45 SE Amplifier upgrade

Replacing driver for 7193 valve

Well, after nearly 12 months of playing relentlessly my 45 SE amplifier, had an unexpected failure in the power supply (passive regulator) which forced me to do maintenance to the amplifier. It was a great opportunity to remove the 6J5 driver and do a quick swap for the greatly respected 7193 (i.e. military version of the 2c22)

45SE Amplifier upgraded with the 7193 drivers

Bias point remains unchanged: 7mA and 260V for maximum swing and minimum distortion. Need to look at my notes, but I remember that I was something around 0.30% at 100Vpp driving the 45 (which is not an easy load for anyone).  Driver configuration was not changed, so had the 7193 now loaded with same DN2540 single transistor gyrator and mu-follower output for lower impedance. The valve was biased at about 8V with an LED array.

7193 in action

As a test, played the fantastic Symphony No. 3 from Henryk Gorecki….

 

VT-25 / 10-Y triode curves

VT-25 triode curves 

 

 

 

Everyone loves this thoriated-tungsten DHT valve. I’ve only used it in a preamp and was hooked with its sound. Really warm and nice. Downside is, it’s very pricey these days and also is quite demanding from a filament perspective.  You can check the characteristics here.

For those who like testing their designs with LT SPICE, I produced a model which matches really well the traced curves. Would like anyone to use this one, to drop me a note with any feedback 🙂

 

 

 

RCA10Y VT25 spice model

Here is the VT-25 spice model. Let me know how it works for you!

 

4-65a HT +300V supply

First breadboard of the 4-65a SE amplifier. Will start this long project with tuning the HT supplies. Will need a +600V, +300V and +100-150V supplies to be stacked.

4-65a 300V supply

Here is a simple breadboard using metal rails over an old wooden frame I had around. Oil caps will be glued temporarily and damper tubes will be fit later:

+300V HT supply (breadboard)

 

A snapshot of first test….

First tests

4П1Л/4P1L THD sweet spot

Playing around today with the 4P1L chap found a very good operating point where distortion is minimised at maximum anode power disipation:

  • Va=250V
  • Vg = -21V
  • Pa=9W (7.5W anode + 1.5W screen)
  • Ia=35mA
  • THD measured at Vo=+22.22dBu (10Vrms)

Interesting to see this valve swinging beautifully at just 0.027% THD….

4P1L THD minimum distortion

4П1Л/4P1L triode curves

4P1L triode curves

4P1L pinout

4П1Л (or 4P1L) is probably one of my favourite valves. It was an unknown device to me until was suggested by some friends in the forum. Many discarded it as being a howling beast in pre-amp stages :). I found that albeit it can be microphonic, this can be controlled to a certain extent, but in my opinion this is a great valve in most of the roles: pre-amp, driver or output stage. Preferably is such a linear valve that can easily match 2A3 and 300B characteristics (when arranged in parallel) at a fraction of their cost. You can get a view of this beauty in the datasheet here.

Looking at the specifications, the key points to highlight are:

  • Filament Voltage: 2.1/4.2 volts @ 650mA/325mA respectively
  • Recommended anode voltage: 200V with 150V on the screen
  • Maximum operating voltage 250V anode or screen.
  • Maximum cathode current: 50mA
  • Anode dissipation: 7.5W
  • Screen dissipation: 1.5W

I became aware of this valve when Anatoly (a.k.a. Wavebourn) recommended the 4П1Л  directly heated pentode  which was used in military transmitters. It is very popular now among Russian audiophiles. Apparently is the Russian equivalent of the WWII era German Wehrmacht RL2 / 4P6 RF oscillator / transmitting amplifier tube. It’s a brilliant valve when triode strapped, better than 2A3 / 300B in terms of linearity. See my post around THD here and will see why 4P1L is at the top of the chart with less than 0.03% THD @ +22.22dBu!

As Anatoly suggested, they are very nice for class A in triode, and give up to 2.5W per valve when driven with up to +12V on control grid. It is easy to parallel them, since they are consistent and very linear: paralleling linear valves you are loosing power on mismatch, i.e. the valve with higher transconductance will draw more, no distortion raise caused by mismatch. 2A3, for example, has a pair of paralleled triodes inside. You can parallel ten of 4P1L matching them (it’s easy), to get 100W dissipation and 25W output.

Many found a sweet spot around Va=235V, Ia=40 mA, Vg=-18V providing Pout= 2.5W on a 5K OT, triode connected. Capacitance between anode and first grid for 4P1L is 0.1 pF. Capacitance between screen grid and control grid is about 1 pF. It has a  10 pF Miller capacitance which is not high value, and for 20 KHz it is slight less than 1MΩ impedance. Any driver with 10 mA idle current will make it happy.

4P1L Siberian DHT Preamp

I tested this valve a lot as a DHT preamplifier with great results. Starving filaments and suspending the socket with cord can reduce significantly its microphony to very low levels. I could listen to it perfectly fine whilst my friend Tony still have some issues with a 30sp DHT stage bolt to the aluminium top cover 🙂

I loved the sound of the 4P1L pre-amp. I will build a 4P1L SE in the future, is on my list…

4P1L Siberian DHT pre-amp

Recently, a friend from the diyaudio forum asked me for the 4P1L curves which I posted previously. Here is a new trace of the curves under the following testing conditions:

  • G2 and G3 tied to anode
  • Filaments are in parallel, so If=650mA powered by DC supply.
4P1L-triode curves

 

You can create your model or use the curves to produce your load lines, etc.

Hope this helps