I have put back the EL84 and now the hiss is gone, it sounds just I like it, soft en silki and still fast and with full bass.
The 6n6p was yhe cause, possible because of the paralelling of the two sections, tube sections are not always the same what can cause oscilations.
regards
I have done the gyrator like this, it sounds wonderfull, a lot less harsing then ccs in my case, I have keeping the CCS for the cathode follower who will be changed for EL34 to drive the mosfet gate capacity, EL84 is now in it.
I see that calculating a gyrator is difficult, I have only change R5 R6 and R7 for tube idle of 1.2mA and 155 volt across the tube, higher idle did,nt work nicely the sound get very harsing.
regards
I see that calculating a gyrator is difficult, I have only change R5 R6 and R7 for tube idle of 1.2mA and 155 volt across the tube, higher idle did,nt work nicely the sound get very harsing.
regards
Attachments
Yes I see, it is drawn wrong in schematic as a testing purpose, I have in real live tapped from anode, but maybe it is nice to tap from source for low impedance driving a EL34 to drive multiple mosfets.
regards.
Last edited:
Ahh oke thank you, I have it on anode but have plans to make a extra cathodefollower with this double triode to drive the EL34 later on, I have a free section left, I have because of the low current of the ecc803 where it sounds so relax that the mosfet hybride is like almost a real tube amp (but not completely this will never happen but close), the idea to change the fet for a low current one, so capacitance are low like the ones you already use, I live in the netherlands so parts from phillips is not a problem.
The voltage over R5 I have plus minus 5 volt like you advised in some topics, mine was now 7 volt..
I now listening to a jazz music, it is really relaxed, it is a big improvement with CCS, but only at anode tapped ac, no differents with cathodefollower where I have a CCS driving the mosfets (verticals 2sj201 2sk1035).
The voltage over R5 I have plus minus 5 volt like you advised in some topics, mine was now 7 volt..
I now listening to a jazz music, it is really relaxed, it is a big improvement with CCS, but only at anode tapped ac, no differents with cathodefollower where I have a CCS driving the mosfets (verticals 2sj201 2sk1035).
Oke thanks, I have a version with two mu followers and feedback. It is the kimmel like version, but that is not a gyrator, so I presume only change the fet and the two gate resistors has to be switched, it is mirror like..
First I try this I have a section of the tube free, so it can easely act as a cathode follower, so now with the P fet and ac taken from penthode it sounds very pleasant, replacing the irf mosfet with bsp tiny one gives a very low distortion, so capacitance act as a non linearity in the gyrator and afcourse the ecc803/12ax7 tube is low current and capacitance get a problem as the mosfet is to big, so I have now a 120 mA mosfet in use like you do with the bsp versions...
I don,t understand the fase shift around 700 hz on the ac analyses picture maybe it is not 100 procent perfect simulation, it never be but sometimes very close. .
Rippelrejection with the cathodefollower will be not as good, but I have a very good supply.
see pictures
First I try this I have a section of the tube free, so it can easely act as a cathode follower, so now with the P fet and ac taken from penthode it sounds very pleasant, replacing the irf mosfet with bsp tiny one gives a very low distortion, so capacitance act as a non linearity in the gyrator and afcourse the ecc803/12ax7 tube is low current and capacitance get a problem as the mosfet is to big, so I have now a 120 mA mosfet in use like you do with the bsp versions...
I don,t understand the fase shift around 700 hz on the ac analyses picture maybe it is not 100 procent perfect simulation, it never be but sometimes very close. .
Rippelrejection with the cathodefollower will be not as good, but I have a very good supply.
see pictures
Attachments
like Wavebourn says, R5 is needed for compensation on very high frequency end where capacitances of MOSFET start playing role.
Some kind of degeneration resistor?
__________________
The devil is not so terrible as his mathematical model!
Wavebourn: We Create Creativity!
Last edited:
I am back again, wel the news is with a 12ax7 on 1 mA idle I get slewing with the gyrator, an square wave become a sawtooth on 18 khz, the sound however is very nice.with a sweep I miss some around 15 to 20 khz, it seems that miller capacity is playing a role here, I think more it is the mosfet in gyrator (DS capacity), I use a irfu9310 mosfet, I think I get to use a small one if I use low idle currents, I have use the 12ax7 as a anodefollower and the other section as a kathode follower, I had a tip to use a cascode circuit, but then a feedbacked one, like in example, 10k anode resistor is a gyrator in my simulation, I have not a measurement in real yet, if it is mosfet problem then here also I get problems, because cascode has high resistance on output and I need low...
Picture of scoop was with a CCS to anode, driving a EL84 directly without a kahode follower and that was fine like you see, (20Khz) nice square wave.
Last edited:
Welll CCS is back again, then I have no problems with 12ax7 anymore, sound is indeed more harsh with a CCS but stage is deeper, so I have idle the follower higher so I get some warmth, later on I try the bsp25 if I can get it here, because I live in a phillips country., maybe I get answer about this problem of miller here?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.