Help with adding diode distortion to a tube preamp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

sonolink

Well-known member
Joined
Feb 15, 2010
Messages
1,277
Location
London-Madrid
So I built a JTM45 preamp clone inside a pedal here:
https://groupdiy.com/index.php?topic=74809.0

This is the final schematic of the pedal:


When used in the return fx loop of an amp it would be cool to be able to add some more distortion. I guess that would mean adding gain right?

I was wondering if diodes would be a good and simple option for this?

Thanks
Sono
 
Why bother with tubes if you're just going to use diodes. Here are 3 alternative things to try:

1) Add "bright" caps. Specifically, add a 500p across the Drive pot pins 2 - 3. And maybe another 500p across the 270k grid stopper.

2) Overdrive it. Put an op amp in front running on 30V (use a simple zener reg to make 30 if you don't already have that avail) and a virtual ground with 20+dB of boost. You might even go high voltage and get an OPA551 and use 60V supply. Then you can probably do 40dB. Or re-wire the thing to use the other tube for more gain. Or try a pedal that has gain (but anything running on 9V probably isn't going to be enough). If you have a mic transformer with a 1:10 step-up, drive that with some pedal and run it into the input. But the op amp (or other tube) is the easiest and most direct solution.

3) Increase the 270k grid stopper. Grid current limiting with a large grid stopper (like 1M) will give a little more edge because it will make the clipping on positive swings sharper and the envelope filter created by it will decay longer. Usually large grid stoppers are on output tubes but there's no reason why it shouldn't work in a preamp. But it won't work at all unless you add enough boost in front because you need to push the tube into grid-current-limiting.

Generally speaking, that circuit just doesn't have enough gain to really get nasty. You need another 40dB in front.
 
squarewave said:
Why bother with tubes if you're just going to use diodes.

While I might generally agree,  have you ever played a silver jubilee?  Quite a nice sounding amp.  Sometimes the combination of diodes and tubes is just what you need.  A tubescreamer into an all tube amp is another classic combo.
 
john12ax7 said:
Look up the marshall jubilee for an example.

You mean doing something like this?



squarewave said:
Why bother with tubes if you're just going to use diodes.

Well that's the whole point really :) The idea behind this is to have a tube amp without having to carry a heavy cab, being able to use it as a distortion/overdrive/coloring pedal in  front of an amp (which it does great, rendering a huge palette of colors and textures), or as a real tube preamp in front of a power amp (using the FX Loop return). In this situation, if you want some overdrive/distortion as you would from a tube amp, you would strap a pedal in front of the amp (Tubescreamer, KoT, SD9, OCD, etc). That's how I thought of using diodes, since to have another tube gain stage would make the unit very bulky, the circuit more complicated, etc.

squarewave said:
Add "bright" caps. Specifically, add a 500p across the Drive pot pins 2 - 3. And maybe another 500p across the 270k grid stopper.
Would that actually provide overdrive/distortion? I'd have to make it switchable.

squarewave said:
2) Overdrive it. Put an op amp in front running on 30V (use a simple zener reg to make 30 if you don't already have that avail) and a virtual ground with 20+dB of boost.

You mean something like this?




Thanks a lot to both for your time and help
Cheers
Sono
 
sonolink said:
Would that actually provide overdrive/distortion? I'd have to make it switchable.
Not exactly. The first cap wouldn't do anything with the gain control in the maximum position. But in any other position it would let the highs through. There's usually a lot of high passing / bypassing in guitar amps whereas your circuit has very little (just the partial bypass of the cathode resistor in the other channel). I was wrong about the second cap. It would have no effect on your particular circuit.

sonolink said:
You mean something like this?
Sort of. You would need to make 30V somehow and then split that to make a virtual ground and then connect the input resistor and lower feedback resistor to it.

But making the 30V (or +-15) is non-trivial depending on what your supply is. Maybe you already have it? What is your power supply?

But if you do use an op amp you might as well use a dual and use the other half to buffer the output because your output Z is way high.
 
squarewave said:
But making the 30V (or +-15) is non-trivial depending on what your supply is. Maybe you already have it? What is your power supply?

Attached :)

squarewave said:
But if you do use an op amp you might as well use a dual and use the other half to buffer the output because your output Z is way high.

Sorry, I got lost...what do you mean?

Cheers
Sono
 

Attachments

  • PSU.jpg
    PSU.jpg
    239.8 KB · Views: 18
sonolink said:
Sorry, I got lost...what do you mean?
The output impedance of your circuit is 200K+. So whatever you plug it into has to be very high impedance (like 1M of another tube amp) or you will loose level, tone, load the output, etc. So preferably it should be buffered. If you do insert an op amp, make it a dual and add a simple output buffer.

Unfortunately your B+ is not ideal for making the 30V (or +-15) that you would need. Your tubes probably use <4mA. But if tap into B+ for your op amp supply that could easily become 10mA+ which might be too much for your little flyback switcher. Your supply would work great as is with 18V input (max for NE555). Then you could make a +-9V virtual ground which would be good enough for the op amp and not tax your B+.
 
squarewave said:
The output impedance of your circuit is 200K+. So whatever you plug it into has to be very high impedance (like 1M of another tube amp) or you will loose level, tone, load the output, etc.

