source (or the emmiter), the stage is non-inverting (typically buffers)
- if the signal in an opamp goes into the + input, the stage is non-inverting.. (some gain stages, and buffers)

now, follow the signal through each stage of the pedals, checking the polarity of each output..

do it like this:
for each stage asign number 1 if non inverting and -1 if inverting, for example , a signal chain:

inverting--> non-inverting-->inverting----> inverting
is:
-1 ---> 1 ----> -1 ---> -1

then multiply the numbers:

= -1*1*-1*-1= -1...then the entire chain is inverting

as a rule, if the chain has an odd number of inverting stages, the output will be inverted..

then..you have to invert the signal again on those pedals having inverted outputs..

So yes. The Tubescreamers do NOT invert their output while the Rangemaster does. This isn't the end of the world though. I just have to use an inverting buffer before the Rangemaster instead of the Tubescreamer style buffer I had planned on.

[ianmgull]

The problem is that since the other two Tubescreamers are being buffered by the standard Tubescreamer buffer I believe I want the impedance of the rangemaster buffer to be some what similar to the others, or does this matter? Could I just get away with throwing an inverting op amp stage before the Rangemaster?

[wavley]

My problem is that both of my signal paths were great until I turn on my EHX 16 Delay, it inverts.  So now I leave it in buffered bypass and flipped the phase of the other channel where it splits, I already had a buffered iso transformer there so all I had to do was add a switch so I could chose my phase.