Loud and (somewhat) clean amp

Overdrive Special, Steel String Singer, Dumbleland, Odyssey, Winterland, etc. -
Members Only

Moderators: pompeiisneaks, Colossal

Post Reply
mobisimo
Posts: 71
Joined: Wed May 11, 2005 12:19 pm
Location: Massachusetts

Loud and (somewhat) clean amp

Post by mobisimo »

Thus endeth the laborious first chapter in my twin overhaul saga. Some background: I bought this amp—a 1980-82 master volume ultra linear twin reverb—used probably ten years ago. It was my first real tube amp! After a year or so I built an 18 watt (with the help of some great folks from that forum) and the twin started collecting dust. It really was a clean machine—did not distort and shook the house—but I was sick of buying overdrive pedals that I ended up not liking ($$$!!!) so I settled on amp overdrive and am still happy about it. About a year ago I opened up the twin and took out some of the plumbing (master volume, boost switch), changed pi values to more at7 friendly ones, and converted the matching circuit to a real bias adjust but the thing just sat around after that. All this recent talk of loud, clean amps on the forum got me to start pondering the twin again, so I opened it back up.
First order of business was to install a post-pi cathode follower. There was much informative conversation on this topic and I am grateful to all of the participants. I took the fairly easy route and jacked the super twin version—12at7/470k/2k2/33k—because it involved the fewest parts and I had some old at7s lying around. I removed the tremolo circuit because I never used it; I also removed the other channel for the same reason. That created some space inside the chassis so I moved the filter caps into the chassis. Made a couple mounts and moved the rectifier board to the dmble-approved location (it’s still a bridge rectifier). The preamp conforms to the third ‘normal channel’ mod from blueguitar. This all amounts to some fairly conventional and sensible modifications and isn’t much to get worked up about, so here’s the fun part: this is mostly done with parts from the twin, and the rest with leftovers. I had only two eyelets lying around and no turrets so the pi/cf layout is a little bootleg and the boards are genuine hack-a-whack. I also took out the brass plate and star grounded the whole thing. (Chipping out the old chassis grounds with a chisel and big @ss hammer was therapeutic.) However, having taken that route I can say almost certainly that I will NEVER AGAIN hack up an amp like this. The mods themselves would have taken about 3 seconds had I made myself a new set of boards. Did I save a couple bucks? Yes. Would mowing a couple lawns instead have made up for the cost and taken substantially less time? Yes—definitely.
The amp is functional now but only consists of a preamp, phase inverter, post-pi cf, output stage, and power supply. With an ax7 in the pre and the pi and the at7 it did distort on first fire-up. I changed those to old stock 5751, ax7 and at7, respectfully, right before a band practice and it made a big difference. I also went from the 1k/4k7 power supply string to 2k2/22k/2k2 and that seemed to make a positive difference (nothing connected at the 22k-2k2 node). I’m looking to add reverb sometime soon according to heisthl’s post found here: https://tubeamparchive.com/viewtopic.php?t=4463 about halfway down the page, and will post results; might change the power supply string around again when the reverb goes in—any thoughts? I’m still using the ul taps on the transformer but installed switchable 820/100/225uF global NFB. Choke is still the original 2k7 resistor. Was using the original 30-year-old power tubes (!) because I broke one of my SEDs, but I recently acquired a set of old Audio Classic 6L6s. If there were ever proof that the world is unjust, it’s that you can break a single tube of a matched quad. I’m thinking these three things are contributing to the grit the amp produced on first fire-up, although the brief break-in and 5751 in V1 tamed that. I’m still tossing around the idea of throwing in some step filters although it would require more planning and plumbing.
How it sounds now:
The thing smoothed out over the few hours since it came online again. I was pleased during practice that the distortion seemed to iron itself out and it gave way to a much sweeter sound by the end. The pops have disappeared and it is possible to get a bit of controllable feedback while the sound is still convincingly clean. I expected the thing to be a bit louder, although I could drown out my buddy’s 1x15 ampeg bass combo without much trouble so maybe I’m being unrealistic here. I’ll post pics within a day or so of its current incarnation; I’ll try to get some clips up as well but I’ll be relying on a cheap computer mic.
Thanks all for reading!
mobisimo
Posts: 71
Joined: Wed May 11, 2005 12:19 pm
Location: Massachusetts

Re: Loud and (somewhat) clean amp

Post by mobisimo »

As promised...
You do not have the required permissions to view the files attached to this post.
mobisimo
Posts: 71
Joined: Wed May 11, 2005 12:19 pm
Location: Massachusetts

Re: Loud and (somewhat) clean amp

Post by mobisimo »

Internals
You do not have the required permissions to view the files attached to this post.
mobisimo
Posts: 71
Joined: Wed May 11, 2005 12:19 pm
Location: Massachusetts

Re: Loud and (somewhat) clean amp

Post by mobisimo »

Last batch of pics. I drew up this layout for the ppicf after some hemming and hawing--hope other folks can get some mileage out of it also, as it is battle-tested. I'd really appreciate any comments thus far.

Thanks to everyone on the forum for their helpful contributions--especially Brandon and Chris for responding to really particular questions.
You do not have the required permissions to view the files attached to this post.
mobisimo
Posts: 71
Joined: Wed May 11, 2005 12:19 pm
Location: Massachusetts

Re: Loud and (somewhat) clean amp

Post by mobisimo »

Hey guys,

I'm installing the reverb on this amp and will post pics when it's done. I've got a quick question meanwhile though. I'd like to experiment with feeding the 6l6 screens from B+ rather than the stock ul configuration; however, this power supply would put about 500v on the screens as it is now from B+2. The way I see it, there are a few ways to lower the screen voltage: 1) increase the choke resistor or add another resistor in series; 2) add a parallel node post-choke with some series resistance and another filter cap, as seen for instance in creating a supply voltage for some reverb circuits; 3) install an OC3 regulator to bring the voltage down ~100v; or 4) create a voltage divider at B+2 a la bleeder resistors on B+1, but mismatch the resistors to achieve the desired voltage, say 1k and 10k.

Solution #4 is probably the most obscure. Tony Albany had great success IMHO with the OC3 in his KT-88 non-HRM (https://tubeamparchive.com/viewtopic.ph ... sc&start=0), however in a very different sort of amp. I realize that the ul method might very well be the best option but I'm just looking to have a little fun before I tie a bow around this project.
Post Reply