diyAudio (
-   Tubes / Valves (
-   -   Quicksilver 8417 Design - Stay with it or modify? (

john65b 7th June 2013 09:49 PM

Quicksilver 8417 Design - Stay with it or modify?
2 Attachment(s)
Hello all,

I am finally getting ready to start an amp project that I have been planning for years...A Quicksilver 8417 with either KT88 or EL34 (got em both - the 8417 tube was way out of my reach). I planned on using the directions from Triode Electronics website that convert the 8417 circuit to a KT88 or a EL34 (bias resistors, grid resistors, etc).

But need some advice - I know the original had the 12AX7 and the 12BH7, but I found some 6N1P, 6H30, 6922, 12AU7, and 12AT7 in my stash of tubes (as well as the 12AX7 and 12BH7 I need to complete the amp)....wondering if anyone has some advice what I could make that would best the 8417.

The schematic for the 8417 is attached. I have a pair of Edcor 4.2k 60W P-P output trannies, a pair of Edcor power Tranny (B+ should be around 500V), and the pair of 1.5 chokes...

I'm thinking a 6922 input amp and 6H30 Phase Splitter would be the best combo??

john65b 10th June 2013 12:50 AM

Also plan on using SS Rectifier instead of the tube rectifier...

DF96 10th June 2013 11:02 AM

So you want to build a particular design, but you want to change every active device in it to something different (with presumably knock-on effects on passive component values too). In what sense are you still building that design rather than simply a different circuit which happens to share the same (fairly common) architecture?

john65b 10th June 2013 12:53 PM

Question was really around a possible better tube choice, as I don't really like seeing a 12AX7 in the amps "pole position". I have read that the 6H30 makes a great splitter, and I have one. Worth the swap?

8417 tubes are expensive, and not easy to source. If I had them, I would use them...

Is that your advice? Stick to the original? OK. Thanks.

DF96 10th June 2013 01:11 PM

The 12AX7 is an excellent linear voltage amplifier in the right circuit. It needs a high impedance load, so either a CCS or a highish supply rail and a high value anode resistor. 439V should be enough. Using a 6922 as input stage would reduce the open loop gain so increasing the closed loop distortion.

That type of phase splitter has so much negative feedback that it will provide low distortion with almost any valve, apart from very high mu ones (such as 12AX7).

john65b 10th June 2013 05:53 PM

OK, I will reconsider the 12AX7....Stick with the 12BH7 as the splitter?

The 12BH7 and 6H30 have close mu (17 vs 15), and maybe not a big deal enough to change it...

The 6H30 is that tube everyone always talks about, so it has that ooh - ahhh factor ...but can anyone really hear differences in a phase splitter??

DF96 10th June 2013 06:52 PM

The issue is not that it is a phase splitter, but a stage with 100% feedback. Apart from gross design errors, that more or less guarantees low distortion.

SY 10th June 2013 07:37 PM

The thing to watch out for is voltage swing. The proposed new output tubes have lower gain than 8417, so require more grid voltage. The cathodyne has gain somewhat lower than unity, so the first stage has to swing the peak to peak requirement of the output stage plus the voltage "used up" by the cathodyne (which is significant for lower mu triodes). Then all of that has to be within the reach of the cathodyne as well.

artosalo 10th June 2013 09:06 PM

The original design has high gain output tubes as well as high gain triode as a voltage amplifier.
You plan to use 6922 (mu = 33) as voltage amplifier and EL34 as output tubes having gm = 12,5 mA/V, which is some 60 % of the gm of 8417.
This means that the whole "gain plan" is very far from the original desing.
If also some amount of GNFB is used, the sensitivity of the amplifier could be several volts (rms) to have full power.

One clear fault - especially with different output tubes - is the direct connection between voltage amplifying stage and cathodyne. These should be separated and biased individually.

GoatGuy 10th June 2013 10:31 PM

Dunno... I'm just fond of not letting one "hanging triode" do all the phase shaping. My advice - since DF96 said it first - is that since you're potentially willing to change most of the circuit path to fit your bottles ... is that you shoot for much more modest gain at the first stage, then use a 12AU7 as a 2nd gain stage, giving larger voltage swing and making more symmetric the 'signal shaping' of the triode on the signal. Hardly makes the circuit more complicated - just another triode stage and a DC blocking cap of high quality - and you're home free. Take the square root of intended "first stage gain", and let each triode be responsible for that part. Gain can be raised a bit overall (addressing various other comments) to better match your non-original-spec output tubes.

And to think: your project will have yet another cute little glowing glass jug of vacuum goodness on the topside! Worth it at near-any cost!


All times are GMT. The time now is 05:54 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2