A friend in DIYAudio came up with a great idea of using the 4P1L in a different way for a pre-amp/line stage. Given availability of IT and its gapped current, he suggested connecting the 4P1L differently. Instead of using the anode as the anode element, the screen is used as anode. The 4P1L screen has a maximum current of 10mA and dissipation should be within the 1.5W.
The 4P1L curves in this mode
Let’s look at an initial transfer curve with Uak=100V:
I wired the 4P1L in the following way to allow tracing with the uTracer:
The connection is slightly different as the one suggested by Indra. If you are looking to implement filament bias, you will have to rearrange the anode and suppressor grid connection and expect a slight shift on the curves given the change in bias. Filaments are in series here, however is preferred to wire them in parallel when using this valve in filament bias as a smaller filament resistor will be required given there is twice the filament current when filaments are connected in parallel. This will help to keep the output resistance lower as the size of this resistor is smaller (remember it is reflected multiplied by μ+1 times.
Now plotting the transfer curves in this mode:
Don’t ask my why I didn’t proceed with the expected 4V step below -12V. It just happened that my brain did the wrong basic math when typing fast!
Approximating an SPICE model for these curves we can see that it matches very well:
Some thoughts
If we have a tranformer gapped between 5 and 10mA then we can bias the 4P1L at 100-113V. The anode resistance will be between 2K9 and 3K7 approximately. This is even less than half of what the 26 anode resistance is. In fact, the 4P1L in this mode has higher anode current and power capacity as well as three times of the transconductance. All in all, is a real winner on paper at least. Haven’t heard the 4P1L in this mode, but am sure it will sound really well after using 4P1L in pre-amps for many years.
I’d probably aim to run it at 8mA if possible to reduce anode resistance and operate the 4P1L out of the knee.
The Spice model
Here is my version f the SPICE model for you to play with, you can download the file from the link below.I’ve been a bit lazy and haven’t updated the capacitances. Also, we don’t have Cg1g2 (now Cg1a) and Cg2k (now Cak) in the datasheet. We can plug in the Cg1k which is about 8.5pF.
**** 4P1L SCREEN AS ANODE2 Composite DHT ***************************************** * Created on 08/04/2014 16:39 using paint_kit.jar 2.9 * Curves and model created by Ale Moglia / [email protected] * www.bartola.co.uk/valves * Plate Curves image file: 4P1L Screen as Anode2.png * Data source link: www.bartola.co.uk/valves * * Capacitances not updated !!! * RFIL is 4.45V/330mA*---------------------------------------------------------------------------------- .SUBCKT DHT_4P1L_SCREEN 1 2 3 4 ; P G K1 K2 + PARAMS: CCG=3P CGP=1.4P CCP=1.9P RFIL=13.6 + MU=9.6 KG1=1545 KP=111.6 KVB=256.7 VCT=0.374 EX=1.43 RGI=2000 * Vp_MAX=200 Ip_MAX=20 Vg_step=2 Vg_start=4 Vg_count=12 * Rp=4000 Vg_ac=55 P_max=40 Vg_qui=-48 Vp_qui=300 * X_MIN=108 Y_MIN=158 X_SIZE=416 Y_SIZE=490 FSZ_X=1128 FSZ_Y=758 XYGrid=false * showLoadLine=n showIp=y isDHT=y isPP=n isAsymPP=n showDissipLimit=y * showIg1=n gridLevel2=n isInputSnapped=n * XYProjections=n harmonicPlot=n harmonics=y *---------------------------------------------------------------------------------- RFIL_LEFT 3 31 {RFIL/4} RFIL_RIGHT 4 41 {RFIL/4} RFIL_MIDDLE1 31 34 {RFIL/4} RFIL_MIDDLE2 34 41 {RFIL/4} E11 32 0 VALUE={V(1,31)/KP*LOG(1+EXP(KP*(1/MU+V(2,31)/SQRT(KVB+V(1,31)*V(1,31)))))} E12 42 0 VALUE={V(1,41)/KP*LOG(1+EXP(KP*(1/MU+V(2,41)/SQRT(KVB+V(1,41)*V(1,41)))))} RE11 32 0 1G RE12 42 0 1G G11 1 31 VALUE={(PWR(V(32),EX)+PWRS(V(32),EX))/(2*KG1)} G12 1 41 VALUE={(PWR(V(42),EX)+PWRS(V(42),EX))/(2*KG1)} RCP1 1 34 1G C1 2 34 {CCG} ; CATHODE-GRID C2 2 1 {CGP} ; GRID=PLATE C3 1 34 {CCP} ; CATHODE-PLATE D3 5 3 DX ; FOR GRID CURRENT D4 6 4 DX ; FOR GRID CURRENT RG1 2 5 {2*RGI} ; FOR GRID CURRENT RG2 2 6 {2*RGI} ; FOR GRID CURRENT .MODEL DX D(IS=1N RS=1 CJO=10PF TT=1N) .ENDS *$
The model can be found here as a well: “4P1L Screen as anode SPICE model“.
