Hi!
I need to have a vacuum tubes pre-pilot for a current BJT.
I need 40 volt rms maximum at minimum THD+N (0,5% at max) ad very low output impedence (max. 100 Ohm)
So I'm looking for a scheme of an Hybrid SRPP i think the best solution and very simple.
The idea is to use a paralled 6922 as the lower tube and a transistor (Bjt) like MJE 350 or 340 as upper dispositive.
The power supply voltage shold be around 300 volt to have the swing I need.
Any suggestion??
Thank you very much.
Gian Maria. Rome. Italy.
I need to have a vacuum tubes pre-pilot for a current BJT.
I need 40 volt rms maximum at minimum THD+N (0,5% at max) ad very low output impedence (max. 100 Ohm)
So I'm looking for a scheme of an Hybrid SRPP i think the best solution and very simple.
The idea is to use a paralled 6922 as the lower tube and a transistor (Bjt) like MJE 350 or 340 as upper dispositive.
The power supply voltage shold be around 300 volt to have the swing I need.
Any suggestion??
Thank you very much.
Gian Maria. Rome. Italy.
hi
One solution would be a 6as7 cathode follower and
no need mje350 because it s very linear.
www.loginnovation.com/jeapel
zout (open loop)=100ohm
zout (close loop 6db feedback)=50ohm
http://www.zenn.com.sg/Why hybrid.htm
Or 6as7 with mje350(+ heatsink) but 100ohm zout
with tube is not a easy thing without a big
feedback and i dont like big feedback 10 db feedback
would be for me a max for a good amp.
One solution would be a 6as7 cathode follower and
no need mje350 because it s very linear.
www.loginnovation.com/jeapel
zout (open loop)=100ohm
zout (close loop 6db feedback)=50ohm
http://www.zenn.com.sg/Why hybrid.htm
Or 6as7 with mje350(+ heatsink) but 100ohm zout
with tube is not a easy thing without a big
feedback and i dont like big feedback 10 db feedback
would be for me a max for a good amp.
I just do a very very quick check maybe i m wrong :-(
1-ccs with mje 350 + heatsink ( ~60v @ 70ma)
2-6as7 (~125v @ 70ma best is around 60% of pa max)
3-R cathode + bypass cap (bias ~50v)
for a supply of ~ 235v
test with 6db or 10db feedback to get 100ohm zout
good test it s a interesting project
1-ccs with mje 350 + heatsink ( ~60v @ 70ma)
2-6as7 (~125v @ 70ma best is around 60% of pa max)
3-R cathode + bypass cap (bias ~50v)
for a supply of ~ 235v
test with 6db or 10db feedback to get 100ohm zout
good test it s a interesting project
Hi and thank you.
I don't want to use feedback.
I just need a simple schematic of Hybrid SRPP (no mu follower), using a 6h30 or 6922 paralled or ecc99, with a MJE 350 or MJE 340.
The schematic reference for the ccs is onhttp://www.machmat.com/sales/kits/ccs.htm, but I don't need the simple ccs, because the output impedence will be the Rp of the tube, to high for my purpose.
I need an SRPP, like Diego Nardi uses, and not mu follower that needs, by other, one more capacitor.
Thank you so much to everybody. G.M.
p.s. Maybe mark_titano already has a solution...?
I don't want to use feedback.
I just need a simple schematic of Hybrid SRPP (no mu follower), using a 6h30 or 6922 paralled or ecc99, with a MJE 350 or MJE 340.
The schematic reference for the ccs is onhttp://www.machmat.com/sales/kits/ccs.htm, but I don't need the simple ccs, because the output impedence will be the Rp of the tube, to high for my purpose.
I need an SRPP, like Diego Nardi uses, and not mu follower that needs, by other, one more capacitor.
Thank you so much to everybody. G.M.
p.s. Maybe mark_titano already has a solution...?
Hi!
First of all thanks.
Yes I know this page, but still we have complex schematic with feedback, mosfets, pentodes or mu follower.
So I repeat I would need an easier schematic SRPP with a BJT (transistor) not mosfet, strating from the CCS on the site I mentioned before.
But as I repeat, the CCS above is excellent circuit, but lack, for my needs, of a very low output impedence.
Thanks again, G.M.
First of all thanks.
