• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

One more 4P1L SE

Deco-doctor

Have a look at the 126B and 126C at Bluebell Audio - look at Transformers > Interstages. They are exactly the same price as Danbury and probably at least twice as good. Also they are bifilar wound and have a known and stated inductance. Philip at Bluebell is very helpful and knows his tubes.

Hi Andy

Thanks for your thoughts. I am already using Hammond 126A interstage in my 45 SE amplifier. I use a triode strapped D3a input/driver running at 160V 12mA. And yes the 126 series are super value. The square wave response is astonishingly good.

I could try the 126A as a load for the 4P1L at up to 15mA if it sounds as good as you say at low current.

Martin
 
Hey Ale,
At what load did you get the sweet spot?

Hi Lars,
I run the preamp at 90V, 20mA and about -4V in filament bias with slight filament starvation at about 500-530mA. I did play around quite a lot. Recently found that is more linear at bias greater than -10V providing you don't exceed anode voltage limits. But haven;t tested how it sounds in this region.

Definitely likes it working about 15mA where possible.

You can read some of the learning experience here:

http://www.diyaudio.com/forums/tubes-valves/190857-4p1l-dht-line-stage.html

Cheers
ale
 
Hi Lars,
I run the preamp at 90V, 20mA and about -4V in filament bias with slight filament starvation at about 500-530mA.
Cheers
ale

Hi Ale - what was the value of your cathode resistor in filament bias?

I'm playing around with cathode resistors and trying to find the right value right now. I won't know what different currents sound like until I get my setup working properly. It may sound better at higher currents like everyone says. The 4P1L drivers are running at more like 17mA in my PP 6C4C amp (26 inputs), and that sounds the best of all. Filament bias in SE is a different story.

Andy
 
I did a 4P1L---4P1L SE DC coupled amp which suffered quite bad microphony , if the chassis was tapped there would be a noticeable 'poing' that lasted a few seconds , sometimes when I hit pause there would be a bit of a poingy sound too . I find as output stages microphony is fine but as linestages microphony poor although I do use 98dB speakers so just about any hum/buzz/filament clanging comes through . Running the fils under 4v seemed to help a little to tame the microphony

316A

Your setup might be experiencing oscillation.
 
Right - some results! This for a projected driver stage using the Hammond 126C interstage. I don't have a 126B so higher currents than 15ma not available right now. This is in filament bias

Vp = 160v
Ip = 13mA
Rk = 20 ohms, 20W
Vk = 13v
Vfil = 1.9v
Ifil = 580mA
Supply voltage to filaments = 15vDC

Pins 2+3+4 connected to plate, filaments from 8 and 7+1

Sounds very nice - full bass, plenty of detail, very lifelike. Not too microphonic in my system - not an issue.

This is using it as a preamp. Next step to install it in a 300b SET and see what it does.

Andy
 
Bass is deeper at 13ma than at 7mA. Just installing the stage in my 300b amp today - hoping to listen soon and will report results. It will be 26 filament bias with 2x 156C and teflon coupling cap into 4P1L into 126c into 300b. Actually, I'll probably just connect up the 4P1L to the 300b in the amp and then try different preamps.

Andy
 
Last edited:
Well - 300b SET is operational. Basically it sounds very good, though far from optimised. I now have a "generic" sound of 26 into 4P1L into the output stage. So I can compare the SET with differential pairs of 26, 4P1L into 6C4C/2a3. I'm happy with this combination of tubes, though it's possible that another 4P1L stage may replace the 26 input. There's arguments for each - 26 more silky, 4P1L cleaner but not by much. I'm a bit suspicious that I slightly prefer the 4P1L.

The differences between SET and PP above are exactly what you'd expect - 300b is a touch nicer on strings (vln, cello) and female vocals. PP is better on bass and cleaner in orchestral climaxes. My speakers could do with more than the 18W plate dissipation of the 300bs at present - they're running at 330v p-k. So I don't know where the overload is coming from - intermodulation or just plain not enough watts. The 4P1Ls are running at about 12.5mA and that could be higher but not into the 126Cs.

But the basic sound is there to be improved on - very promising and now replaces the previous 46 driver stage, which wasn't in filament bias. This is and that makes a difference. So yes, it can be done and I think can be done well.

Incidentally, 7N7 has done all the building and a lot of the operating point calculations here so he's very much part of the whole project.

Andy
 
Thanks for the credit, Andy and for your nice comment Ale.

I'm enjoying all this R&D, and delighted to have become acquainted with the excellent 4P1L, the first small directly-heated valve I have come across that has a creditable specification (ducks...) with reasonable mu and good gm.

I should like to know how far one can push the upper voltage limit; with its beef and good gm it would make a superb driver for a big power stage if it could be persuaded reliably to swing sufficient voltage. 845 would be out of the question, but triode-strapped 813 at 900V and 100mA requires only about -75 - -80V bias.

The only other small DH valve I know of that has an interesting specification is YL1130, again relatively new to me, but it does not have 4P1L's linearity; I suppose it might though make an interesting differential pair, but then 4P1L makes an excellent diff. pair!

7N7