Ale,
If it is going to sound as good as it looks on the paper then it is well worth to go for it with a LL2745/8mA replacing the 26 in your Gen 3 26 line stage ;)). It will increase the capability to drive long cable ( cut the Zout in half).
I am tempted to try. I’ve built a line stage based on your design and it is not very hard to mod it. I just have to replace (adapt to the hole) sockets, mod Coleman regs, adjust SSHVs and replace filament to ground resistors. (see my built here http://simplepleasuretubeamps.wordpress.com/2014/06/26/26-linepreamp/ )
My only problem is that I don’t have a Phono stage yet ( I am in process of building Kevin Kennedy’s “Muscovite”and eventually Salas’s simplistic) and in my actual setup I don’t need a line-stage ( I have the PSE 4P1L amp fed with short Cable (3ft) from a CD player (Marantz 5001).
Thanks for sharing,
Radu
Hey,
Vg=-8V Vak=108V Ia=8ma Ra=choke 200H Mu=7 real
Piotr
Hi Piotr,
Did you listen to this or was just a bench test?
Cheers
Ale
Hi Radu,
Yes, it’s a perfect drop in for a quick test. Unfortunately I’m running the 4P1L Gen3 after modding the 26 preamp. Not easy way to go back, however will breadboard it when I get the chance.
If you’re using an SSHV2 you can dial the anode voltage and change the filament resistor. The Rod Coleman boards can be tweaked to adjust the 4P1L filament current.
A must try if you can play with your 26. You can compare sound of both and report results. Would be very interested in your findings.
Cheers
Ale
Hi Ale,
Only test.
Cheers
Piotr
Ok. Difference in mu may be due to filament wiring. You should expect mu to be between 8 and 9 generally.
I am wondering what the rationale, the raison e’tre is for this? I mean I am no conformist, but is there some advantage?
If you don’t have an OT gapped at 20mA or more you can’t use 4P1L.
I haven’t listened to this valve wired this way so can’t comment at this stage further on sonic or microphony improvements
Hi, Ale. I just had to try this 4P1L “screen-as-anode” (SAA) mode. I have a linestage with a Salas reg for HV and a pair of Coleman regs for filament power. It uses filament bias and an LL1660/10ma for output. To change tubes, all I usually need to do is swap out the filament resistor, then change the HV and filament current. I’ve used 26, 301A, and 112A so far.
Changing to the 4P1L required new sockets, of course. I used a 16-ohm filament R (2 8-ohm Mills 12W in series, because that’s what I had). I connected the filaments in parallel and set the current to 550ma (filament bias of about 8.9V). I then adjusted the HV for about 8ma screen/plate current, which ended up being about 130V. I have (2) 4P1L’s on hand (more on order!), so one tube pulled about 7.5ma, the other about 8.2, still well within the 1.5W rating for screen dissipation.
I’m stunned! I like a big, bold, colorful image, and this really delivers! The clarity is outstanding. Clearly the best so far! Bass goes deeper than any of the tubes mentioned above, but this is probably because I’m driving the LL1660 from a lower plate resistance. It’s certainly not as laid-back as the above either, so there might be a bit of peaking in the treble response of the transformer, which is also to be expected due to the lower impedance drive.
Anyway, kudos to you for following up on Indra’s idea. FWIW, I once heard an amplifier using this mode, and it made an impression on me. It used an EL509 I think. It put out 2 really fine watts!