Yes I know this page, but still we have complex schematic with feedback, mosfets, pentodes or mu follower.
So I repeat I would need an easier schematic SRPP with a BJT (transistor) not mosfet, strating from the CCS on the site I mentioned before.
But as I repeat, the CCS above is excellent circuit, but lack, for my needs, of a very low output impedence.
Thanks again, G.M.
Relatively complex, yes, high output impedance, no. Have you noticed there are two outputs - including a "MU" output (from the source of one of the MOSFETs)?
Konnichiwa,
The short answer is "You cannot do that".
Here is why:
1) Your BJT is an enhancement Device (meaning for the NPN device it's control electrode [base - which is analogue to the valve's grid] needs to be positive with respect to the common electrode [emitter - which is analogue to the valve's cathode]).
2) Further not only are you asking to use an enhancement device, but also one where the control electrode draws significant levels of static and signal dependent current.
The "long" answer is: "You can achieve something that is close to what you want, BUT it requires great complexity and is unlikely to perform well."
Here is what would be needed:
1) You need a "ring of 2" CCS using NPN Transistors (or an NPN Transistor and a TL431 for better DC precision).
2) The "Pass" transistor needs to be able to withstand the voltages involved AND must be able to provide a high current amplification at the kind of currents flowing, implying probably a darlington device or arrangement, or indeed (dread of dreads) a Mosfet.
3) You need to feed the "error amplifier" transistor collector in the ring of 2 Circuit from a high voltage, high impedance current source, be aware that this current flowing in the error amplifier transistor is added to the current flowing in the pass transistor and that the impedance of this CCS will be in parallel with the impedance of the main circuit.
4) You need to bypass the error amplifier transistor's emitter/collector connection for AC using a suitable capacitor.
Now, all that effort to void using a single and inexpensive depletion mode Mosfet which will provide materially better performance strikes me as irrational.
Any particular reason you "MUST HAVE a BJT Hybrid SRPP" and the refusal to use a Mosfet?
Sayonara
gianmaria said:
The short answer is "You cannot do that".
Here is why:
1) Your BJT is an enhancement Device (meaning for the NPN device it's control electrode [base - which is analogue to the valve's grid] needs to be positive with respect to the common electrode [emitter - which is analogue to the valve's cathode]).
2) Further not only are you asking to use an enhancement device, but also one where the control electrode draws significant levels of static and signal dependent current.
The "long" answer is: "You can achieve something that is close to what you want, BUT it requires great complexity and is unlikely to perform well."
Here is what would be needed:
1) You need a "ring of 2" CCS using NPN Transistors (or an NPN Transistor and a TL431 for better DC precision).
2) The "Pass" transistor needs to be able to withstand the voltages involved AND must be able to provide a high current amplification at the kind of currents flowing, implying probably a darlington device or arrangement, or indeed (dread of dreads) a Mosfet.
3) You need to feed the "error amplifier" transistor collector in the ring of 2 Circuit from a high voltage, high impedance current source, be aware that this current flowing in the error amplifier transistor is added to the current flowing in the pass transistor and that the impedance of this CCS will be in parallel with the impedance of the main circuit.
4) You need to bypass the error amplifier transistor's emitter/collector connection for AC using a suitable capacitor.
Now, all that effort to void using a single and inexpensive depletion mode Mosfet which will provide materially better performance strikes me as irrational.
Any particular reason you "MUST HAVE a BJT Hybrid SRPP" and the refusal to use a Mosfet?
Sayonara
I think that gianmaria is looking for something like Diego Nardi's circuits:
http://www.webalice.it/jlc891/schemapraseto.htm
http://www.webalice.it/jlc891/schemizjal.htm
He uses an hybrid srpp with an mje340 and to polarize the transistor he used a special Hewlett Packard AlGaAs (low current) led.
This one, HLMP-K150:
http://www.fairchildsemi.com/ds/HL/HLMP-K150.pdf
Only with this one is the design works well ( Nardi's words). I suspect that the reason is because it works as coltage ref even if only 1mA pass through it.
In this way the current through the reference chain does not disturbs the tube.
Kuei, how does ti looks?
Mark
http://www.webalice.it/jlc891/schemapraseto.htm
http://www.webalice.it/jlc891/schemizjal.htm
He uses an hybrid srpp with an mje340 and to polarize the transistor he used a special Hewlett Packard AlGaAs (low current) led.
