Soul Sonic - Crackle NOT Okay!

soulsonic · Post by **soulsonic** » 29 Nov 2009, 05:27

Thanks! I'm glad you guys are diggin' it!
It is definitely very loud.

fosnal1950 · Post by **fosnal1950** » 29 Nov 2009, 12:15

It is definitely very loud.

Couldn't you just change R4 into a 100K pot to get a bit more control ?

RnFR · Post by **RnFR** » 29 Nov 2009, 19:57

yes, you could always add a master volume control.

trad3mark · Post by **trad3mark** » 09 Dec 2009, 10:55

hey soulsonic, thinking of having a go at adapting this into a super duper, and also, use a tube instead of fets. ok if i borrow parts of your schem to give it a bash?

soulsonic · Post by **soulsonic** » 09 Dec 2009, 23:46

trad3mark wrote:hey soulsonic, thinking of having a go at adapting this into a super duper, and also, use a tube instead of fets. ok if i borrow parts of your schem to give it a bash?

Sure, but with all those changes it would certainly be hard to tell which parts came from me.

One thing about tubes, though - I don't think the SHO-style feedback biasing will work with tubes. This is a biasing style that works with BJT transistors and MOSFETs. Tubes bias in a manner very similar to JFETs, and JFETs do not function in this kind of circuit. I recommend reading up about "Common Cathode" tube stages.

trad3mark · Post by **trad3mark** » 10 Dec 2009, 14:31

had a little go of it just there. You get a lot less gain than with your crackle not okay. I made a tonne of changes, but as it's still essentially based on yours, i thought it'd be the right thing to do to check with you first.

I then tried it with the normal SHO circuit, and although you get similarly low levels of gain, there's no crackle, which i'm guessing is partly cos of the tubes. I think i'm onto something, but the question now is, keeping it running on 9v, what's going to be the easiest way to get the gain up. I'm guessing if i muck about with the resistor between the +9v and where the drain WAS, then that'd help. Any ideas on other values i could try that would affect gain (in a big way)?

I think it'd work better as a sort of Super Duper Valve in One...

soulsonic · Post by **soulsonic** » 10 Dec 2009, 19:32

Are you using an ECC83 type? Maybe try a tube that's intended for lower voltage, like one of the subminis. The only time I've ever gotten a decent sound from a high voltage tube running at low voltage is when it was used as part of an overdrive like in a Tube Driver.
If you want to stick with an ECC8x type, then I highly recommend trying it at higher voltage. You can run a simple transformer setup like EHX uses for their tube series and easily get a usable high voltage supply. Even many of the low voltage submini tubes work much better at voltages higher than 9v. By sticking with 9v, you're putting an insurmountable brick wall limit of how much work you can physically get the tube to do.

Mr. G. · Post by **Mr. G.** » 10 Dec 2009, 20:02

Just noticed a post on TGP about the crackle not ok... http://www.thegearpage.net/board/showth ... 978&page=2 post #24

Thought it might be relevant to this thread.

soulsonic · Post by **soulsonic** » 10 Dec 2009, 21:49

Mr. G. wrote:Just noticed a post on TGP about the crackle not ok... http://www.thegearpage.net/board/showth ... 978&page=2 post #24

Thought it might be relevant to this thread.

I saw that a week or two ago. I'm banned from there, or I would have probably replied (which probably would have resulted in me being banned... of course!).
Clearly Vex has no clue how an opamp works, or the basic concept behind using feedback to control gain.

earthtonesaudio · Post by **earthtonesaudio** » 11 Dec 2009, 13:55

I attempted to set Zachary straight in post #38. So far no rebuttal.

5thumbs · Post by **5thumbs** » 19 Jan 2010, 01:56

First off, Mr. Earthtonesaudio, that was an excellent reply to Mr. Vex. Respectful, but firm...well played.

Next, Soulsonic, here is the CNO-SHO v2 I mentioned in the other thread.

