Hi Ale, Yes, it is short of voltage. C2 should have about 10V across it to set a stable current in the P-FET, and The FET needs about 20-30V of Vds, minimum. When the drain-source voltage is too low, the current source shows a lowered dynamic impedance, which makes a bad current source. The LED string needs to increase to at least 25V for good CCS behaviour.
The 6Ж51П circuit looks really good!
Ale -
Thanks for the starved filament data! I'm reviewing the filament voltages in my amp. I just finished a new one tonight and listening to it. It's pretty damn good! First stage is 46 in unbypassed cathode bias (2.1K, 24v) into a Hammond 126C. This is an old favourite combination. Running the B+ here from a couple of glow tubes VR75 and VR150, giving 218v on the anodes. The 300b has cathode bias 1.3K with 100uF ICW PW polypropylene bypass, 66v right now and 388v on the anode. Running about 50mA but I can push this up since OPT is LL1660/80mA.
I'm a little short of volts in the power supply - about 400v HT under load. I'm using two 6BY5A double damper diodes with 3.3H chokes at 300mA and a bunch of 40uF motor run polypropylenes. Input cap is a 10uF polypropylene - about the max for a 6BY5. Mains transformer is 366v AC on the secondary. 400v AC would be better. I can see a super new power supply coming! Also need to convert to Rod's regs - using LM1084s right now. But sound is extremely detailed and nice smooth treble. The usual seductive 300b SET midrange! Listening to a bunch of female singers - what else with a 300b SET!
Happy tonight.
Andy
Thanks for the starved filament data! I'm reviewing the filament voltages in my amp. I just finished a new one tonight and listening to it. It's pretty damn good! First stage is 46 in unbypassed cathode bias (2.1K, 24v) into a Hammond 126C. This is an old favourite combination. Running the B+ here from a couple of glow tubes VR75 and VR150, giving 218v on the anodes. The 300b has cathode bias 1.3K with 100uF ICW PW polypropylene bypass, 66v right now and 388v on the anode. Running about 50mA but I can push this up since OPT is LL1660/80mA.
I'm a little short of volts in the power supply - about 400v HT under load. I'm using two 6BY5A double damper diodes with 3.3H chokes at 300mA and a bunch of 40uF motor run polypropylenes. Input cap is a 10uF polypropylene - about the max for a 6BY5. Mains transformer is 366v AC on the secondary. 400v AC would be better. I can see a super new power supply coming! Also need to convert to Rod's regs - using LM1084s right now. But sound is extremely detailed and nice smooth treble. The usual seductive 300b SET midrange! Listening to a bunch of female singers - what else with a 300b SET!
Happy tonight.
Andy
Hi there,
Just modded my existing '26 linestage into the filament biased version.
Having built and auditioned cathode biased, battery biased and led-biased '26 preamps, my ears are convinced this is the way to go..
Great, thanks to you all for sharing!
I use Tribute interstage 2:1 in reverse as load and power supply is a capacitance multiplier. As filament supply a C-L-C-R-C and VCCS is used. '26 is biased at 4.7Volts and 110Volt on the plate.
Worth mentioning: I experienced some serious noise (hiss) at the output of the preamp which came from the used 12Watt resistor which became too hot.. Replaced it for a 100Watt type and now the noise is almost gone. Guess paralleling is the way to go here? Anyone tried graphite resistors?
If interested I can post diagram and/or photo's.. 🙂
Thanks, Rob
Just modded my existing '26 linestage into the filament biased version.
Having built and auditioned cathode biased, battery biased and led-biased '26 preamps, my ears are convinced this is the way to go..
Great, thanks to you all for sharing!
I use Tribute interstage 2:1 in reverse as load and power supply is a capacitance multiplier. As filament supply a C-L-C-R-C and VCCS is used. '26 is biased at 4.7Volts and 110Volt on the plate.
Worth mentioning: I experienced some serious noise (hiss) at the output of the preamp which came from the used 12Watt resistor which became too hot.. Replaced it for a 100Watt type and now the noise is almost gone. Guess paralleling is the way to go here? Anyone tried graphite resistors?
If interested I can post diagram and/or photo's.. 🙂
Thanks, Rob
Nice Rob, But my wife wonders why there is no gap between the transformers.... shouldn't the sound be able to creep through? and how about cross talk?? OK we're a bit "melig" here...
No gap because the maker has generously put in a shield... 🙂 (and it looks better)
..and since when do you consult your wife for amplifier design?
..and since when do you consult your wife for amplifier design?
)
I see that you used a variable resistor. I don' t think that paralleling is necessary.
I use a 10ohm 95W resistor and with Rod' s regulators there is no noise.
Also grid stoppers may not be necessary. I did not have any issues with my #26. You could try and see.
Last edited:
OK, thanks.
With a little help from Bas I will be able to give Rod's regulators a try.. Curious though, since I used the VCCS total noise- and humfree on a '45 as output stage on my DAC.
Rob
With a little help from Bas I will be able to give Rod's regulators a try.. Curious though, since I used the VCCS total noise- and humfree on a '45 as output stage on my DAC.
Rob
I would not use 6J51P here for cathode follower, it wants more voltage. 6J52P would fit much better.
Or 6J9P
Or 6J9P
Hi Anatoly,
You're probably right. See attached trace of the 6J51P. I should have extended the anode voltage swing beyond 100V, but it doesn't look that is the most linear region there.
I have a couple of 6J9P around and the filament current will fit the 01A as well, so will try them instead
Thanks
Ale
Attachments
Last edited: