Thoughts on an interesting issue...

[bradicusmaximus]

I'm hoping someone with a bit of experience with some 'wreck type amps might have some thoughts.

A buddy of mine brought me an amp for repair. His description was basically, "it just doesn't sound right and seems funky". Given many other people who have brought amps to me, it's actually a much more helpful comment than it seems. So, to the bench it went. Unfortunately, I've never dived into the 'wreck world too much so while it seems familiar, it's also a bit odd to me. Anyway, it appears that it was probably built as a Weber kit or at least built based off that layout.

Upon pulling it out and playing with it, I'm getting a high frequency oscillation that seems most apparent when the channel volume is set around 2-3. It subsides if it gets higher. Chopsticking the board, I get a noticeable variation in the frequency around pin 6's lead. However, the fun part is if I put my hand over the preamp section, I can make it react like a theremin. I did ask my buddy if he wanted a 'wreck theremin, but he seemed to want the actual amp for guitar - so I figured I'd try to fix it right.

My initial thought is that it seems like a lead dress issue, but before I started fumbling blind, I thought I'd at least ask here and see if anyone had any alternate suggestions.

Thanks!

Brad

[xtian]

You're exactly right. 'Wrecks are known for being on the edge of oscillation. New builders are encouraged to copy exactly the original, known good layouts, because the circuit can very easily become unstable.

Post photos for us, and we'll try to help.

[RockinRocket]

Does the amp have grid stoppers on v1 grids? Shielded wire from input to V1 is standard.
I can get my builds to not osscilate with everything dimed with no grid stoppers, but from now on, when I part with my amps grid stoppers come on just in case.

[bradicusmaximus]

Thanks for the replies. I've added a few pictures of the guts. The one shot has a blue box around where I seem to be getting the most reaction. While I'm not 100% certain, I've also included what I believe to be the kit / layout used by the original builder - it took a while to track down, but it seems to match up.

I've considered re-running some of the signal wires with shielded instead of that solid core cloth covered wire everywhere. Is there a better way to run some of the leads that are a bit more "conventional" or proven to have better results in this design?

I'm open to any thoughts or suggestions you might have.

Thanks!

IMG_2682.jpg

IMG_2683.jpg

IMG_2684.jpg

java_layout.jpg

java_schem.jpg

[Littlewyan]

The two wires for the bright caps are way too long. It's also not ideal having the switch so far from the volume pot but if you shorten those wires so they run direct and not over the board first that should help.

That's quite a big mess of wires around the output section so I don't know how much we can do to help but we'll try . Not sure why they put the treble pot as far from the volume pot as possible. What's underneath the black tape? Looks like a small PCB.

I'm away until Sunday night now but will reply when I can.

[Colossal]

Hi Brad,

I don't mean this is a mean way, but honestly, if it were me, I would gut the amp to the bare chassis and start over. That layout is not at all conducive for that much gain. With an amp like a Trainwreck, there is some small forgiveness, but not much. If you look at the originals, the signal path is very short and the components are given breathing room. That is not by chance. Even then, some guys chase oscillation with seemingly perfect builds. If you look in the For Sale section right now, Dynaman has a beautiful alternative layout. Maybe you can chopstick around and get some stability, but there was another guy on here before that had that same Weber build and if I remember correctly, he was chasing problems too.

Just my two cents.

[bradicusmaximus]

The two wires for the bright caps are way too long. It's also not ideal having the switch so far from the volume pot but if you shorten those wires so they run direct and not over the board first that should help.

That's quite a big mess of wires around the output section so I don't know how much we can do to help but we'll try . Not sure why they put the treble pot as far from the volume pot as possible. What's underneath the black tape? Looks like a small PCB.

I'm away until Sunday night now but will reply when I can.

Thanks for the input. I'll definitely be re-routing those wires. It is pretty much a mess in there - especially those transformer leads. I've got some good amount of cleanup, but I'll start with that bright switch.

[bradicusmaximus]

Hi Brad,

I don't mean this is a mean way, but honestly, if it were me, I would gut the amp to the bare chassis and start over. That layout is not at all conducive for that much gain. With an amp like a Trainwreck, there is some small forgiveness, but not much. If you look at the originals, the signal path is very short and the components are given breathing room. That is not by chance. Even then, some guys chase oscillation with seemingly perfect builds. If you look in the For Sale section right now, Dynaman has a beautiful alternative layout. Maybe you can chopstick around and get some stability, but there was another guy on here before that had that same Weber build and if I remember correctly, he was chasing problems too.

Just my two cents.

No worries at all on being mean. It's not my build and I don't even really know the dude who did it - I'm just helping a buddy of mine out by seeing if I can get this in service for him.

I don't know if I really want to get into a complete rebuild, but I'll probably buckle down and figure what will be the most effective use of my time (of course after I'm stubborn and spend way too much time trying to chopstick and whatnot - lol).

From what I've seen in my own research, this does seem to be universally noted as not an optimal layout. I will make my way over to the for sale forum and check out the other one.

Thanks!

[Colossal]

No worries at all on being mean. It's not my build and I don't even really know the dude who did it - I'm just helping a buddy of mine out by seeing if I can get this in service for him.

I don't know if I really want to get into a complete rebuild, but I'll probably buckle down and figure what will be the most effective use of my time (of course after I'm stubborn and spend way too much time trying to chopstick and whatnot - lol).

From what I've seen in my own research, this does seem to be universally noted as not an optimal layout. I will make my way over to the for sale forum and check out the other one.

Thanks!

