New driver for SR-45

[2wo]

I Love the sound of my SR-45 but compared to most of my stuff, it has rather low gain. I know that low gain is not a subject that usually comes around here and I thought rather than pad everything else down I might try to boost this amp up.

 I most often don’t run with a line stage, for other than LP. I have a Ah Njoe Tjoeb  CD, which has variable output and volume control, The problem is with my modified Squeezebox, with or without an external DAC it won’t run the SR-45 to full power. If you can call 1600mW full power;)

What I was thinking was to change the driver for the 45 to a 12AT7. The SR-25 uses a 6CM7, half for the driver and the other for the shunt regulator. The driver half has a mu of about 21 and an rp of 10,500.  The 12AT7 has a mu of 60 and  looks like it would work well at the same operating point.  I have plenty of power supply capacity to support additional tubes but for now I just want add one socket temporarily as this amp is in a finished cabinet but I will not rule out adding two if that’s what it takes.   
Any comments or suggestions?...John         

[Paul Joppa]

The 6BN7 is fairly close, and has a mu of 28 - don't know if that is enough though. Any idea how much more gain you need? You might get away with something in the 30-40 range, in which case there are many more options among the single as well as dual triodes - as long as you are willing to add a socket. Maybe a 5842 if you are feeling sporty... anyhow, one of the frame-grid VHF triodes might be good.

Of course, the 12AT7 operating point IS well established for this application... :^)

The shunt reg can be segregated to a single tube as well, possibly an EL-84 or 6V6/6AQ5, triode wired. Those pentodes aren't rated for enough voltage, but a zener between plate and screen will make them triodes with better longevity.

[2wo]

Thanks Paul, I hadn’t even considered a different dissimilar triode. A mu of 28, may be enough. Certainly easier to rewire an existing socket than to try to add new ones. Come to think of it, I could replace the sockets with octal’s without choping up the cabinet. something to think about...John     

[Caucasian Blackplate]

I think PJ is maybe being a little bit generous about the flexibility of the design.  The need for two tubes in place of one will leave you a little stressed in terms of heater current on the Paramour power transformer, and you will likely need to adjust both R1 resistors on the driver boards to bring things back into equilibrium.

I would suggest either placing a SUT or a preamp between your sources and the SR45.  I think your squeezebox probably has low enough impedance to drive a nice pair of 600:600CT Altec/Peerless bridging transformers.  Drive the end of one winding to the CT with the source for each channel, then take your output off the 600 ohm tap.  If you really like the sound and functionality of this, you could add octal sockets in your SR45's to accommodate the transformers in the amplifiers.

If the output impedance of the SB is too high to make that work, there are plenty of 10k alternatives that should work quite well.

-PB

[Paul Joppa]

The Bottlehead PT-2 can supply 1.2A RMS of heater current at 6.3v.

I agree, any change of the driver and shunt reg tubes will call for adjusting the currents through these tubes, and the driver bias - in other words, re-designing the operating points.

[2wo]

So I finally got around to changing the driver for my SR-45. I ended up with a 6201, a 12AT7 variant. I went with this because with the mu of 60, I thought I could get the gain I wanted and I thought it would work well at the 3.6ma that the 6CM7 was running at.

Well nothing is ever as easy as it looks. My SR-45 was a quote, unquote, finished amp and was built into a found, Ikea enclosure. No popping the top plate off to make changes and I didn’t want to drill a new socket hole till I was sure this would go somewhere. So I added a "pony board " as a temporary measure with leads down to the existing circuit.

Changing the resistors for the 431 in the cathode circuit would have required ripping the PCB out, so I added  LED’s for bias.     

I fired it up and the left channel voltage were not right. After some checking and head scratching. It turns out the left channel was oscillating like mad. This was without the input shorted but grid stoppers seemed like a very good idea.

now  there is about 160V at the plate at 3.6ma, with a 1.5V cathode bias. Time to play.

The CD player was up when I hooked it up and it generally has enough gain to run the amp to a good level but  I was immediately  impressed by the huge increases in the dynamics. It was hard to believe that it was a 45 driving my ~96db speakers

Next day I hooked it up to my very modified  Squeezebox, which was the point to this whole exercise. So it plays louder, still not to full output without a linestage but boy does it sound good. the added drive of the 6201 , shunt regulated seems to drive the 45 with one plate tied behind it’s back. The formally 3D sound stage is getting scary holographic.

I am really liking this. 

So were it stands,  Custom choke input power supply Cree rectified,   MQ iron and bcp16 grid chokes, Obbligato PF caps and for now at least K40 PIO caps  and a little this and that…John           

[Paul Joppa]

Very cool - thanks for posting the steps on this adventure!

Here's another idea which I didn't think of before - it is possible to drive a 45 with a 6SL7. Run it with an HLMP-6000 LED for the cathode, and about 2.0mA plate current. The reason such a wimpy tube can drive a 45 is the very low Miller capacitance of the 45's grid, about half that of a 2A3 or 300B. Gain is 70, a bit more than the 6201, and the tube is very linear.

I don't guarantee you'll like it, but it has great potential and is quite different from the 6201 - so whatever happens, you will have tried more variables and you will have learned more.