Tweet
I'm working on a project amp going on 4 years now. I finally have it together to where I can get some sound, and while I expected oscillation issues, its hard to figure out where to start sometimes to try to solve them. My build has about 460v B+ on three power stages.
This project amp of mine uses two chassis, with the power amps one, and preamps in another. The power amp chassis was from a 1964 Conn organ, and has one power transformer, and 3 output sections, each with a PI and two output tubes, and their own output transformer. The preamp chassis I made myself and there are 3 preamps in there. I spent a lot of time with the layout and design to try to head off issues, but even so, channel A and B are oscillating, and channel C, the 2204 channel isn't even hooked up all the way yet.
Channel A is an AA864/AB165 Bassman based on that article about converting one to the other thats on the web, but I also have a Marshall style presense control and a single triode Princeton Trem circuit in it. It uses a 6L6GC pair with a 12AT7 PI, and the preamp is a single input, single channel with 3 stages plus the oscillator all using 12AX7's. It sounded fantastic until I hooked up the tone controls the other day...now its an oscillating machine. The OT is a 4000 ohm primary so I should get 50W easy out of this channel. The trem isn't working yet but the presence works fine. NFB is a little less than a stock AA864 bassman.
Channel B uses a 7868 pair with a 5E5 PI, and the preamp is an EF86 into a CF and then the tone stack, volume control, and a 12AY7 gain stage. I built this basic circuit into my brother's Silvertone 1484 that I modded and it sounds great in there. The 1484 is a 6L6 pair and this amp is a 7868 pair though so maybe I'll need to reduce gain? My PI is also different than the Silvertone 6CG7 cathodyne arrangement. In this amp the circuit is very squishy up until about 4, and then starts oscillating. The tone controls don't really work right either, and it sounded great until I hooked them up. This channel has a hi/lo input setup. This channel uses a smaller 8000 ohm OT so I'm guessing power is going to be around 30W here.
Channel C is a JCM800/2204 circuit using an EL34 pair and 12AX7 PI just like the Marshall 2204. This is using a hi/lo series Jfet switching circuit than an engineer friend helped me come up with.....but this channel is not hooked up completely yet. The OT here is a 4000 ohm primary also so I would guess a little more power and a little less breakup than a Marshall, but I'm not sure on that part.
Power amp is all PTP with the exception of a bias tracking circuit I put in from KOC that is on two boards. Each power output section has a screen standby of it's own. It uses a single point buss ground setup and stock Conn layout scheme as far as sockets and transformers arrangement.
Preamp chassis is my own layout similar to Fender's with a board, sockets behind the board, and controls on front, although to allow myself to work on it, front controls are on a plate that screws into preamp chassis and so wire runs to it are a little longer than normal. Grounding here is also single point buss ground with each of the preamp channels having it's own buss ground up to chassis point right next to socket that goes to power amp chassis. Layout is straight line and nothing crosses back on itself and wires that have to cross to do at right angles to each other.
The whole thing has 15 tubes and fits into one head about the size of an SVT head and almost as heavy too. I've got a Weber Fourhead (the last one actually) to allow me to run only one cabinet at a time safely. Should be cool when done but in the meantime its a bitch to troubleshoot. I've incorporated CT biasing on the filament string up to 50v or so to reduce hum. Each stage is seperately decoupled from each other. The isolation between channels is so far quite good but then its not all working yet either. The latest thing that came in along with the tone controls being hooked up was a clicking sound that speeds up if I take tubes out in the preamps and slows down where they're all in there. Its been a good project and I'm learning a lot, but its damned hard to track stuff down sometimes.......I know its a crazy ambitious project too, and I'm probably nuts for trying something like it, but what the hell....
I'll try to post pics this weekend....
Greg
This project amp of mine uses two chassis, with the power amps one, and preamps in another. The power amp chassis was from a 1964 Conn organ, and has one power transformer, and 3 output sections, each with a PI and two output tubes, and their own output transformer. The preamp chassis I made myself and there are 3 preamps in there. I spent a lot of time with the layout and design to try to head off issues, but even so, channel A and B are oscillating, and channel C, the 2204 channel isn't even hooked up all the way yet.
Channel A is an AA864/AB165 Bassman based on that article about converting one to the other thats on the web, but I also have a Marshall style presense control and a single triode Princeton Trem circuit in it. It uses a 6L6GC pair with a 12AT7 PI, and the preamp is a single input, single channel with 3 stages plus the oscillator all using 12AX7's. It sounded fantastic until I hooked up the tone controls the other day...now its an oscillating machine. The OT is a 4000 ohm primary so I should get 50W easy out of this channel. The trem isn't working yet but the presence works fine. NFB is a little less than a stock AA864 bassman.
Channel B uses a 7868 pair with a 5E5 PI, and the preamp is an EF86 into a CF and then the tone stack, volume control, and a 12AY7 gain stage. I built this basic circuit into my brother's Silvertone 1484 that I modded and it sounds great in there. The 1484 is a 6L6 pair and this amp is a 7868 pair though so maybe I'll need to reduce gain? My PI is also different than the Silvertone 6CG7 cathodyne arrangement. In this amp the circuit is very squishy up until about 4, and then starts oscillating. The tone controls don't really work right either, and it sounded great until I hooked them up. This channel has a hi/lo input setup. This channel uses a smaller 8000 ohm OT so I'm guessing power is going to be around 30W here.
Channel C is a JCM800/2204 circuit using an EL34 pair and 12AX7 PI just like the Marshall 2204. This is using a hi/lo series Jfet switching circuit than an engineer friend helped me come up with.....but this channel is not hooked up completely yet. The OT here is a 4000 ohm primary also so I would guess a little more power and a little less breakup than a Marshall, but I'm not sure on that part.
Power amp is all PTP with the exception of a bias tracking circuit I put in from KOC that is on two boards. Each power output section has a screen standby of it's own. It uses a single point buss ground setup and stock Conn layout scheme as far as sockets and transformers arrangement.
Preamp chassis is my own layout similar to Fender's with a board, sockets behind the board, and controls on front, although to allow myself to work on it, front controls are on a plate that screws into preamp chassis and so wire runs to it are a little longer than normal. Grounding here is also single point buss ground with each of the preamp channels having it's own buss ground up to chassis point right next to socket that goes to power amp chassis. Layout is straight line and nothing crosses back on itself and wires that have to cross to do at right angles to each other.
The whole thing has 15 tubes and fits into one head about the size of an SVT head and almost as heavy too. I've got a Weber Fourhead (the last one actually) to allow me to run only one cabinet at a time safely. Should be cool when done but in the meantime its a bitch to troubleshoot. I've incorporated CT biasing on the filament string up to 50v or so to reduce hum. Each stage is seperately decoupled from each other. The isolation between channels is so far quite good but then its not all working yet either. The latest thing that came in along with the tone controls being hooked up was a clicking sound that speeds up if I take tubes out in the preamps and slows down where they're all in there. Its been a good project and I'm learning a lot, but its damned hard to track stuff down sometimes.......I know its a crazy ambitious project too, and I'm probably nuts for trying something like it, but what the hell....
I'll try to post pics this weekend....
Greg
Oscillations caused by capacitive coupling in your layout can be a real PITA. I once spent a whole day chasing one. It turned out the feedback wire from the speaker terminal had got loose and wandered too close to one of the earlier stages.
Comment