I hear ya, it's a lot more that you probably want to deal with. I've been there before; trying to help a buddy out. You might get lucky with a few wire moves, hopefully so. Good luck!

[bradicusmaximus]

No worries at all on being mean. It's not my build and I don't even really know the dude who did it - I'm just helping a buddy of mine out by seeing if I can get this in service for him.

I don't know if I really want to get into a complete rebuild, but I'll probably buckle down and figure what will be the most effective use of my time (of course after I'm stubborn and spend way too much time trying to chopstick and whatnot - lol).

From what I've seen in my own research, this does seem to be universally noted as not an optimal layout. I will make my way over to the for sale forum and check out the other one.

Thanks!

I hear ya, it's a lot more that you probably want to deal with. I've been there before; trying to help a buddy out. You might get lucky with a few wire moves, hopefully so. Good luck!

That is my hope. My plan of attack is to try and hit some of the low hanging fruit in cleaning up the power supply area and then re-working the bright switch wires. I'll probably throw in some shielded cables where appropriate and hopefully that will get it to a workable status. Otherwise, my buddy will have the world's coolest theremin. lol

[M Fowler]

The main problem I'm sure is the presence circuit which is right in the boxed area you show.

Disconnect the presence and all the noise will go away.

Then you can try figure out the presence or make it a cut control and ground the tail resistor.

Mark

[romberg]

A java kit was the very first amp I ever built. I still have mine. It has been slightly modified and works fine as far as I am concerned. I am able to dime all the controls and the amp does not oscillate. Which is one of the typical "bench tests" for this amplifier. Sure this may not be the best layout for this circuit. But it can be made to function. I've not gone back and rewired mine because to me, it sounds good.

The pictures you have show some things that could use some tidying up. Heck mine could use some as well. But I don't think it is at the point where one would be better off gutting it and starting over. I agree with the notion that the brite switch is probably one of the main issues here. The brite switch *is* connected to and thus *is* the grid of the second gain stage. As the switch is wired now those wires are hanging all over other parts of the circuit. It should be moved to be right next to the volume pot and not on the other side of the input jacks. One could mount something else in the unused hole or leave a non-functional switch in there as a mystery "tone" boost for the discerning ear. I'm sure someone will hear a difference when it is engaged .

Below are some pictures of my amp. Which does function fine without any oscillations. I have made these modifications to the circuit to make it a pretty close match to the kelley amp documented here: http://ampgarage.com/forum/files/kelly90schemlayout.zip

- Add brite switch (right next to the volume pot)
- Shield run from volume to stage 2 grid (kelley does not have this but the run is long enough that it is a good idea)
- Take out/don't use cathode follower
- 1MA treble pot
- 1M bass pot
- Remove grid stopper on stage 2 (R34) 1k (did not notice any difference in "tone")
- cathode bypass cap on stage 2 to 22uf
- PI tail resistor (R25) 22k -> 10k
- LarMar (type II) master volume

After these changes my circuit is more or less identical to the kelley with the exception of the power supply and grid stopper on V1. The changes that made the biggest difference were 1) Take out the cathode follower (if it has been wired up). 2) Replace the two pots in the tone stack. 3) PI tail resistor. Also, using a type II master volume means you can get rid of those extra caps.

Finally a few pictures of mine along with a schematic for the LarMar/Type II master volume.

[bradicusmaximus]

A java kit was the very first amp I ever built. I still have mine. It has been slightly modified and works fine as far as I am concerned. I am able to dime all the controls and the amp does not oscillate. Which is one of the typical "bench tests" for this amplifier. Sure this may not be the best layout for this circuit. But it can be made to function. I've not gone back and rewired mine because to me, it sounds good.

The pictures you have show some things that could use some tidying up. Heck mine could use some as well. But I don't think it is at the point where one would be better off gutting it and starting over. I agree with the notion that the brite switch is probably one of the main issues here. The brite switch *is* connected to and thus *is* the grid of the second gain stage. As the switch is wired now those wires are hanging all over other parts of the circuit. It should be moved to be right next to the volume pot and not on the other side of the input jacks. One could mount something else in the unused hole or leave a non-functional switch in there as a mystery "tone" boost for the discerning ear. I'm sure someone will hear a difference when it is engaged .

Below are some pictures of my amp. Which does function fine without any oscillations. I have made these modifications to the circuit to make it a pretty close match to the kelley amp documented here: http://ampgarage.com/forum/files/kelly90schemlayout.zip

- Add brite switch (right next to the volume pot)
- Shield run from volume to stage 2 grid (kelley does not have this but the run is long enough that it is a good idea)
- Take out/don't use cathode follower
- 1MA treble pot
- 1M bass pot
- Remove grid stopper on stage 2 (R34) 1k (did not notice any difference in "tone")
- cathode bypass cap on stage 2 to 22uf
- PI tail resistor (R25) 22k -> 10k
- LarMar (type II) master volume

After these changes my circuit is more or less identical to the kelley with the exception of the power supply and grid stopper on V1. The changes that made the biggest difference were 1) Take out the cathode follower (if it has been wired up). 2) Replace the two pots in the tone stack. 3) PI tail resistor. Also, using a type II master volume means you can get rid of those extra caps.

Finally a few pictures of mine along with a schematic for the LarMar/Type II master volume.

Very cool - thank you for the input. It's gonna take me a bit to get through the meat of your post, but I definitely get what you're saying. Unfortunately, this got bumped from my bench while I take care of another amp that came in on a higher priority, but that bright switch definitely seems to be an issue. I'm in the process of re-routing and trying to get the noise down.

Thanks!