Russian rod tubes in a battery all-tube mini amplifier prototype

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, the lower one. I thought it was the hotter bias, but that makes sence. I will try a 300-200 ohm resistor and see how it changes.

Edit: Changed the initial stages bypass resistor to 150 and 300 at the lower power tube. That worked, now both OP tubes drop 1.8v, while the preamp tubes are dropping 1.3v at the heaters. The 22R resistor is dropping 0.85v, resulting in 38.6mA.
 
Last edited:
Where did you get the finals' operating points from? Did you chose/calculate them by yourself? Are the tubes working in class A or AB orB? Anyway, I'd follow Hearinspace's suggestion to tentatively operate each tube's filament separately, e.g. from NiCd cells, adjust the operating points, measure the plate and screen currents, and finally design the heater circuitry with all the required shunt resistors.


Best regards!
 
I had to define some priorities.
Series heating was one of them.
The second was try to run it in class AB.
And the impedance of the OT was also one point.
This all running from 2 Li-Ion batteries, and a SMPS for the high voltage.

With that in mind, I tried to achieve the bias at the power tubes by positioning them at the the higher side of the series string and playing with the position of the grid leak resistor.
I think my schematic was wrong, because I changed the position of the grid leak of the higher tube to achieve a similar bias. From grid to the negative side of the filament I have -3.67v at the higher tube and -3.48v at the lower. (this was before the addition of the bypass resistor, now I need to check it again). The datasheet only has curves for the screen at 45v, so I used paintkip, traced the pentode curves (not as accurate as a good model) and increased the screen. I used the obtained curves as an estimation of how it would look like. I choose the coldest setting (lower power tube leak resistor to ground).

For the preamp tubes there was just a little negative bias available, so I used it and reduced the plate resistor to come closer to center bias.

Something that must be accounted for is the discharging of the batteries. (don't want another SMPS in there). This means, that the bias changes with battery life. Mostly due to the SMPS voltage. Fresh batteries gave me 100V after the SMPS, now I'm at 90v and batterias adding up 7v. At 6v, the lowest charge it will get worse, but I can live with it, since at this point it's time to recharge anyway.
 
Yes, I think I'm well aware of your design goals. Series heating even of filament type tubes is possible, even biasing the power tubes by voltages that occur along the heater chain. But it's much more delicate than for indirectly heated tubes. Think of bypass resistors and capacitors etc., because the filament next to ground (most possibly that of your first and most sensitive tube) carries all the plate currents of the upcoming (power!) tubes.



Best regards!
 
Member
Joined 2008
Paid Member
Yes, I wasn't criticizing your approach. (I think it's a cool project!) My suggestion to power the tubes individually wasn't about your amp design. It was about establishing the tubes' real life operating characteristics (as opposed to data sheet or sim), giving a firm base for figuring out why the voltages were where they were. Nothing more than that.
 
I tested it with pots in the place of the bypass resistors and adjusted them to have the correct voltage drop at each filament.
That worked nicely, it even increased the output a little (I think now I'm at 0.3W, with 2.41v at 8 ohms, measured with my oscilloscope and a sine wave).

I ended up with 3 different resistors, because the PI was also forcing some extra current through the gain stages. It also helped with the oscillations!

I still need to build a test gig for testing each tube separately. Guess I could use small protoboards...
 
While working on a layout I procrastinated a little bit with some DM160 tubes:

lqnGJcll.jpg


even made a video>
YouTube
just to show it blinking...

maybe I should try this amp in parallel SE too, to see how louder the PP is.
 
Hi, going back to the first amp, here some pictures of the current board:
TOXWjDNl.jpg

ENFXj6ql.jpg

and the used schematic:
Wj1bmD2.png


The buzz went away with the board, but now I had some squeal with everything on full. I solved it by bypassing part of the signal at the OT using the 470pF capacitor. I tried it at different ano resistors, but the resistor at the transformer was the only one that really solved the problem. It still feedbacks, what is kind of nice.

I could also remove the cap bypassing the 330k resistor after the second stage. But I think I fixed it in some way.

Something interesting is that with the volume maxed I can't obtain clean tones, but the distorted tone is really fuzzy, and sounds really good. Checking with the oscilloscope showed that after a certain level the PI distorts too. This unbalances the PI and gives the nice fuzz/high gain/heavy metal tone from the late 70s. Instead of solving this I think I'm going to put some indication at the faceplate for the last 30 degrees turn of the volume pot.

When I reduce it gets the normal sound distorted sound, which can be cleaned with the gain control or even with the guitar volume control.

I'm not sure if I liked the layout with the tubes parallel to the board, I made a second layout with the tubes perpendicular to the PCB, but the anode at the top is always a problem. I couldn't find a ceramic cap for those tubes, like some bigger tubes have, to hide the anode.
I think the board will be standing inside the box, so the tubes are at least oriented vertically to the box, and hide the PCB behind an aluminium plate, like in the orange amps, but with holes where the tubes are located.

Cheers,
Thomas
 
I am severely infected!
Yesterday I've got a bunch of these subminiature 1J24-B tubes for little money, but I wasn't able to acquire the 1P29-B's. Instead, I've opted for 3A4 power tubes (by CIFTE, €5 each). Surely this won't end in a microwatt amplifier like your's, rather in the 1 watts area, as these 3A4's really are the big blocks among the miniature battery tubes.
I am planning to provide three switchers as PSU's, one for the heaters in parallel (1.4V @ 0.5A), one for the plates (150V @ 30 mA) and one for the power tube's bias voltage (-10V @ 1 mA approx.).
Did you manage to get your hands on a valid datasheet for the 1J24-B's?


Best regards!
 
This was the best I could fin:
https://ssli.ebayimg.com/images/g/nkoAAOSwVm5Y9hwC/s-l640.jpg

But there is a whole page over Radiomuseum. Russian Subminiature Tubes

I'm interested on how the switchers will work toguether. I was affraid that using another one for the heaters would make them heterodyne and introduce some noise.
This happens when using a 12v SMPS with a 12v to 200v SMPS.

Will it be battery powered? I think there is a MAX CI that converts anything between 4v and 3v to steady 5V.
 
And the final result>
socVZS0l.jpg

SWbRg66l.jpg

9K7Atyul.jpg

ElsDwb9l.jpg


It has a small squeal, almost inaudible. I need to check that. Could be some bad lead dress or even a microphonic tube. Maybe I should cut some high end along the stages, since these tubes work well in the high frequency domain.

The box was an old ammeter. The faceplate now covers where the meter once was. I still need to rewind the meter to bring it back to life. For the time being the box is being used for something else. Really liked the grill made of a perforated metal sheet with a cross pattern. I made the grid to protect the tubes. I wanted to show the tubes, but at the same time keep them protected, so the metal grid helped.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.