Troubleshooting 2204 clone oscillation

[kenrinc]

I posted previously regarding this 2204 clone I did using Jason Tong's PCB. Got some excellent help from many members who helped me get it up and running. After playing it for a month or so I've had more time to play around and look at all of it's idiosyncrasies. First off, it sounds good. Really good actually. I'm at a point now where I'm just trying to get everything working perfectly. There are a few issues that are "interesting" but not deal breakers and I'll eventually sort that out but right now I'm working on one that has me stumped.

I'm trying to resolve an oscillation problem in what I think is the output but not entirely sure. Basically with the amp on (out of standby) and the guitar plugged in but volume off, if I turn the MV past 6 on the amp goes into oscillation. Actually the guitar doesn't have to be plugged in. If I crank the MV and lower the Gain I can hear what sounds like a blocking distortion with the MV all the way up. You can't actually play with that tone since even though you get output it's in between oscillation and feedback. Troubleshooting is made more difficult by the fact that I can't get near the amp with the guitar on. It just screams. If I'm 10ft away on the couch, no problem. Bias is currently set at 35ma @ 490v. New JJ E34Ls.

Ken-

 

[Guitar-Rocker]

I built a 14watt 2203 version of Jason PCB. Having built a few hundred amps, I still found the board a bit challenging at times. Can you post photos of your build? You may have something that either needs a jumper or a wire that is in the wrong position.

 

[StingRay85]

Are you sure you have your PI coupling caps connected to the right EL34 and side of the OT? If not, you get global positive feedback and you might get the result you are described.

 

[dtier]

If you have not verified the global feedback phasing you should. Its typically done with a signal generator and scope but I suppose it could be done by ear. It should be louder with feedback disconnected, not quieter. You should be able to play next to the chassis without it going crazy, just noisy. Do not reach into the chassis while holding the strings!

 

[kenrinc]

Thanks guys. Appreciate it.

I've verified that the coupling caps connect to the correct output tubes but I've never been fully sure that the OT primaries were correct and that's just because the Hammond documentation for the 1750N is not really helpful. The wire colors definitely don't match so I just looked at the transformer schem and wired it as the 2204 schematic shows. I'm fully aware what swapping them does as the amp was setup that way when I built it and all it did was howl the first time I powered it up. But if someone wants to confirm the output wiring of the 1750N I'm all ears. Currently Red goes to V4 and Blue goes to V5.

I do have a scope. It will take some time to get it working though.

Ken-

 

[mickeydg5]

Swap the primaries to figure out what is what.

 

[kenrinc]

Swaped the primaries and get same problem only worse. Starts early on the MV dial, around 3 instead of 6. On the schematic it shows the bottom side of the OT primary going to V5 and the top going to V4. With the Hammond it's opposite. I'll take a look at the rest of the NFB stuff tomorrow.

Ken-

 

[AndyD]

Bias is currently set at 35ma @ 490v. New JJ E34Ls.

490v b+ on a 50w amp seems very high.

 

[playloud]

Lead dress could be improved. In particular, take a look at the signal feed to your power tubes (i.e. the green wires going to pin5 of each). Any interaction with the OT secondaries or orange NFB wire is a potential source of oscillation.

A scope will help you isolate the problem to a specific part of the circuit.

 

[william vogel]

The two green wires going to the power tubes grids need to be twisted and run further away from the output transformer wires.

 

[kenrinc]

Thanks. I've tried all the suggestions. I put the amp up on the bench and played through it while chopsticking around. The oscillations don't go away when moving around the obvious signal wires. OT sounds like a buzzer once past 4 on the MV dial. All take a break and think about this. I can scope it but It may just come down to pulling the board and rewiring everything to the board anyway. The transformers never had long pigtails so I was sort of forced into doing what I had to. I do have an old HP scope out in the shop but not sure I have all the needed connectors. I'll see if I can get it and my audio generator up and working. What should check first?

PS: Plate voltage is 472v.

Ken-

 

[kenrinc]

Loooked at it again tonight. Scope isn't working. Separate project. I pulled the 1st tube, powered up and the problem isn't there so I'm guessing it's in the first stage area? I've swapped the 1st tube with multiple known good tubes; makes no difference. Also tried diming the master and then bringing up the preamp. With the master on full I can get the preamp to about 2 and then same problem so it's obviously in the preamp. I've got a 50pf cap between plate and cathode on V1B. That may be too high. Again, moving wires around in the output section doesn't seem to do much of anything to the noise. I'm beginning to suspect that the tube sockets I have in the preamp section are either faulty or at least contributing to the issue. When swapping tubes I noticed that the pins are very loose and I don't see any way to tension them. There is quite a bit of noise when poking at the bases from underneath. Currently have new Mullards in the first 2 stages and it sounds great just as long as I keep the pre and master at around 7 and 4 beyond that, it's a high frequency screamer and then a motorboat.

