|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
198.208.159.19
In Reply to: RE: 26 tube filament ripple posted by andy evans on January 27, 2017 at 08:01:45
Andy, you meant with grid fixed bias you eliminate the cathode capacitor. Right?
Follow Ups:
Andy is referring to filament bias. In filament bias, the bias resistor is connected in series with the filament, and the filament power supply is connected with the positive at the free end of the filament and the negative at the ground of the bias resistor. In this situation the capacitor across the bias resistor can be eliminated.
To power the filament and to provide the bias, a higher filament supply voltage is needed. For instance, in the case of the 26 with a filament voltage of 1.5v and currrent of 1.05A, and say bias of 9v and .007A current, a total of 1.5v + 9v =10.5v and 1.05A + .007A = 1.057A need to be provided. The bias resistor needs to drop 9v so the resistor needs to be 9v/1.057A = 8.5 ohms. Power rating required would be 9v x 1.057A = 9.5 watts times two for safety equals 20 watts.
I have a 26 preamp and with filament bias, it is very quiet. Previous to filament bias, even with a regulated filament supply there was still some hum. So for me, filament bias solved my hum problem and it also eliminated the cathode capacitor which is another plus.
Ben, I do not get it. Float the filament DC power (no reference to ground) and provide +9 volts from one side of filament & negative to the ground bus?Or, are you saying provide 10.5 VDC + to one side of filament and other side of filament gets a dropping resistor to ground. The - from the 10.5 volt filament power supply goes to ground. The dropping resistor dissipates 9 volts.
Edits: 01/27/17
"Or, are you saying provide 10.5 VDC + to one side of filament and other side of filament gets a dropping resistor to ground. The - from the 10.5 volt filament power supply goes to ground. The dropping resistor dissipates 9 volts."
Yes, this is what I meant. The resistor goes to ground along with the supply negative. Sorry if I was not clear.
So the add to the plate resistance is ~9R*(u+1) or less than 100R added to a plate resistance of ~ 8,000 Ohms. Trivial indeed...:)
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.
Sounds like a great idea and no cathode capacitor.
And I did not even use Coleman regulators. I did it on a budget with my own simple diy regulated power supply for each channel comprising a pi filter followed by a LT1085 voltage regulation stage and finished with a LT1085 constant current stage. I used LT1085 because I had them but any of the three pin regulators probably will also work.
Why use any regulator. Use bridge into 6uH choke to 20000uF cap and after a 15 ohm resistor a 200uF cap. This set up makes only .001 AC ripple.
Sure - you can just use chokes in filament bias. Thomas Mayer does just that and he was the guy the popularised filament bias in the audio community. Look at his site Vinylsavor.But I still prefer a combination of Rod's regs and choke input. I don't think you can go too far towards overkill with filament bias. You really need a smooth supply. And Rod's regs are also about impedance matching, which is an added benefit.
As for LT1085 regs, they can be a bit noisy and Rod's regs are much more evolved with better parts. They're pretty cheap as well - really worth using. He's now on the 7th evolution of his regs. See Lyrima.com
The best 26 design currently is probably Ale Moglia's, as shown - with filament bias. See Bartolo Valves website. His Gyrator is really good. And again, like Rod's regs comes on an easy to build PCB. Cheap and well worth using. You can see from the 850mA supply that the 26 is used in "starved filament" mode which can give gains in lower distortion and longer tube life. I've been building 26 preamps since 2009, and I would use the gyrator in preference to an OPT. More detail and generally a more immediate sound.
The 26 circuit is in essence extremely simple. But the gyrator and Rod's filament regs - while a lot more complex than other options in terms of parts count, though as above on handy PCBs - really brings the last ounce of excellence out of the 26. The alternative is the 01A which is almost identical to the circuit shown and which you can find on Ale's site as 01A Gen 2 preamp. That's the one I use myself and Ale prefers it too.
Edits: 01/28/17 01/28/17 01/28/17
I was tired of trying to get rid of the hum so I went to the max. I also had the parts lying around. If you have chokes, then give it a try. That's the beauty of diy. If it doesn't make you happy, then try something else.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: