Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.||New: visit Schematic Hell!
The sunn still shines online!
|12/14/2004 10:54 AM|
||MOSFET cathode follower in Soldano-type preamp|
Hi all. I'm thinking about building a Soldano-style preamp into an existing amp, but need an additional triode stage for the cathode follower driving the tone stack. Have any of you tried the RG Keen "Mosfet Follies" large MOSFET as a cathode follower in such a preamp, and do you have any comments on the tone? Does the Soldano use the CF for distortion? It looks like it does to some degree. It would be much easier to use the MOSFET than to cut another hole in the chassis for one triode stage.
|12/14/2004 2:15 PM|
can you use the mosfet to replace some other follower?
the direct coupled follower in the soldano/marshall topology generally has some LARGE voltage swings on it. although i've never tried a mosfet there, i'd wager that it will NOT react in the same way, especially as the gate approaches the rail voltage.
|12/23/2004 8:36 PM|
I worried about that myself. Near as I can make out, the MOSFET is going to be linear outside the reaches of the tube follower it might replace, so in one sense, no, it won't react exactly the same as the tube - it will refuse to color the sound if that one point (the follower) was what was doing an audible amount of the overall sound.
The MOSFET will actully run closer to the rails linearly than a tube will.
However - I've not seen a case where a cathode follower was what was causing audible distortion inside an amp operating normally.
So - how do you tell???
|12/24/2004 9:18 AM|
what i'm thinking, RG, isn't so much the CF itself distoring, but rather the (assumedly) direct coupled preceeding common cathode stage distorting BECAUSE of the CF. ie, when the grid on the CC stage goes low, Ip gets cut off, Vp rises, then at some point when it's approaching B+, the CF grid will start to pull current from THAT stage. this will tend to "soften" the hard clip that would have otherwise occured as the CC stage's plate hits B+. the effect can be increased by reducing the Rk of the CF stage, since the increase in Ik would cause more Ig to flow.
|12/24/2004 12:24 PM|
Possibly. It's worth doing head to head comparisons (of course, that's always true.)
In the case where the driving tube is completely off, the circuit looks like the driving tube's plate resistor pulling the CF's grid to B+, and depending on the relative size of that resistor and the follower load resistor, the tube may or may not pull up to B+. It is possible to bias a triode with only a large resistor to B+. The tube is said to be "running in clamp" i.e. the grid "clamped" to B+ by the resistor. At that point, the cathode may or may not make it up to anywhere near B+. Depends on the resistor values. Some older circuits were biased that way.
In that case, yeah, there may well be an audible difference, as the MOSFET is going to follow until the gate is at B+, and will then still follow with the source one Vth down from that. The MOSFET will accurately be following its gate, the tube will be getting jerked around by grid current - if any. Some situations and resistor values will let it follow near B+. That's probably only true for high-value cathode resistors, though.
I guess the right answer is - it depends. If the CC stage driving the CF is clipping, yeah, you might get some softening depending on how the CF is set up.
Interestingly, the impedance driving a triode grid determines how sharply it clips when the grid is driven positive. A low impedance will make it softer because it can supply the grid current to pull it above the cathode, and that lets even more current through than 0V g-k, so there's no hard clip on positive grid excursions - as you know from your AB2 work. The opposite is true. The higher the impedance driving a triode grid, the harder and sharper the clipping at Vgk=0 gets. Tubes running in clamp typically use megohm sized resistors and clip flat as rulers when Vgs=0. With Vp of the CC stage at 100K to 220K, I'm not sure which way it would go, and what a 10K to 100K cathode resistor would do to the balance of available emission electrons.
The MOSFET doesn't care, of course. It's linear the whole range, so all you hear is the driving tube.
|12/24/2004 9:23 AM|
R.G., I see your point, and it makes sense. What is puzzling is how the tone is said to change with a CF. So in your experience, is the tone change mainly due to the unloading of the prior gain stage?
Can a standard gain stage with cathode follower (Marshall tone stack driver) actually drive the cathode follower into its nonlinear region, or doesn't it have enough swing at the anode of the gain stage? If that's the case, I could see why the mosfet would sound the same. And I will start spraying them all over the place, gotta love a low drive impedance...
|12/24/2004 12:27 PM|
Unloading a triode plate with a follower is the same as increasing the AC loading to equal the plate resistor, removing any following loading of the driven network. I suspect that it also improves the frequency response, so you get more trebles.
I don't play much with Marshalls so I can't say if they overdrive the follower.
|Page 1 of 2||Next>||Last Page>>|