by dcgillespie » Fri May 13, 2011 6:25 am
How did you like the change to KT88s in your Quick Silvers? I would think that was a good move sonically besides resolving basic issues associated with the 8417. But in going from 17s to 88s as you did, besides the bias and FB considerations, I would also want to make sure that the phase inverter stage could adequately drive the KT88s as well.
To me, the 8417 is just a weird tube. It is a frame grid tube, that (IMO) is not nearly as beefy as more traditional big bottle output tubes are. It is also a cavity anode design, which is what I believe gives it some of it's rather peculiar characteristics. Finally, while it's capabilities appear to be on par with that of the 6550, it must be remembered that the maximum ratings for the 8417 were developed under the Design Maximum guidelines, where as those for the 6550 were developed under the older more conservative Design Center standards (at least as published by RCA). On an equal playing field, I believe that the 6550 is a much more hearty tube.
A decade or so before the Internet, I could find little to no design data on the 8417, so at the time, I manually ran and developed my own set of plate curves. It's been a long time since I reviewed the data, but I remember the tube as being a much lower impedance device than traditional ouput tubes are. However, in my testbed amplifier incorporating a Dynaco A431 OPT, the tubes consistently delivered about 10 watts less available power than 6550 class or EL34 tubes could develop when using tapped screen operation. After that, I really did very little more with the tube.
In the MK VI, the OPT provides a rather typical load environment for the tubes, with the OPT representing a 2200 ohm primary, and screen taps placed at 40% of the winding. I've never had a MK VI on my bench to know of its actual performance capabilities, but this is not significantly unlike what a pair would be loaded with when using an A431 (or an A470 for that matter). The Quick Silver shows a nearly identical 4200 ohm load being supplied to the tubes. While I don't know what position the screen tap occupies on the QS's primary winding, based on my experiments anyway, I would think that the move to KT88s was a positive one.
In the end, I think I became so gun shy of the tube, that I never performed any significant listen sessions with it. I'll be interested to know what yours are.
Dave