Sorry, but I still don't understand this. The output will be plugged into the front or into the FX loop return of an amp. The only thing I miss is being able to have some more overdrive when plugged into the power amp. Wouldn't the diodes be a simpler solution? Sorry for the language barrier :)

squarewave said:
Your supply would work great as is with 18V input (max for NE555). Then you could make a +-9V virtual ground which would be good enough for the op amp and not tax your B+.
You mean substituting the 9v input of the PSU for an 18v one, right?

What about using this?
The 9v PSU I'd be using provides 1700mA
 

Attachments

  • voltpmp2.gif
    voltpmp2.gif
    4.3 KB · Views: 10
sonolink said:
You mean substituting the 9v input of the PSU for an 18v one, right?
Yes. But then again you would need to run your heaters on 12.6V because the current drawn through the regulator would be a lot more than necessary. Meaning swap the 78L06 with 78L12 and tweak your heater wires for 12.6V.

sonolink said:
What about using this?
That could work but note that that IC is also a switching mode device and if you mix switchers you can get beating noises at the difference in switching frequencies which could easily end up in the audible range.

sonolink said:
The 9v PSU I'd be using provides 1700mA
Ok but just realize that if you draw 4mA @ 195 = 0.78W / 9V = 87mA from your input. So if you draw 10mA for example (because you derive an op amp supply from B+) that would be 10mA @ 195 = 1.95W / 9V = 217mA (and that's on top of what your heaters are consuming). Meaning your mA on the input might be much more than you think. Meaning just adding one little op amp makes your input current jump 130mA! So you probably want to prefer solutions that use your input and not tax B+.

Or just use diodes!
 
sonolink said:
Is my diodes schematic ok?
I don't know anything about putting diodes in a tube amp. But what strikes me about this is that it's going to limit the level so much that your other stages are probably not going to clip at all. You might want to put them on the output of your follower stage. But there's also another problem which is that you're basically shorting the poor tube they're connected to. You should probably have like a 10K or 100K in series before. But maybe look at that "jubilee" amp john12ax7 was talking about and see how they did it.
 
squarewave said:
But maybe look at that "jubilee" amp john12ax7 was talking about and see how they did it.

It's what I did :)

This is the Jubilee schem. I was hoping someone could tell me if my adaptation is correct.

https://i0.wp.com/oldschematic.com/wp-content/uploads/2019/05/Marshall_25aniv_silverjubilee_50_100w_2555_00.png?fit=2633%2C1813&ssl=1

JohnRoberts said:
or use a guitar efx pedal.

Thanks for chiming in, John. But isn't that the same as using diodes or putting an opamp at the input?
I was hoping to include it inside the unit itself...

Cheers
Sono
 
Why not just run the opamp off the 9V you already have? Like almost any guitar pedal out there. 8Vpp is plenty to boost a tube frontend.
 
volker said:
Why not just run the opamp off the 9V you already have? Like almost any guitar pedal out there. 8Vpp is plenty to boost a tube frontend.
An e-guitar can make 4Vpp if you jam on it just so. That limits the gain you could add to ~6dB before the OA starts to clip. Maybe you stretch that to 10dB reasoning that you want to clip transients anyway. But any more than that and the OA is potentially going to clip / latch-up in funny undesirable ways. Gain would need to be more like 20dB before it's worth it but even that is way lower than what a typical tube stage makes (30dB+). Now if there was a neat way to do an OA stage that clipped "nicely" at +-3V instead of +- 0.7V like the typical diodes in the feedback path of an OA does, that might be useful in this case.
 
sonolink said:
Thanks for chiming in, John. But isn't that the same as using diodes or putting an opamp at the input?
I was hoping to include it inside the unit itself...

Cheers
Sono
Sorry, do whatever floats your boat... Most guitar players I know use multiple different pedals.  Keeping the EFX outboard makes it easier to tweak.

JR
 
squarewave said:
An e-guitar can make 4Vpp if you jam on it just so. That limits the gain you could add to ~6dB before the OA starts to clip. Maybe you stretch that to 10dB reasoning that you want to clip transients anyway. But any more than that and the OA is potentially going to clip / latch-up in funny undesirable ways. Gain would need to be more like 20dB before it's worth it but even that is way lower than what a typical tube stage makes (30dB+). Now if there was a neat way to do an OA stage that clipped "nicely" at +-3V instead of +- 0.7V like the typical diodes in the feedback path of an OA does, that might be useful in this case.

On a superhot humbucker, the first attack, maybe. But yet, the big majority of players don't deem this to be a problem. The thousands of different pedals doing "clean boost" running on 9V prove that it doesn't bother guitarists in practice. You might be surprised about the amount of distortion guitar players perceive as clean.

But if you are concerned about opamp clipping, use a simple transistor circuit like the ZVex SHO or Jack Orman's Mini Boost. Clips more gracefully and still provides more than enough gain.
 
volker said:
But if you are concerned about opamp clipping, use a simple transistor circuit like the ZVex SHO or Jack Orman's Mini Boost. Clips more gracefully and still provides more than enough gain.
Well there ya go. That looks like a good circuit for this scenario. It's single supply, high Z, probably pretty high gain and should clip nicely. Perfect fit here.
 
Back
Top