Nothing shocking here in the way of additions. This is similar to how I implemented the CNO-SHO v1 in a couple of my designs. I wanted to reduce series resistance (for noise reasons), but I didn't want to go back to the "Crackle Okay" SHO. By adding the op amp buffer before the CNO-SHO v1, I could reduce the size of the pot/resistors used in the CNO voltage divider, but kept the ratio between them the same as Soulsonic's original CNO design. All of this while still enjoying the high input impedance (or higher) of the original SHO via the buffer.

RnFR · Post by **RnFR** » 19 Jan 2010, 06:38

does the MOSFET bias up with those small resistors? did you try this?

soulsonic · Post by **soulsonic** » 19 Jan 2010, 06:55

yeah, I wonder about that too. When I was experimenting with the SHO stuff, it really seemed like 100k was about as low as is practical.

5thumbs · Post by **5thumbs** » 19 Jan 2010, 07:15

RnFR wrote:does the MOSFET bias up with those small resistors? did you try this?

Not yet. 100K series is still better than 1M, but that's still a pretty big resistance (at least for the project I'm trying to squeeze those out of.)

IvIark · Post by **IvIark** » 19 Jan 2010, 17:06

Here's a turret layout for anyone interested:

: Crackle Not OK - IvIark.png (38.77 KiB) Viewed 1909 times

5thumbs · Post by **5thumbs** » 19 Jan 2010, 17:10

OK, didn't sleep well last night, part due to this friggin' cold I've got today, part due to thinking about CNO-SHO v2 and its return to 100K series resistance in my design.

If the MOSFET won't bias with low series resistance as proposed, then perhaps an op amp-based solution might work? R.G.'s Adjusticator comes to mind, but as he spec'd it, it too has a 100K series resistance before the opamp. However, on at least some single op amps such as the OPA134PA, the data sheet indicates that it might bias with much lower series resistance in front of the inputs. I'm going to have to test this to confirm proper operation, but here's the circuit I'm thinking of.

5thumbs · Post by **5thumbs** » 19 Jan 2010, 17:14

And here's the corrected CNO-SHO v2 schematic, with series resistance and biasing voltage resistors at their 2M2 original value. This iteration has a higher input impedance than CNO-SHO v1, but still has high input series resistance (100K) due to biasing requirements for the MOSFET.

earthtonesaudio · Post by **earthtonesaudio** » 19 Jan 2010, 18:01

You may have reduced series resistance, but using 10M to bias the op-amp will likely make even more noise than the original.

The buffer will also present the guitar with a fixed load, so you'll lose the interaction with the FET stage as you increase the gain.

5thumbs · Post by **5thumbs** » 19 Jan 2010, 18:18

earthtonesaudio wrote:You may have reduced series resistance, but using 10M to bias the op-amp will likely make even more noise than the original.

The buffer will also present the guitar with a fixed load, so you'll lose the interaction with the FET stage as you increase the gain.

I agree. It is the change in the latter that has me more concerned than the former. I think all the iterations as-currently-posted would work, but I'm not convinced we've come any closer to an improved (crackle-free) emulation of the SHO than Soulsonic's original.

The question that the designer will have to answer for themselves might be, "Rank from most to least important the following: 1) Varying MOSFET/Guitar load interaction at different bias points (features of both the original SHO and CNO-SHO v1; 2) Low series resistance; and 3) High input impedance." It appears that you can't have all three, so pick two that are most important and then pick your CNO-SHO topology to match your prioritization.

Silent Fly · Post by **Silent Fly** » 20 Jan 2010, 09:29

5thumbs wrote:(...) I wanted to reduce series resistance (for noise reasons), but I didn't want to go back to the "Crackle Okay" SHO. By adding the op amp buffer before the CNO-SHO v1, I could reduce the size of the pot/resistors used in the CNO voltage divider, but kept the ratio between them the same as Soulsonic's original CNO design. All of this while still enjoying the high input impedance (or higher) of the original SHO via the buffer.

You could try this solution as well:
https://www.freestompboxes.org/viewtopic.php?t=5811

fig.5 in the PDF file should do what you are looking for.

Soul Sonic - Crackle NOT Okay! [documentation]