Quicksilver 8417 Design - Stay with it or modify?

[john65b]

I would recommend a lower value pot frankly of 5K or 10K with more current flowing through it. This will make the bias less prone to changing due to grid current, for the same reason adhere to the recommendations about max grid circuit resistance for the 8417.

A 10K pot and raise the series resistor from 10K to 50K?

[kevinkr]

Series resistor gets scaled in the same direction as the pot so for a 10K it would go to 2K, and for a 5K pot it would go to 1K. You want exactly the same voltage and drop across the pot at 5X or 10X the current. (I would probably select 10K pot combo for 5X the current.)

[Original Burnedfingers]

I have some info in my Quicksilver section that might be of some interest to you. The Quicksilver actually sounds much better with 6550's in it. Individual tube bias yes. I would NOT suggest changing the front end. Keep the 12AX7 and the 12BH7. The 12AU7 sounds like syrup.

As for the Quicksilver sounding bad per one member here..well to each his own. The sound is a result of several factors... the preamp or line stage, the source gear and the speakers.

[Original Burnedfingers]

With respect to the short life of the 2) 5AR4 rectifiers.... back in 2002 I fitted my Quicksilvers with diodes prior to the rectifiers. The result is the amp retains the rectifier sound and the slow B+ warm up. The power supply caps were left stock. The amps STILL have the same 5AR4's that I had installed years ago. Thus the rectifier problem has been solved.

Info in my Quicksilver section.

I've also carried this idea to every Dynaco that I come across and the results are very positive. What a difference a couple of dollars worth of diodes do make in increasing the life of the rectifiers.

[john65b]

I finished this amp a while back and used a NOS set of KT120s from Tung Sol. Sounded good. Got better after time. I had 420v on plates and 60ma per tube. 12AX7 into a 12BH7. grid resistors are 47k and left the .22uf caps (still need to raise these to .47uf with 47k grid resistors)....I used the KT120s after I initially had an 8417 flash/die on initial bias up. These 8417 were just too expensive to lose.

Last night I was feeling a little bold and decided to try the 8417 again (I have a stash of 10 now). they dialed up to 60ma each fine and I monitored the bias for another runaway event. Plate voltage is 430v. Everything appeared stable. I set up a listen and it sounds great. It sounds very clear and detailed. I think I may like it better than the KT120s.

I am running them at 74% of max plate dissipation. I also ran PP TubeCAD PP calculator and the curves look very good at this operating point.

The thing that worries me if turning my back on the amp and having it go into another bias runaway. Never had to worry about this with the KT120s....

[tyu]

john....how an why did you setup the amps for 430B+??....what is that 30-40 watts
What speakers are you useing....ESLs??
Looks like great Diy amps....an as far as anyone here.... can know how you amps sound with 8417 are any other tube.... would have to have your same transfourmers..... there not stock QS
Back in the 90s i had the QS 8417 amps..an setup more than one pr for other people..never like the sound....even with KT88,6550s.....an had lot of good preamps...like Melos,cj,ARC,Krell....8417 tube was made to run at higher B+ i thought.............have fun with tubes.....
all just one mans O-pine..

[tyu]

Well i gees my brian has stop working.....an only after 65years....i got tubes older than that!
I gess all this tells you is As nice a guy as Mike is....back befor the net.....even his preamps sounded bad....
i dont remimber the B+ be that low..... more like 460................430 thats less the a Dynaco ST70...
what am i missing here??
..good luck with geting more than 35watts out of that....

[john65b]

..good luck with geting more than 35watts out of that....

From PP Calculator, with 430 B+, 60ma each tube, and 4200 output tranny, I am getting 55wpc with 4 ohm speakers with the 8417...I am using them on Acoustat 2+2. Sound great, even better than the Tung Sol KT120s.

[john65b]

The 8417 has 35w max plate dissipation, and a max plate voltage of 660V.

Yes, I could do 550V and 50ma each tube for a Class AB1, but my PS caps are rated for 450v...and according to the calculator, the curves look best at my operating point of 430v, 60ma.

[Johan Potgieter]

Originally Posted by artosalo

The key point seems to be forgotten.
Direct connected voltage amplifier and cathodyne phase splitter is a compromise.
It is working well only when a pentode amplifier is used, since this operates at optimum point with sufficiently low anode voltage.

Was waiting for this to come along. Why are we so bent on using a high gain triode input? In NFB circuits the fact that the input impedance is left at the mercy, thus may be anything from low to several 100K can affect the circuit stability or at least optimal h.f. performance, by virtue of the first triode's variable Miller capacitance. (Personally I have never seen a demand for an ECC83 in power amplifiers. I start with ECC88 or better.)

This is not the case with an input pentode, with the several other advantages. Something like an ECF80 works well within it limits, with hardly any capacitace at the signal input. Generally input pentodes also generate about 25% of the distortion of triodes. This not a great tribution to the total; still.

Lastly - sowhat OT: What is the general life span of an 8417? I read several reports that they were not rugged and generaly had short life-spans?

Any experiences - the OP allowing?