TH Black Eye (Soldano GTO) prototype build

The question we should be asking is "Why do we need to add a buffer to cure squealing?" Squealing is just the symptom, not the root cause of the problem.
Very confined PCB with Input & Power traces close together combined with High Gain would be a possible cause!
The Original was Noisy & is equivalent to 3- 1590BB's.
The Sib Cuda runs the same 9v to 200v + Power converter but only One valve is involved.
 
DC-DC converters are by their very nature noise sources. It takes a lot of work to
a) reduce noise at the source
b) minimize coupling
c) decrease susceptibility of the high-gain circuits

We have to address filtering, board layout, grounding and off-board wire routing to accomplish those things.
 
The question we should be asking is "Why do we need to add a buffer to cure squealing?" Squealing is just the symptom, not the root cause of the problem.
I suspect that it's a routing issue. The layout is more compact than it probably should be on account of the tube socket mounting setup, so the routing probably isn't following best practice for high gain design.
 
The question we should be asking is "Why do we need to add a buffer to cure squealing?" Squealing is just the symptom, not the root cause of the problem.
I've stayed out of this one for a while and privately told certain individuals what I think but ultimately, you and I are in agreement that putting a buffer in front of it is like putting a painting over a hole in the wall. You don't see the hole anymore but it is very much still there.

So what is the buffer doing?
driving signal current and masking the parasitic current causing the squeal.

The obvious answer here is that the layout is too compact and poorly designed. I'm sorry if that hurts the feelings of the vendor but we learn from fixing mistakes, not hiding them and providing a buffer pcb is a duct tape and baling wire solution at best.

A number of answers can be found here https://www.aikenamps.com/index.php/white-papers. Merlin's book on tube amp design is a grail. Following vendor instructions has you grounding the input jack to a power supply ground. It also has you running the input signal unshielded and parallel to both power supply and heater runs. There's the previously mentioned impedance issue allegedly inherent to the design (no verified trace with pictures, etc. currently exists to confirm). I could go on but I see others are posting replies so I'll cut this short
 
I do want to leave one last thing, this problem isn't unique to this particular pedal. I've seen this out of a number of high gain, solid state amp in a box pedals. I once pulled an all-nighter fighting with a DIY SansAmp GT2 to get rid of squeal and still ended up with a hissy mess with the gain maxed. Sometimes you can't round the corners off enough to get that square peg into the round hole.
 
So how did that different buffer workout? less noisy?
It was less noisy, but it also dropped the gain a bunch. I think I may have messed up somewhere in my vero layout. I'll have more time to work tonight than I did last night, I'll see if I can figure it out and try again.
 
I suspect that it's a routing issue. The layout is more compact than it probably should be on account of the tube socket mounting setup, so the routing probably isn't following best practice for high gain design.
Have you considered offloading the boost converter to a daughterboard that mounts through header pins to the main board? You could likely fit something along the input side or output side wall of the enclosure (the output side is probably a slightly better choice). The additional room on the main board may allow for more effective routing.
 
Vigilant normally works with smd on completed pedals, better component construction and smaller sizes… what fits in a smartphone now could fill a mid 70’s marshall jtm 800 cab with bits out the back in through hole and spacing for RF avoidance easy.
 
Have you considered offloading the boost converter to a daughterboard that mounts through header pins to the main board? You could likely fit something along the input side or output side wall of the enclosure (the output side is probably a slightly better choice). The additional room on the main board may allow for more effective routing.
I haven't really considered it. The goal of the whole project was to make it easy to assemble, once I open the door to daughter boards it completely changes things. The easiest way would be to put the tubes on daughter boards and mount the PCB horizontally, which is what I do for my usual builds and it gives me a lot more room to work with.

My SMD version of Black Eye fits in a 125B because the components are so much smaller, but I actually used a 4-layer PCB for that so I could route it more cleanly, and it has never had oscillation issues. I will admit I spent a lot more time on that layout than I did on this one.
 
Quick note. I put 12au7 tubes in. No squeal. I'm going to probably try a combo of 12au7 and 12ax7/12at7. I like the 12au7s but I need a little more gain. I didn't need a buffer, either. I really like the 12au7s but would like to have 12ax7 build and 12au7 build. Different pedals but both really good.
 
Quick note. I put 12au7 tubes in. No squeal. I'm going to probably try a combo of 12au7 and 12ax7/12at7. I like the 12au7s but I need a little more gain. I didn't need a buffer, either. I really like the 12au7s but would like to have 12ax7 build and 12au7 build. Different pedals but both really good.
@tcpoint, how much gain do you get with the 12AU7s in the BE? Can you get it "clean"? I'm doing that with my Space Heater and a 12AU7 but I'm wishing for some basic EQ, would be cool to be able to do it with the BE.
 
When you use 12AU7s, the gain know is really only useful from about 11am - noon and up. When I turn the volume all the way up and turn the gain to unity, it is somewhat clean unless I hit the strings hard. If I do that it goes from edge of breakup and a little bit past that. That's with humbuckers. With single coils, it stays pretty clean. I do like this with 12AU7s but it's more of low - low/medium gain pedal. I'm going to experiment with different combos, tomorrow. I'll probably figure out where the squeals come from (V1 or V2). My feelings, this is DIY and that implies tweaking and sometimes it doesn't turn out the way you envision. This is a fun pedal and it's going to eventually earn a spot on my board.
 
Quick update. I put a 12AX7 in V2 and it sounds great. No squeal (no buffer needed). I'm going to try a 12AT7 in V1 (I have a 12AU7 there now). I'm convinced that squeal came from V1. I tried a 12AT7, but I can't remember if I tried it in V1 or V2.
 
I may have missed it, but what’s the status of the board and the squealing? Is the solution an integrated buffer on the switching daughterboard, or are there plans to revise the board to address the issue?
I had been experimenting with a buffer on vero and it seems like that does the trick. I ordered new footswitch PCBs with a built-in buffer that just shipped from China today, so I should see them next week. Once I have the chance to verify that they're going to work the way I want I'll let everyone know and start getting them out to people.
 
Back
Top