Hi Luca,
Thank you very much.
BTW, according to Thorsten, diodes, including LEDs, are less linear than any electrolytic capacitor.
Hi Phil,
There is no problem with using the right voltage regulators in phono stages.
There is a problem with VR tubes, since they are noisy.
Context and analysis are a better approach than pronouncements.
Keep the voltage swings required in mind when thinking about linearity in general.
Remember that your PSR for a pentode input is effectively nonexistent.
Good luck.
He may or may not be correct, depending on the situation and the context. Your reduction to a rule is clearly incorrect.
Hello,
using a LED to bias doesn´t solve the stability problems. I´ve tried it. You still will have problems to find matching pairs. The operating point is solid as a rock with the positive g1 voltage and the high value cathode resistor.
I first was a bit afraid about the additional coupling capacitor which is needed between MC step up and g1. Surprisingly, it sounds better with it.
Regarding the VR tube, I only changed one channel to a 0A2. I couldn´t hear any differance regarding noise. But it sounds clearer. Less distortion.
Best regards,
Martin
using a LED to bias doesn´t solve the stability problems. I´ve tried it. You still will have problems to find matching pairs. The operating point is solid as a rock with the positive g1 voltage and the high value cathode resistor.
I first was a bit afraid about the additional coupling capacitor which is needed between MC step up and g1. Surprisingly, it sounds better with it.
Regarding the VR tube, I only changed one channel to a 0A2. I couldn´t hear any differance regarding noise. But it sounds clearer. Less distortion.
Best regards,
Martin
Hi Luca,
It looks nice.
Is the phono stage completely quiet, free of any hum?
Happy New Year.
Hi Martin,
What is stability issue you encountered with the original design? Was it difference in anode currents between the 2 channels, or what?
What capacitor did you put before g1 of the first tube?
Can you describe the sound difference when you added that capacitor?
That regulating the voltage supplied to g2 decreases distortion does make sense. However a maida-like regulator can do the job well. It looks to me better to avoid any noise-producing elements inside a phono stage.
Happy New Year.
Hello,
I have all the irons (transformers, chokes for PSU and EQ) custom made in Italy.
Because sometimes I collaborated with the manufacturer I don't know if it is correct to disclose here who is.
The unit itself is quiet enough (no noticeable hiss) and with little hum.
But since then I have to deal with external MC pre-pre.
Until now I can't make it quiet enough because both SUT and single FET (similar to that proposed here by Salas
) are quite humming.
Best regards
Luca
I have all the irons (transformers, chokes for PSU and EQ) custom made in Italy.
Because sometimes I collaborated with the manufacturer I don't know if it is correct to disclose here who is.
The unit itself is quiet enough (no noticeable hiss) and with little hum.
But since then I have to deal with external MC pre-pre.
Until now I can't make it quiet enough because both SUT and single FET (similar to that proposed here by Salas
) are quite humming.Best regards
Luca
Hello,
Different anode currents resulting in quite different voltages. Also the gain was quite different. Now I can plug almost any E810F resulting in a very stable operation point. AC as well as DC. Give or take 1V.
220nF with a 910K grid leak resistor. In theory a 10nF would be enough but it sounded better with a larger cap. By the way, by using the postive g1 voltage you can increase the grid leak resistor up to 3M5.
I always have problems to explain how things sound. I guess you´ll have to find it out by yourself. I like it
Distortion reduction only worked by connecting the 0A2 to the cathode. There was no change in distortion if connected to ground. There was also some noise.
Happy New Year!
Best regards,
Martin
Different anode currents resulting in quite different voltages. Also the gain was quite different. Now I can plug almost any E810F resulting in a very stable operation point. AC as well as DC. Give or take 1V.
220nF with a 910K grid leak resistor. In theory a 10nF would be enough but it sounded better with a larger cap. By the way, by using the postive g1 voltage you can increase the grid leak resistor up to 3M5.
I always have problems to explain how things sound. I guess you´ll have to find it out by yourself. I like it
Distortion reduction only worked by connecting the 0A2 to the cathode. There was no change in distortion if connected to ground. There was also some noise.
Happy New Year!
Best regards,
Martin
Hi Jushua,
If you go for the LED bias option, use an HLMP 6000 series LED for the bias; these have really nice curves.
The E810F will oscillate in the box, carbon comp stoppers on the grids and ferrites on all pins; the one I'm building now has 2k2 grid stoppers.
Don't worry about the pentode partition noise, E810F is designed for instrumentation use, very very quiet!
As Martin said AC reference g2 to the cathode, I used a 10uF cap, bit like the attached.
If you go for the LED bias option, use an HLMP 6000 series LED for the bias; these have really nice curves.
The E810F will oscillate in the box, carbon comp stoppers on the grids and ferrites on all pins; the one I'm building now has 2k2 grid stoppers.
Don't worry about the pentode partition noise, E810F is designed for instrumentation use, very very quiet!
As Martin said AC reference g2 to the cathode, I used a 10uF cap, bit like the attached.
Attachments
Hi Andrew,
Thank you.
I intend to start with Thorsten deign as it is. Once it will be running and (hopefully) singing, I may try some modifications.
Carbon composite grids stoppers and ferrite beads are obvious, as well as mounting the tubes sockets on some rubber O-rings. (Rubber in a manner of speaking, since plain rubber rings will lose their elasticity under heat).
One thing I may try is about 0.3V bias (not much higher), to maintain as much transconductance as possible and provide some stability. I also may try positive voltage to g1 (with the obvious needed adjustment to the cathode bias). However, should I do that I'll go for bypassed resistor, not nay diode or LED.
I may also try maida-like regulator to g2.
In any case, referencing g2 to the cathode is an excellent idea, in the way of ultra-path.
Thank you.
I intend to start with Thorsten deign as it is. Once it will be running and (hopefully) singing, I may try some modifications.
Carbon composite grids stoppers and ferrite beads are obvious, as well as mounting the tubes sockets on some rubber O-rings. (Rubber in a manner of speaking, since plain rubber rings will lose their elasticity under heat).
One thing I may try is about 0.3V bias (not much higher), to maintain as much transconductance as possible and provide some stability. I also may try positive voltage to g1 (with the obvious needed adjustment to the cathode bias). However, should I do that I'll go for bypassed resistor, not nay diode or LED.
I may also try maida-like regulator to g2.
In any case, referencing g2 to the cathode is an excellent idea, in the way of ultra-path.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.