Traynor amp hybrid.

[Nixon101]

Would I have any problems possibly mounting a tube socket for the driver tube in-between the transformers? This particular amp I have is just after they quit putting chokes in the amp so I have mounting holes for a little board and I could possibly enlarge the one whole to mount a tube socket. 1... is it just too far away from the phase inverter and power tube, as in, trying to keep the leads short and away from other stuff. 2.... would the transformers themselves induce nose in the signal path? Before I do anything permanent, I'm going to rig it up and try it out to see if I like it first. As big as this chassis is, there's not a lot of real estate left inside other than where I put the existing bias supply, which would be replaced anyway.

[Nixon101]

Well I made a board with the bipolar power supply, and kind of rigged up a temporary tube socket to experiment with. Hopefully back at it tomorrow.

[Nixon101]

So I have this bipolar supply and cathode follower tube temporarily jumpers in, everything except the actual audio path from the phase inverter, and to the power tubes. Just trying to get my voltages right. What I'm seeing is -41v and -40.5v on the cathodes and -37.5v on the grids, so about 3 to 3.5v more positive than cathodes. I'm pretty sure that sounds about right. I just need to adjust the bias resistors to get the pot more in range, maybe swap the 10k for a 20k, but it's there. Now after all the schematics and layouts I looked at, what I followed was the layout manning posted where the 470k grid leaks are coming off the wiper of the bias pot. (I just altered the values but the topology is the same). Now in other sss schematics the bias pot is wired in differently. Are there any pros and cons to each style of bias pot wiring? And ya i kind of messed up the drawing but the 1 megs are to the wiper off the pot.

[martin manning]

This is the bipolar supply powered CF I have for the SSS 002. The bias voltage is aimed at 6L6's, so you might have to adjust the 75k. The grids should be below the cathodes as usual.

[Nixon101]

I see. My grids are a few volts more positive than my cathodes, rather than being more negative. I missed that. Im not sure why but ill rewire my bias pot and see if that gets fixed. I have the 100k's and 1 meg's the same as yours, but what I'm wondering is the other resistors.... are the values really critical, or is the ratio between the values what really counts? I ended up with 51k/ 15k with the 10k pot rather than your 75k/24k which are roughly the same ratio but would I be better off upping the values?

[martin manning]

The values aren't critical, but the current should be kept down to a couple of mA. I like to wire the bias pot as shown because if the wiper contact fails it goes to max negative voltage.

[Nixon101]

The values aren't critical, but the current should be kept down to a couple of mA. I like to wire the bias pot as shown because if the wiper contact fails it goes to max negative voltage.

Ya that's what I thought about the bias pot, but I followed that layout anyway just to see. Anyway, I rewired it like you said and I'm still getting less negative voltage on the grid than the cathode. Is this because I don't have it hooked up in circuit, or am I really getting something wrong here?

[martin manning]

That's not making sense to me. The only other parts that will affect the CF bias points are the 470k's to ground at the outputs.

[Nixon101]

That's not making sense to me. The only other parts that will affect the CF bias points are the 470k's to ground at the outputs.

Yup. I rewired the bias pot like you suggested. I did play around with the resistors. I had 15k/51k...... tried I think 24k/91k.....and something like 220k/860k. All did the same thing. Here's a voltage chart and schematic. Still not connected from phase inverter/to power tubes grid.

Edit: the only other connections I have are the ground from the bias divider over to the center ground point of the supply board, which is jumpered to chassis. And the 60vac winding jumpered to the supply board.

Also I tried another 12ax7. Same thing.

[Nixon101]

I should also ask.... when I flip the power switch on, I have full voltage on the plate and cathode is seeing full -170v for about 10 or 15 seconds until the heaters warm up. Would this cause damage to the driver tube or for some reason the power tubes? Also, I really appreciate everyone who has chimed in to help, especially you manning. You seem to be one of the more active members and always giving advice and helping others. I hope one day to have as much knowledge as you. Thank you!