This one, HLMP-K150:
http://www.fairchildsemi.com/ds/HL/HLMP-K150.pdf
Only with this one is the design works well ( Nardi's words). I suspect that the reason is because it works as coltage ref even if only 1mA pass through it.
In this way the current through the reference chain does not disturbs the tube.
Kuei, how does ti looks?
Mark
Hybrid SRPP
Ciao GianMaria.
Ok I understand that you want a scheme of an Hybrid SRPP, but what you want and what you ask it's different. Diego Nardi makes his hybrid, because he's a professionist, ( he cannot makes or give a scheme coping the work of another )The best choice is mu follewer with mosfet see AUDION N° 11 or Alan Kimmel works It's all you need. Of course mje 340 with ALgaas led is more COOL😉
Ciao GianMaria.
Ok I understand that you want a scheme of an Hybrid SRPP, but what you want and what you ask it's different. Diego Nardi makes his hybrid, because he's a professionist, ( he cannot makes or give a scheme coping the work of another )The best choice is mu follewer with mosfet see AUDION N° 11 or Alan Kimmel works It's all you need. Of course mje 340 with ALgaas led is more COOL😉
Konnichiwa,
Good lord, now that is what I call a subidal circuit if what we want is a CCS or indeed an SRPP with Distortion cancellation.
Even with the "special LED" it will have a very non-linear impedance and with only an MJE340 the base current modulation with signal will cause material troubles. And at the best you will get around 82K Anode load, which is not much better than any Resistor. Actually, most Dual Valve SRPP's will do a lot better than that.
As I did not suggest it's use as reference voltage generation and as I suggested to feed it from a CCS the resultant circuit would be failry close to a true CCS. But compared to a basic Mosfet circuit (which is much simpler) it would still perform very poorely indeed.
So, "how does TI looks" does not make much sense in the context.
Sayonara
mark_titano said:
Good lord, now that is what I call a subidal circuit if what we want is a CCS or indeed an SRPP with Distortion cancellation.
Even with the "special LED" it will have a very non-linear impedance and with only an MJE340 the base current modulation with signal will cause material troubles. And at the best you will get around 82K Anode load, which is not much better than any Resistor. Actually, most Dual Valve SRPP's will do a lot better than that.
mark_titano said:
As I did not suggest it's use as reference voltage generation and as I suggested to feed it from a CCS the resultant circuit would be failry close to a true CCS. But compared to a basic Mosfet circuit (which is much simpler) it would still perform very poorely indeed.
So, "how does TI looks" does not make much sense in the context.
Sayonara
Kuei Yang Wang said:
This is what I've always suspected.
Nardi said that this circuit sounds better than a standard ccs. I really don't understand why it should be better.
Mark
PS: ehmm...sorry for spelling mistakes in the previous post
Konnichiwa,
For one, good technical perfoemce does not equate with good sound to each and all individuals, two, DN's circuit does provide a fairly low output impedance, which may make more of an improvement to some, subjectively speaking than the drawbacks.
Sayonara
mark_titano said:
For one, good technical perfoemce does not equate with good sound to each and all individuals, two, DN's circuit does provide a fairly low output impedance, which may make more of an improvement to some, subjectively speaking than the drawbacks.
Sayonara
Hi and thanks to everybody!
One of the reason I'd like to use a BJT, instead of a Mosfet, is all the schematic with mosfet are mu-follower and not SRPP, so for example they need one more capacitor on the signal path...
But maybe is not possible to use simply BJT and have good performance.
So I'll get thru a mosfet schematic.
Any ideas about schematic of Hybrid Mosfet SRPP?
Thank you very much, and sorry for my deep inesperience...
G.Maria
One of the reason I'd like to use a BJT, instead of a Mosfet, is all the schematic with mosfet are mu-follower and not SRPP, so for example they need one more capacitor on the signal path...
But maybe is not possible to use simply BJT and have good performance.
So I'll get thru a mosfet schematic.
Any ideas about schematic of Hybrid Mosfet SRPP?
Thank you very much, and sorry for my deep inesperience...
G.Maria
Konnichiwa,
The BJT Hypbrid SRPP as per DN uses a LED as coupling element to relay the AC from the anode of the valve to the base of the transistor, an altogether much more questionable part than a good quality capacitor IMHO.