Hi John,
Many thanks for sharing your experience with us and I’m glad you just confirmed what I suspected after tracing and testing the 4P1L in this mode. The 4P1L is very linear and delivers a clear tone. Given it’s low anode resistance in screen as anode mode compared to other DHTs as you well mentioned, you will notice an improvement in bass without a doubt. And your ears have confirmed this. I use 01a but with in a hybrid mu-follower configuration with a depletion FET gyrator as the top load. This changes things a bit when driving the amp or any subsequent line stage. I still find the tone of the CX301a impressive, but is a personal impression. 4P1L is a wonderful valve that doesn’t stop from surprising me all the time. I started 5-6 years ago playing with it and used them as preamp, as driver and as output valve! It can work brilliantly in most of these roles. Only drawback I did find was when I listened to a 4P1L-4P1L-4P1L system, it didn’t sound as great as other combinations. Not ruling this out, but was let down only on this occasion with the 4P1L.
Still is available at a reasonable cost, but I suspect that won’t last for long despite the still big number available of sellers offering NOS. They are also very close to spec typically and are easy to match, which is a bonus!
Cheers
Ale
I’m glad I was able to confirm your “suspicions”!
FWIW, I prefer to use a step-down transformer on line-stage output. I do not need the full gain, and this also helps with driving cables.
I love the 301A’s, too! They have such a beautiful tone. But like so many other 80-year-old tubes (!), dealing with microphonics is a way of life. The SAA 4P1L exhibits just a tiny bit of harsh wide-band ringing when you tap on the tube. I think this is why it’s so clear sounding.
Agreed, better to step-down for sure. Miller is not a good friend 🙂
I have had a superb experience with the 01a so far. I run them with starved filaments and are dead quiet. No microphonic noise at all. On the other hand, 4P1L is a real singing lady, as I wrote time ago…
cheers
What are your operating points for the 01a? I.e. filament current, anode voltage/ current?
John, have a read through this: http://www.bartola.co.uk/valves/dht-pre-amplifier/01a-preamp-gen2/
You will find it useful I hope!
I shall study! Thanks.
Hallo Ale,
In 1997 I tried to mimic a mesh anode power tube because I found the sound of mesh anode powertubes superior to the ones with solid anode plates. I therefore tried the pl519 (and pl509 of course) because of it’s high power dissipation (9watts) and also a low mu (3.8) like the 2A3.Especially single ended low mu tubes sound best imo.
The driver consisted of a mu-follower ecc 85 (6aq8). The amp sounded very good : big sound stage, very detailed and heavier bass than the otherwise very good sounding 2A3. Va 300v, Rk 1k, Zo 3k (not critical) .Several friends built similar amps varying driver stages and number of 519’s (1-4) all with very good results.
An American acquaintance (Jonathan Weiss) of one of my friends (Paul) was given the material to build this amp and some time later Paul called me to tell that Jonathan was now selling this amp. You can find his ad as Oswalds Mill Audio in the amp section.
I emailed J. because the halleluya story in the ad about how he used a televion deflection tube in a completely new way with g2 as anode did not mention where he got it from. I told him I didn’t care about the commercial use of the idea but that I wanted my name attached to it. And so he did.
Tonight I was curious to see if anybody had also tried this idea as J. had in fact given all the details in the ad so that anyone with some experience could build an amp like that. That’s how I found your site.Among others I found some Chinese firm selling a headphone amp where the driver uses g2 as anode.
I use the 519 with 13 watts g2 dissipation without any problem. It becomes a dull red only at 16 watts dissipation! Also the Pl 504 can be used. The overall sound very much resembles that of the 519 except for the bass which is a little lighter. When you’re interested I can give you more information as to what specific tube properties determine the sound you can expect from any power tube used this way
(I’m sending this email from my girlfriend’s place. My email address is [email protected])
Best regards! Kees Brakenhoff, Leiden, The Netherlands
Hi Kees,
Many thanks for sharing your experience, this is very interesting. I will drop you a line shortly.
Ale
I would only like to add that I have heard Jonathan Weiss’s amplifier at Oswald’s Mill and it is very special indeed!