[martin manning]

I should also ask.... when I flip the power switch on, I have full voltage on the plate and cathode is seeing full -170v for about 10 or 15 seconds until the heaters warm up. Would this cause damage to the driver tube or for some reason the power tubes?

The power tubes shouldn't have any trouble with that voltage on the grids. The 470k to ground will divide the Vh-k at start-up down, but I think I would reduce those to 270k to bring it nearer to 120V. That should be fine for most 12AX7's. See if the voltages line up better with the CF connected into the circuit. It's possible there is some oscillation happening.

Interesting experiment here, happy to help out.

[Nixon101]

I haven't had time to experiment today but just throwing some ideas out there.... is it possible to run the plate and cathode off the bipolar supply, while biasing the grids of the driver tube from the existing bias supply? Why I started thinking this, was because for some reason I was getting positive bias on the grid conpared to cathode, maybe thinking that decoupling the negative supply for the grid, from the negative supply for the cathodes might correct whatever is wrong. I highly doubt it but who knows. I'm going to try that and see what it does. The second reason I thought this was that I could have heater winding and bias winding start with ac on switch, then possibly the +170/-170vdc supply hooked to the standby switch so it doesn't get hit with voltage on the plate and cathode before the filaments are warm. Now I have an issue with how my standby switch is wired. It's a dpdt with two lamps wired to one side, and the ground side of the full wave bridge wired to the other. Which when the ac switched on but standby off, sees -330v. I was thinking of just hooking the ground of the bipolar supply to the same switched contact as the bridges negative side but I'm sure I can't do that seeing voltage on it, thus entering the bipolar supply and probably blowing up some caps and diodes. Can the ground switch be repurposed to switch the bipolar supply on/off? This is alot of dicking around that I'm not going to do, but just throwing ideas out there for sake of learning.

[Nixon101]

That's not making sense to me. The only other parts that will affect the CF bias points are the 470k's to ground at the outputs.

So I just found out, when I take the two measurements independently from ground, my grids are 3.5 volts more positive than cathode. Now when I take the measurement with common on the cathode the grid measures -0.3v. Any ideas why?

[martin manning]

I haven't had time to experiment today but just throwing some ideas out there.... is it possible to run the plate and cathode off the bipolar supply, while biasing the grids of the driver tube from the existing bias supply?

Yes, you could.

Why I started thinking this, was because for some reason I was getting positive bias on the grid conpared to cathode, maybe thinking that decoupling the negative supply for the grid, from the negative supply for the cathodes might correct whatever is wrong. I highly doubt it but who knows.

I highly doubt it too.

The second reason I thought this was that I could have heater winding and bias winding start with ac on switch, then possibly the +170/-170vdc supply hooked to the standby switch so it doesn't get hit with voltage on the plate and cathode before the filaments are warm. Now I have an issue with how my standby switch is wired. It's a dpdt with two lamps wired to one side, and the ground side of the full wave bridge wired to the other. Which when the ac switched on but standby off, sees -330v. I was thinking of just hooking the ground of the bipolar supply to the same switched contact as the bridges negative side but I'm sure I can't do that seeing voltage on it, thus entering the bipolar supply and probably blowing up some caps and diodes. Can the ground switch be repurposed to switch the bipolar supply on/off?

The bias supply should be up and running before you switch the high voltage on. This CF is now the source of negative bias for the power tubes, so it needs to be stable before you take it off standby. I would just lower that 470k's to 220k as suggested above.

[Nixon101]

Now I'm having some more weird stuff happen. I separated the negative supply from the grid and biased it from the existing bias supply built in the amp, through the 1meg's. Same result. Then thought, wait I don't need these 1meg grid leaks (or do i) and now everything is reading all good!!!! I have around -1.5v negative bias on the grids! I have no idea why one way didn't work and now this is working. Maybe separating the negative cathode supply from the negative grid bias supply might help with tremolo. I would think not fluctuating the cathode supply but only the grid bias would be better.