How about using this one:
Simply take the output from the "K" Terminal of the of the IXCP10M45S.
More reading here:
Experiments with an Ixys current regulator as a CCS plate load
Little PCB's and parts including the IXYS CCS IC can be had from www.diyhifisupply.com, they do not yet list this but I have a set of 6pcs here. The CCS IC can handle 450V and you can in effect view it as a high transconductance pentode....
Very useful little critter and sounds good, though in hybrid SRPP mode a little bit of "solid state" sound creeps in, but it does that as well with BJT CCS's and Mosfet Mu-Followers.
An all valve alternative I can strongly recommend is the "Gomez Stage", in effect a combo of SRPP & White Follower, needs four triode sections but has it all, low output impedance, low distortion and great sound.
Sayonara
gianmaria said:
The BJT Hypbrid SRPP as per DN uses a LED as coupling element to relay the AC from the anode of the valve to the base of the transistor, an altogether much more questionable part than a good quality capacitor IMHO.
gianmaria said:
How about using this one:
Simply take the output from the "K" Terminal of the of the IXCP10M45S.
More reading here:
Experiments with an Ixys current regulator as a CCS plate load
Little PCB's and parts including the IXYS CCS IC can be had from www.diyhifisupply.com, they do not yet list this but I have a set of 6pcs here. The CCS IC can handle 450V and you can in effect view it as a high transconductance pentode....
Very useful little critter and sounds good, though in hybrid SRPP mode a little bit of "solid state" sound creeps in, but it does that as well with BJT CCS's and Mosfet Mu-Followers.
An all valve alternative I can strongly recommend is the "Gomez Stage", in effect a combo of SRPP & White Follower, needs four triode sections but has it all, low output impedance, low distortion and great sound.
Sayonara
Kuei Yang Wang said:Konnichiwa,
How about using this one:
Simply take the output from the "K" Terminal of the of the IXCP10M45S.
More reading here:
Experiments with an Ixys current regulator as a CCS plate load
The CCS IC can handle 450V and you can in effect view it as a high transconductance pentode....
Very useful little critter and sounds good, though in hybrid SRPP mode a little bit of "solid state" sound creeps in, but it does that as well with BJT CCS's and Mosfet Mu-Followers.
An all valve alternative I can strongly recommend is the "Gomez Stage", in effect a combo of SRPP & White Follower, needs four triode sections but has it all, low output impedance, low distortion and great sound.
Sayonara
Hi Kuei, yours are always first-rate councils. 🙂
Do you think that the IXCP10M45S can be a good solution on the cathode of a long-tailed pair stage as ccs? I'm planning to built a little el84 PP.
Regarding Gomes Stage, is this one?
http://www.tubecad.com/2004/blog0015.htm
Mark
Konnichiwa,
Hai.
Hai.
And ignore the meter reader behind the curtain....
Sayonara
mark_titano said:
Hai.
mark_titano said:
Hai.
And ignore the meter reader behind the curtain....
Sayonara
The issue I have with single mosfet CCS's (and I include the 10M45S with the single mosfets) is the high shunt capacitance. Yes, the performance curves look a lot like pentodes but they don't share pentodes low capacitance numbers.
A simple way around the high capacitance problem is to use 2 devices and configure them in a cascade connection. In the cascade configuration the shunt capacitance is reduced greatly.
In the CCS development work I've been doing the shunt capacitance has turned out to be the bad player for sound quality. Each time the CCS's design reduced the shunt capacitance the sound quality improved.
Very low shunt capacitance CCS's do a very good job keeping the solid state sound out.
Stuff on my webpage that may be of interest:
For a look at the differences between single mosfet and cascade connected dual mosfet CCS's take a look at the "CCS's and signal current controll" presentation that I gave at one of the Bay Area Triode Festivals in San Francisco a couple of years ago.
Also, you can see the differences in performance between several different CCS variations including the 10M45S and the Supertex DN2540 in the CCS performance measurment section.
Gary P's DIY page
Member
Joined 2009
Paid Member
Maybe this thread is too old ... but the last post has some interesting comments about the impact of CCS design on sound quality that I'd be interested to read about - only the link at the bottom to Gary's page is no longer working ?
- Status
- Not open for further replies.