Traynor amp hybrid.

[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.

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?

Are all the grounds tied back to the chassis?

[martin manning]

By tying the CF grids directly to the negative bias you’re ac grounding them, and that will ground out the signal too. This makes me think there was oscillation happening before. Do yo have an oscilloscope?

[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?

Are all the grounds tied back to the chassis?

Yes grounds all connected. Still experimenting.... thought I had it figured out but I just realized, if I omit the grid leak resistors, the bias is correct, but I'm just hooking both grids together which can't happen. If I run the bias voltage through the grid leaks, I end up with a positive bias. Now I'm back where we were before where you said put it in the circuit to see if it acts correctly. The only difference is I'm running the existing bias supply to the grids instead of from the -170v supply. I noticed there was some debate on whether the extra 470k power tube grid leaks to ground were necessary, considering the 100k's on the cathodes were already grid leaks. (Plus I think those are what you were referring to, that I should add to tame the start up voltage) are these necessary for this cathode follower to operate correctly?

[martin manning]

I think the 470k’s are there to reduce the heater cathode voltage at start-up. They are not needed for the CF to operate.

[Nixon101]

By tying the CF grids directly to the negative bias you’re ac grounding them, and that will ground out the signal too. This makes me think there was oscillation happening before. Do yo have an oscilloscope?

I'm starting to understand now. I'm just reluctant to tear the amp apart to hook this in until I know it works correctly. You think I can just jump some ac signal to the grids to see if it stops? Or will I need to jump the output as well? Also I have no grid stoppers but I have no signal input anyway so I don't think that's the problem. No I don't own an oscilloscope. But I did find a couple old units locally I was thinking about picking one up, but I'm just not at that level yet. This is my first time doing something this extensive.

[martin manning]

Can you add grid stoppers between the 1Megs and grid pins and see if that settles it down? I guess if you don’t have anything connected yet it won’t make any difference …

[Nixon101]

Can you add grid stoppers between the 1Megs and grid pins and see if that settles it down?

I was just thinking that, but then thought wouldn't the grid leaks act like grid stoppers if i am picking up noise from somewhere, since i have no other connection to the grid? They are the only thing connected to the grids, and right on the socket. I'll try it to see. I've been wrong many many times before lol.

[Nixon101]

Grid stoppers didn't help. Tried a 200k to ground with held start up voltage to 116, so that's good. I just seem to be dropping about 4 volts through those 1 meg's. I tried jumpering in some signal to the grids, still didn't help. Any reason the 1 megs are too large a value?

Edit.... just clarifying that taking measurements from ground to cathode, and ground to grid gives me positive 4v bias. Taking measurements from cathode to grid gives me negative .5v bias. Those are the reading I get.

[Nixon101]

http://www.rfcafe.com/references/electr ... 1-1-40.htm

I think I may be on to something here. Very good information.

[martin manning]

I realized (see above) that the grid leaks are essentially in the position of stopper as it stands. It's possible that your probes leads are causing the issue depending upon where you place them. The right answer is the grids should be below the cathodes by ~1.5V. Re the link above, I'm not sure what part of that you are referring to.

[Nixon101]

I realized (see above) that the grid leaks are essentially in the position of stopper as it stands. It's possible that your probes leads are causing the issue depending upon where you place them. The right answer is the grids should be below the cathodes by ~1.5V. Re the link above, I'm not sure what part of that you are referring to.

The information I thought was relevant to me (and I may be wrong) was section 1-33 and 1 -34. Grid leak biasing. Where it explains that a triode wants to self bias through charging of the input capacitor. I'll re-read through it again just to make sure I understand it correctly, but it sounds like I may need it coupled to the stage previous to it. It's my last day of Christmas holidays before I go back to work, so it may be awhile before this project gets done.

[Nixon101]

Alright so this is where I'm at...... I built a board and a bracket to hold the tube socket, didn't like it, didn't fit, had to redo it all. Now I have it all installed, power supply hooked up, bias hooked up, not in circuit I was getting the same readings as before. BUT...... when I jumper the grids over to the coupling caps, the driver grid bias instantly pulled down negative with respect to cathode. About 1.5 to 2 volts negative grid bias. Which is right where I should be. In my totally non expert opinion, I'm thinking the information I was referring to in that article about the triode wanted to self bias through charging and discharging of the coupling capacitor is coming into play here. Could it be that, maybe. Could it be oscillation, also maybe. I don't have the skill or equipment to diagnose it, but I'm just glad it's working. I want to thank you martin manning for all the help. When I get a couple hours to myself, hopefully tomorrow but who knows, I'm going to play the amp how it is, then officially swap the wires over and give it a really good A/B test to see what this really brings to the table. If all goes well, I might start a new thread to break down what I did, and what it adds to an already great sounding amp.

Edit..... no vintage amps were harmed in the making of this. I drilled 0 holes and took out 0 components. Just need to swap wires over and hopefully go.

[martin manning]

... when I jumper the grids over to the coupling caps, the driver grid bias instantly pulled down negative with respect to cathode. About 1.5 to 2 volts negative grid bias. Which is right where I should be. In my totally non expert opinion, I'm thinking the information I was referring to in that article about the triode wanted to self bias through charging and discharging of the coupling capacitor is coming into play here. Could it be that, maybe. Could it be oscillation, also maybe. I don't have the skill or equipment to diagnose it, but I'm just glad it's working.

Ok, great! Looking forward to the results of your A-B listening test. Be sure to minimize the extra wire length added by jumpers when you do that. I'm thinking when you connect it to the coupling cap you are adding a low impedance AC ground (to the power supply), which keeps it from oscillating.

no vintage amps were harmed in the making of this. I drilled 0 holes and took out 0 components.

Excellent!

[Nixon101]

Hey martin (pretty sure it's just you and I here) do you anticipate any problems with my layout? I spent a lot of time trying to figure out where and how to put it all in. I'm still skeptical about mounting the tube inside the chassis, but I got it about as far away from everything else as I could, and have a tube shield to put on too. The pots the tube is close to are, backside view starting from the left..... trem intensity, trem speed, and then reverb. Also the board I made with your bipolar power supply right next to the cathode resistors (red leads), with the grid resistors (yellow leads) a little further away on the lower right, I'm hoping I don't get noise in the signal.

[martin manning]

I think it looks pretty good. That is exactly where I was imagining it might go when an LND150 SF was being discussed. There is very little extra wire length to the power tube grids, and the signal level is high, so less chance of noise from the power supply. If there is a problem, maybe some fast rectifiers or bypass capacitors would help. As far as heat dissipation from the tube, it's only about 2W, so it's not likely to be a problem.

You have the trem to play with yet too, no?