|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
216.232.207.152
In Reply to: Final SE Amp Tube Decision posted by danlaudionut on April 26, 2007 at 16:01:27:
If you haven't already, try a rectifier diode in the screen circuit (in series) between the 100 Ohm screen load resistor and the KT plate. Anything will do, (I use UF4007's only because I have a bunch).Cathode of the diode goes to the resistor , anode to the tube's plate.
You'll hear the difference right away.
Follow Ups:
I mean, what does it do besides reducing screen voltage 0.7V?> try a rectifier diode in the screen circuit (in series) between the 100 Ohm screen load resistor and the KT plate.
Well, as you know, the more the AC control is limited to that of the control grid the more the tube is acting like a triode and therefore ideally, the diode limits the screen's AC interaction with the plate, so the output doesn't contain AC signal from screen involvement, either by its modulating the electron flow through the vacuum , or through its wired connection to the plate. In power tubes it might be even more important to find the right value of screen load resistor as well. What you're shooting for is to get a drop that puts the screen at the potential the space it occupies would be at if the screen wasn't there - making it electrically invisible.
I played around with it for a couple of years (yikes, time is passing `,;^() using smaller pentodes as linestage amps. The difference between with and without was easy to hear.
I eventually turned to other stuff and now am getting into DHT's (listening to a 46 pre as we write this) which I think is awesome, but I wouldn't write off the exploration of triode strapping multi grid tubes. If the natural desire to explore them is there, I can't see any reason why a really great sounding amp couldn't emerge from the experimentation. My guess (and it IS a guess) is that the best results may come from tubes without internally connected grids , like the KT66, so that you could try different configs at different potentials. Some people have written papers on it, but I haven't read anything by anybody who has actully invested the time to perfect it. I don't really have the inclination myself, (mostly because ALL my work gets done on the kitchen table and this project would take a LOT of trying and retrying) but I kind of hope Danl does`,;^) It would be neat to see that all those bazillions of unloved pentodes out there suddenly hold promise.
> ...ideally, the diode limits the screen's AC interaction with the plate, so the output doesn't contain AC signal from screen involvement...I don't see how it can do that. Assuming the screen is always drawing current, the diode will be continuously conducting, with only a 0.7V drop to show for its presence. Regardless of whether we perceive the circuit to be one where the screen modulates the plate, or one where the plate modulates the screen, they are connected together by the diode just as firmly as if it were a small value resistor dropping 0.7V. Only if the plate were to become negative relative to the screen could the diode do anything else. If there really is an audible difference with the diode in place, I intuitively suspect something bad is happening. Have you ever looked at the voltage across the diode with a scope, just to see if it does in fact ever turn off?
Well, I don't see how it can do that either, and if I understand what you're saying, I agree with you . The theory leaves me with many doubts and twice as many questions - especially about blocking the AC. But sizing the R to put the grid in question at a potential consistant with it's position between the cathode and plate makes sense to me. I don't believe that that automatically makes it the same as if it weren't there at all. There's probably all kinds of little effects. (If it was exactly the same it would BE the same , and it isn't, so it's not! `,;^) But it does do something positive and does not in itself create any instability (I'm assuming that's what you meant by "something bad is happening")
I can put it on my to do list but all my Study and Soldering time is spoken for up to mid '08 (yes , I actually have a little book) and by then everybody will have forgotten about this (and you will probably have lost interest!)
Nobody else is is getting involved in this now but if it interests you enough to give it a test I'd bet everybody here would be reading your results if good enough to post them.
I'll keep it in mind for a future experiment. Hopefully, I can get to it before '08, but no guarantees. :) Out of curiousity, do you know of other amp builders using this technique?
I strongly suspect that reducing the voltage by the same amount by any other means will do exactly the same! Reducing it even more will have a greater effect ... but as you intimated, it won't necessarily be good. UL was designed with benefits in mind but its effect on sonics was a retrograde step IME.
IanI saw your post but it disappeared.
I thoughr that you reconsidered the mod.
No, it's not that. I often have difficulty putting things into words. I first posted out of enthusiasm but where voltages and currents are concerned simplicity and clarity are more important. As soon as I thought of a better way to describe it I cleaned it up.
The circuit is good. I don't really understand why more people don't do it.
This post is made possible by the generous support of people like you and our sponsors: