All about electronic valves and hi-fi
I think a gain of about 130-140 should be ok. Perhaps if I get around in adding the 6SF5 stage then it may be good enough.
So this got me thinking. Of course I have on my list 2 driver tests:
Lately I haven’t had any time for audio work unfortunately. Changing nappies to a 4 week old baby whilst working long hours is tough. I can get the odd 30 minute here and there and every time I try to get upstairs to the workshop something pops up. Never mind, hopefully things will get easier in the near future.
I’ve been asked about the 4P1L pentode driver. It’s been a long time since I did those tests and never got around to listen to the driver sound. Tests were promising but never managed to include this driver on my amp.
Driving transmitting valves is a challenging task. Especially if we want to take them to A2-land (unless they operate in A2 whilst in zero grid bias). Driving big transmitting valves like 211, 805, 845, 813 or GM-70 require a large swing of volts for the driver which should do this linearly. The load is quite demanding in particular when we approach the grid to 0V (or biased positively) and using a triode as driver also puts a daunting task to the previous stage due to the Miller effect. It’s not easy to find triodes that can swing 300Vpp with very low distortion.
After the initial tests done with the 4P1L in pentode mode and filament bias, I thought a bit about how this driver could be implemented in practice. I may try this configuration in my 4-65a SE amp, but am not urged by this at all.
The screen supply is formed by a gas valve (SG3S) which provides a very stable reference when feed by a CCS. In this case the cascoded pair M3 and M4 will provide in conjunction with R4 and U2 a very low noise screen current to U1. R9 has to be adjusted on test to set a current of about 15mA on the CCS. The 4P1L will draw 1.8mA at 81V as screen current, so R4 may also need adjustment to set the right operating point.
The driver should provide about a gain of 150. Driving easily a transmiting vale (or why not a 300B 🙂 ) in class A2 with this configuration. My tests showed a maximum THD of about 0.27% at 200Vpp.
Gain could be reduced if needed by tweaking the RL. And if stacked supplies are not used, then a single +400V supply could be used with additional dissipation across the reference currents (M1 and M3/M4).
Keep thinking…
After the early experiments with the 4P1L driver in pentode mode, I decided to look at improving it somehow given advice given. The gyrator load is not a good match for a pentode unless the reflected impedance is low enough to control the gain of the stage. Gary Pimm recommends:
“In the driver experiments the plate resistor was increased to a value larger than in traditional Pentode driver stages to get more gain.A CCS was placed in parallel with the plate resistor to add plate current to compensate for the high value plate resistor. This allows you to have independent controls of the gain and operating current. The resistor is chosen to set the gain and the CCS is used to set the Pentode operating current.So I opted for adding a resistor in parallel (RL) to adjust gain, minimise distortion and improve PSRR:
The load resistor is 68K. I optimised the operating point to reduce distortion at maximum swing (i.e. 200V peak to peak). The input impedance of the soundcard interface which is 100K didn’t produce a significant impact on the distortion when measuring from the anode output or in the mu output:
Interesting to see that distortion is now nearly half of previous operating point and 0.27% for 200Vpp is very good.
The screen current is approximately 1.8mA at 81V bias.
Inspired by Rajko’s model, I tried creating my own Spice model of the 4P1L pentode:
Dmitry’s tool for pentode is difficult to fit to the pentode curves, not as the triode tool which is very accurate. Here’s Dmitry’s explanation:
“The pentode model equations, as defined by Koren, have one very, very important property: they reduce to a pretty accurate triode model of the same device if the screen moves together with the plate (I have a parameter UL for that and for ultra-linear arrangement, UL=1 is triode – try it! – then ul=0.4 or so is for UL connection, UL>1 is so called supertriode connection). That’s the good news, a really good one but that is the end of the good news. The bad news is that pentode and tetrode fitting can be at best very approximate – that is, the knee region and the slopes are not right for most tetrodes and pentodes, with the exception for some small-signal pentodes.”
Here is my model. Hopefully someone can try it and report some results:
** 4P1L PENTODE ************************************************************
* Created on Sun Jan 06 18:21:28 GMT 2013 using tube.model.finder.PaintKIP
* model URL:
*————————————————–
.SUBCKT PENT_4P1L PENTODE 1 2 3 4 ; P G K G2
+ PARAMS: CCG=9P CGP=0.1P CCP=9.5P RGI=2000
+ MU=9.27 EX=1.4139 KG1=1658.84 KG2=3528.0 KP=469.2 KVB=40.504 ; Vp_MAX=450.0 Ip_MAX=0.07 Vg_step=1.0
*————————————————–
RE1 7 0 1MEG ; DUMMY SO NODE 7 HAS 2 CONNECTIONS
E1 7 0 VALUE= ; E1 BREAKS UP LONG EQUATION FOR G1.
+{V(4,3)/KP*LOG(1+EXP((1/MU+V(2,3)/V(4,3))*KP))}
G1 1 3 VALUE={(PWR(V(7),EX)+PWRS(V(7),EX))/KG1*ATAN(V(1,3)/KVB)}
G2 4 3 VALUE={(EXP(EX*(LOG((V(4,3)/MU)+V(2,3)))))/KG2}
RCP 1 3 1G ; FOR CONVERGENCE
C1 2 3 {CCG} ; CATHODE-GRID 1
C2 1 2 {CPG1} ; GRID 1-PLATE
C3 1 3 {CCP} ; CATHODE-PLATE
R1 2 5 {RGI} ; FOR GRID CURRENT
D3 5 3 DX ; FOR GRID CURRENT
.MODEL DX D(IS=1N RS=1 CJO=10PF TT=1N)
.ENDS
4P1L is a sublime DHT. As shown before it’s one of the most linear valves in triode mode. I built a breadboard in filament bias to test 4P1L as a driver using a MOSFET gyrator in mu-follower mode:
My test set can only drive the 4P1L output to 30Vrms and the distortion is only 0.027%!
I was intrigued by the performance of this driver in pentode so did a quick modification to provide a screen fixed voltage instead via a source follower and adjusted the bias voltage to minimise distortion. I found that a bias of about 120V was the best. This setup wasn’t the ideal one as in filament bias the frequency response is really poor as there is no cathode resistor bypass. The gain is about 200 with the gyrator used:
A distortion of 0.58% @ 200V peak-to-peak is really good. The filament bias is forcing the pentode to operate with low anode current so I guess that with a lower bias point performance will improve. I will have to test this.
The measured THD was:
Interesting to see the increase of H3 and H5 as a result of the pentode operation.
The breadboard for pentode can be improved for sure. I will look next at reducing the bias voltage as a first step. Interesting results which show that 4P1L is a great driver both in triode and in pentode modes.