Member
Joined 2009
Paid Member
The SS version is a bit more open and spacious compared to the tube versions. I suspect that this is the lack of microphonics with the MOSFETS.
The SS version has no hint of any "Solid state" sound. Pretty much sounds like the tube versions.
The biasing needs of each pair of enhancement mode MOSFETs makes them more complicated that a pair of depletion mode MOSFETs. The upside is biasing the lower MOSFET at a higher DC voltage that gets the part into a much better operating point with respect to inter-element capacitances. Also, there is a larger selection of HV enhancement mode MOSFETs.
Gary
The SS version has no hint of any "Solid state" sound. Pretty much sounds like the tube versions.
The biasing needs of each pair of enhancement mode MOSFETs makes them more complicated that a pair of depletion mode MOSFETs. The upside is biasing the lower MOSFET at a higher DC voltage that gets the part into a much better operating point with respect to inter-element capacitances. Also, there is a larger selection of HV enhancement mode MOSFETs.
Gary
Last edited:
I love the sound of my 807 based Tabor clone, so same but better sounds good.
My wattage needs are less than 2watts for my vintage German speakers so running some IXCP10M90's at say 200V 50mA would provide me with more than enough output. Cascading would seem the sensible solution to crushing the gate capacitance. Basically i would build out each sudo pentode as a cascade CCS.
Unfortunately i have a habit of frying these chips and they are not easy or cheap to get on this side of the pond.
This whole exercise is both a though experiment and a pursuit of increased efficiency without sacrificing the performance of my valve amps.
Shoog
My wattage needs are less than 2watts for my vintage German speakers so running some IXCP10M90's at say 200V 50mA would provide me with more than enough output. Cascading would seem the sensible solution to crushing the gate capacitance. Basically i would build out each sudo pentode as a cascade CCS.
Unfortunately i have a habit of frying these chips and they are not easy or cheap to get on this side of the pond.
This whole exercise is both a though experiment and a pursuit of increased efficiency without sacrificing the performance of my valve amps.
Shoog
Sounds like a fun project. The expense and rarity of the depletion mode parts is one of the reasons I've stuck with the enhancement mode parts. That and the fact that they work better with the higher voltage available from the local bias circuit. The depletion mode parts also work better with more bias voltage available but that kind of defeats the charm of depletion mode parts.
The enhancement mode parts take a bit more to get them going but they are dirt cheap when you let the smoke out...
Gary
The enhancement mode parts take a bit more to get them going but they are dirt cheap when you let the smoke out...
Gary
Here's a thought.
I have just successfully built a SE spud E55L with an input choke (1:2 step up) and the plate to grid feedback applied to the input choke. It worked very well.
Why not go for a spud FETRON using the same principle. It seems to me that one of these babies should bias up to about -1.5V at 200V and 50mA with a gain in excess of 50. This puts it in the same ball park as a direct drop in replacement for the E55L but in a PP implementation.
I don't see why it wouldn't work as an experiment in extreme simplicity.
Shoog
I have just successfully built a SE spud E55L with an input choke (1:2 step up) and the plate to grid feedback applied to the input choke. It worked very well.
Why not go for a spud FETRON using the same principle. It seems to me that one of these babies should bias up to about -1.5V at 200V and 50mA with a gain in excess of 50. This puts it in the same ball park as a direct drop in replacement for the E55L but in a PP implementation.
I don't see why it wouldn't work as an experiment in extreme simplicity.
Shoog
Have a look here, posts 32 thru 42:
http://www.diyaudio.com/forums/tubes-valves/159892-nice-triode-4.html#post2062625
http://www.diyaudio.com/forums/tubes-valves/159892-nice-triode-4.html#post2062625
Just to carry on the thought experiment a little longer I was trawling through the IXYs range of depletion mode components and came across this one as a suitable candidate;
http://ixdev.ixys.com/DataSheet/DS100178A(IXTY-TA-TP08N50D2).pdf
This is a 500V component rated at 800mA but with the gate to drain capacitance of just 312pf which is considerably lower than almost any MOSFET parts I have looked at. In order to overcome the intrinsic voltage limitation of running one of these in cascade to crush the capacitance, accept the capacitance and brute force it out of the picture. Use an interstage transformer with a step down ratio of at least 4:1 and drive it with at least 30mA of current and the input capacitance should be beaten into submission for a two stage four active devise PP amp.
Thoughts.
Shoog
http://ixdev.ixys.com/DataSheet/DS100178A(IXTY-TA-TP08N50D2).pdf
This is a 500V component rated at 800mA but with the gate to drain capacitance of just 312pf which is considerably lower than almost any MOSFET parts I have looked at. In order to overcome the intrinsic voltage limitation of running one of these in cascade to crush the capacitance, accept the capacitance and brute force it out of the picture. Use an interstage transformer with a step down ratio of at least 4:1 and drive it with at least 30mA of current and the input capacitance should be beaten into submission for a two stage four active devise PP amp.
Thoughts.
Shoog
No Ciss would be 315pF.
Couldn't drag the datasheet for the IXTP01N100D from anywhere so I missed that.
In the configuration I propose the whole depletion mode thing becomes a bit superfluous since positive bias can be applied to the IT center tap very easily.
FQP3N30 is still available and Ciss is only 175pF
http://www.mouser.com/ds/2/149/FQP3N30-109265.pdf
Shoog
Couldn't drag the datasheet for the IXTP01N100D from anywhere so I missed that.
In the configuration I propose the whole depletion mode thing becomes a bit superfluous since positive bias can be applied to the IT center tap very easily.
FQP3N30 is still available and Ciss is only 175pF
http://www.mouser.com/ds/2/149/FQP3N30-109265.pdf
Shoog
Last edited:
Why make life difficult - an all IRF610 design. Readily available and cheap and very low Ciss gate capacitance.
First stage PP running at 100V 30mA each into a 4:1+1 interstage driving a PP pair at 100V 50mA Schaded for triode performance driving an OT. Should be good for at least the required output of 2.5Watts.
Basically this is a Push Pull Bride of Zen driving a Zen amp but with transformers.
I think i'll build it and see how it turns out.
Shoog
First stage PP running at 100V 30mA each into a 4:1+1 interstage driving a PP pair at 100V 50mA Schaded for triode performance driving an OT. Should be good for at least the required output of 2.5Watts.
Basically this is a Push Pull Bride of Zen driving a Zen amp but with transformers.
I think i'll build it and see how it turns out.
Shoog
Last edited:
A bit off topic but why not just try mosfets at low voltage and avoid the output transformer ? RD16HHF1 has low capacitance , is fairly cheap and available . Built a hybrid choke-loaded source follower and had good results , with the low capacitance of this device , may be worthwhile trying these schaded .
316a
316a
http://www.ant-audio.co.uk/Theory/Parametric_Linearisation.pdfYou might try putting a big damper diode tube in series below the source terminal of the driven Fetron to get triode like curves.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- How about an IXCP10M90 as a FETRON