Ken

 

[kenrinc]

Tonight I put a 50pf cap across the plate and grid of V1B and BINGO, I can get the MV to about 9 before oscillation so it's definitely at V1. Both grids are using shielded cable grounded at the jack and gain pot so I'm sort of miffed at that..... onward. This was actually the only gripe I had with the amp in terms of tone. When I would rub my fingers over the strings it could damn near take your head off.

Ken-

 

[dtier]

Just noticed your using non-stock values in the cathode circuits for V1a,V1b. I can't read the values but it looks like your adding a lot of gain there with your bypass caps, which is likely why its freaking out. It would probably be ok with stock values but will take some experimenting to get it under control as is. I had similar problems with a Bogner Ecstacy circuit which is similarly hot in the front end. I think your on the right track.

 

[kenrinc]

I thought the cap across the plate and grid was gonna be the key but it cuts too much out of the tone of the amp. The front end is just the JEL mod (at least it is right now). It's like 5 parts total. Nothing special and it sounds great until about 5 on the MV and then it just screams and once your at 10 it's just a motorboat. It wouldn't matter most of the time but there are some artifacts that show up during muted riffing that bother me. That is why I'm try to get to the core of the problem. I've replaced all the RU174 shielded cable with new shorter runs thinking the old cable might have had an issue. I've traced and checked all the joints at V1 thru V2. Replaced V1 and V2 with brand new tubes.

I've decided to do what I didn't want to do and that was pull the board. Truly a pain in the ass. But it's done. I will rewire the whole chassis from scratch. One question though, this is an aluminum JMP style small chassis. Is the aluminum chassis a good choice? Wondering if it makes any difference.

Ken-

 

[StingRay85]

Maybe the wiring of the socket?

 

[playloud]

Loooked at it again tonight. Scope isn't working. Separate project. I pulled the 1st tube, powered up and the problem isn't there so I'm guessing it's in the first stage area?

Tonight I put a 50pf cap across the plate and grid of V1B and BINGO, I can get the MV to about 9 before oscillation so it's definitely at V1. Both grids are using shielded cable grounded at the jack and gain pot so I'm sort of miffed at that..... onward. This was actually the only gripe I had with the amp in terms of tone. When I would rub my fingers over the strings it could damn near take your head off.

Ken-

The last part is good troubleshooting, but it doesn't mean the oscillation is "at V1" necessarily. Whatever leaves the circuit at V1 gets amplified by subsequent stages and, with a sufficiently high MV setting, could be getting above a certain "threshold" later in the circuit to induce oscillation. Your 50pf cap is keeping you below this threshold by rolling off high end (which is where oscillation often occurs).

Those mods (particularly bypassing the "cold clipper" at V1b with a 3.3k resistor) definitely increase the likelihood of oscillation. You now have 3 very hot plate-driven gain stages + the CF, so your lead dress has to be pretty good if you want to get away with MV close to 10.

Did you try @william vogel's suggestion above? I would try the obvious stuff before redoing the board. Both of the power tube grid wires currently have ample opportunities for interaction with high-voltage signal (via the NFB wire and OT secondaries respectively)! I would also clean up the orange wire going to the presence pot.

 

[dtier]

I don't see the 2k7 resistor that should be paralleled to the gain pot in your photo for the JEL spec. Please ignore if it's there somewhere.

It's worth the effort redoing your build. you'll appreciate it when done.

 

[kenrinc]

Thanks Dtier, the 2.7ks are for the cathodes of V1a and b. Maybe your talking about the 220k parralled with the gain pot? I actually did not use that because it signifcantly reduced gain but I'll check that out again when I put the board back in. I went back last night with my loupe on and checked every trace, every pad, front and back and cleaned all the flux off the board. I gotta give Jason credit here; I've soldered and re-soldered parts out of this board hundreds of times over the last month of tuning and I cannot believe how well the board has stood up! Almost like a turret board. Before I pulled the board, I double checked my grounds and all of them were good low resistance connections. I did see one questionable ground wire that measured ok but was not fully flowed. It was at the far end of the board. I believe PT ground that will all be rewired.

 

[dtier]

Oops, yes I ment the 220k. Or a higher value. As long as your circuit grounds are all local to their B+ nodes along the chassis I can't think of anything else to add.

Note taken on the Headfirst boards. I have a couple but haven't used them yet.