6F12P-6Ф12П passive RIAA

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
During the search of proper tube for building RIAA, I came across a site introducing a triode-pentode tube called 6F12P (6Ф12П) which has very high mu and high Gm for each section. So I purchased a few from ebay. Last week I read the article of Kevin discussing his future MICRO project for 6F12P, thinking it is a good time to try this unfamiliar Russian tube.
After some rough calculation and reference from internet, I drew a schematics as attachment.
Triode section is used for input stage, then direct-coupled to pentode section using pentode connection. Then 5:1 opt out.

It is really simple, so I use a versatile PCB hard-wiring a circuit. As photo.

I don't have much experience about small signal pentode, only hoping all is alright. After power up, the working points seem OK, but each tube had different gain, so I just kept swapping 6F12P to get two channels balanced.

And the sound is pretty good! Although a little bit hum, but this is a good stuff.

I'll manage to build another with different layout.

cheers

Coffin
 

Attachments

  • 6F12P-RIAA.jpg
    6F12P-RIAA.jpg
    119.6 KB · Views: 965
  • BOARDASSEMBLE.jpg
    BOARDASSEMBLE.jpg
    162.7 KB · Views: 876
Administrator
Joined 2004
Paid Member
Not sure what you mean by flickering, do you mean that you saw arcing during warm up, or did the filament "flash" when the power was applied? Can you describe it in a bit more detail.

FWIW I have a design on paper which I will start to work on shortly. In my case the pentode is used as the input in pentode connection and the triode is used as the output. More details will follow in the thread once I start to build a prototype.
 
Administrator
Joined 2004
Paid Member
Yes, it is internal arcing, and it is an indication that the voltages are too high between one electrode and another when the filaments are cold. I would suspect the pentode grid to cathode as one possibility, another very likely scenario is suppressor grid to plate. This will destroy the tube in short order. I suspect this phono stage is powered by an unregulated solid state supply?
 
I found that one of the 6F12P flickered during powering up. Then I check other tubes to find that 3 out of ten will produce this flicker. I have biased the heater at about 50V, which is between the cathode voltages of two section.

weird.


6F12P flicker - YouTube

hi coffin, it's been a long time....
i am not even sure that you need to lift the heaters at all, since you are not using direct coupling between stages...perhaps just grounding the heater centertaps is good....i could be wrong though, but i think it is worth a try...

OT, do you still sell generic polypropylene caps?
 
Administrator
Joined 2004
Paid Member
Umm, it is direct coupled, have another look.. :D

Not a bad idea IMO, but need to watch what is going on with the power supply while the tubes are warming up.. FULL plate voltage is applied to the pentode grid and the spacing between grid and cathode is tiny.. This is one of the few cases where I would recommend a clamp diode from grid to cathode that conducts any time the grid is more positive than the cathode.

I am also very concerned about the large electrolytic on the cathode of the pentode section, it could be a lot smaller and would help.
 
kevinkr said:
Yes, it is internal arcing, and it is an indication that the voltages are too high between one electrode and another when the filaments are cold. I would suspect the pentode grid to cathode as one possibility, another very likely scenario is suppressor grid to plate. This will destroy the tube in short order. I suspect this phono stage is powered by an unregulated solid state supply?

Kevin, That was something I didn't notice. Maybe I can use 6X4 or EZ81 as slow turn-up.
I'm using MOSFET regulator based on Audio Research design.


AJT said:
hi coffin, it's been a long time....
i am not even sure that you need to lift the heaters at all, since you are not using direct coupling between stages...perhaps just grounding the heater centertaps is good....i could be wrong though, but i think it is worth a try...

OT, do you still sell generic polypropylene caps?

AJT,
It is direct-coupled. So elevating the bias voltage of heater is necessary.
And yes I still have some caps on hand…..


kevinkr said:
Not a bad idea IMO, but need to watch what is going on with the power supply while the tubes are warming up.. FULL plate voltage is applied to the pentode grid and the spacing between grid and cathode is tiny.. This is one of the few cases where I would recommend a clamp diode from grid to cathode that conducts any time the grid is more positive than the cathode.

I am also very concerned about the large electrolytic on the cathode of the pentode section, it could be a lot smaller and would help.

Yes! Clampping is a great idea! Thanks.
About the 330u/100V cathode by-pass cap, what I know is when using triode, the real Rk is [(RL+ra)/(u+1)]//RK], but I don't really know how pentode works, so I use the same concept…..please correct me if I'm wrong.


AJT said:
in that case a neon NE2 from grid to cathode of the pentode might help,
or a diode may be tried....

agree on the cap, more like 3.3uf is more than enough, so that film caps can be used instead...

I've never used NE2, is it similar to zener diode?
Or can I use a zener? or a P-JFET as diode? or just simply a 1N4148?


Thank!!

Coffin
 
Administrator
Joined 2004
Paid Member
1N4004 has too much non-linear capacitance (and some leakage current as well) when reversed biased imo, that's why I recommended the 1N4148.. Voltage rating should be a non issue when forward biased, current will be limited by the high resistance in the EQ network, and reverse biased with tube conducting it will see <2V across it.
 
Administrator
Joined 2004
Paid Member
On a high impedance node with very low signal levels and a lot of gain to follow? Pehaps some 1N400x are better than others in regards to their parasitics..

I use NE2 to clamp cathodes in long tailed pairs to <90V, this might still be too high a potential grid to cathode, I believe the limit is somewhere around +/-30V.
 
Administrator
Joined 2004
Paid Member
Yes the EQ string of resistors and the triode plate load are the source, and it will rise quickly to the supply rail while that 330uF cap holds the pentode cathode close to ground. It is the cathode bias in the pentode that will cut off the diode by rising above the voltage on the grid. Make sure that you orient the diode so that the cathode (banded end) is connected to the cathode of the pentode.

I think you might want to use a smaller cathode bypass cap in the second stage. I should think 47uF would be adequate and would reduce stress on the tube as it is warming up.

Consider a LED for biasing the first stage, they come in all sorts of forward voltages, IR leds typically from about 1 - 1.3V, and red leds from 1.7 - 2V.. Other colors offer higher vf, but also higher dynamic impedances in most cases. Very large electrolytics with very little voltage across them don't seem to perform well with audio.
 
Administrator
Joined 2004
Paid Member
There is no classic grid resistor on the pentode per se as it is direct coupled to the plate of the front end triode through the EQ network resistors, there however is a grid stopper resistor of 150 ohms which should be fine. I tend to overkill, but in noise sensitive locations am using values as low as 47 ohms to reduce johnson noise in low level circuitry. The key is close to the pins and sufficient in value to isolate external reactances from the pin. (And check for oscillation of course.)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.