6N1P & SV83 mu stage driving 845?
According to the Svetlana, the 6N1P & SV83 mu stage is suitable for driving a SE 845.
see technical bulletin:
http://www.svetlana.com/docs/TechBulletins/appnoteNO.27.html
Has anyone built something like this?
Thanx for any comments about this combination.
According to the Svetlana, the 6N1P & SV83 mu stage is suitable for driving a SE 845.
see technical bulletin:
http://www.svetlana.com/docs/TechBulletins/appnoteNO.27.html
Has anyone built something like this?
Thanx for any comments about this combination.
Attachments
Yes. Please try it. It's cheap.
If you read the article, 100Vrms with 0.45% THD. This is quite good!
The output impedance is quite low, 100 ohms, so such an amplifier
built with the mu stage ought to take a 211 by simply screwing in the
new tube and back off the fixed bias setting...
Yeah you could use D3a et cetera, but this is cheap to build using
tubes that cost less than usd$10.00 each.
-- Jim
If you read the article, 100Vrms with 0.45% THD. This is quite good!
The output impedance is quite low, 100 ohms, so such an amplifier
built with the mu stage ought to take a 211 by simply screwing in the
new tube and back off the fixed bias setting...
Yeah you could use D3a et cetera, but this is cheap to build using
tubes that cost less than usd$10.00 each.
-- Jim
Re: Yes. Please try it. It's cheap.
BTW, I've never paid more than $US10 for a NOS D3a.
I read the article, and I don't question the swing, but is the gain enough? I've seen plenty of 845 amps biassed around -145V, and I just don't think it's enough in most systems. A 211 is a different subject.jrdmedford said:
BTW, I've never paid more than $US10 for a NOS D3a.
Hi, (6N1P + SV83) this is the Mu-stage schematic that I used for my Line-Amp. The sound is very nice and detailed. In fact I'm using 300V B+ and changed the 6N1P's Rk=2.2K. I had tried the D3a instead of SV83, but I don't like the result! The overall sound is too bright and forwarded, a bit like metalic sound? The D3a that I used is Siemens from eBay http://cgi.ebay.com/4-x-D3a-Siemens...ryZ67816QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
Is there anyone know why?
Is there anyone know why?
I suppose SV83 is EL84 without internal connection G3-K.
Probably it has more precision winding of grids, or just more strict limits in datasheet to meet better linearity needed for TV amplifiers where 6P15P (original name of SV83) were used.
Tubelab, did you torture them? How different they are really from 6P14P?
Probably it has more precision winding of grids, or just more strict limits in datasheet to meet better linearity needed for TV amplifiers where 6P15P (original name of SV83) were used.
Tubelab, did you torture them? How different they are really from 6P14P?
Hi Dr. Strangelove,
My email button is on now, but I can't send you a mail because my account is under moderation, due to my first steps on this board.
About Siemens D3a:
All of my amps utilized the D3a penthode and I can't say anything else than that it maybe have something to do with the way you bootstrapped the tube. I've used it in several 300B and 211 amps.
My email button is on now, but I can't send you a mail because my account is under moderation, due to my first steps on this board.
About Siemens D3a:
All of my amps utilized the D3a penthode and I can't say anything else than that it maybe have something to do with the way you bootstrapped the tube. I've used it in several 300B and 211 amps.
I thought it was pretty decent of 211audio to share his design with you. So what if it isn't exactly what you were looking for, it's a starting point. Perhaps you ought to have asked him how to convert it to a mu stage, and/or why he chose to design it as he did..
Get yourself LT spice (free) and google for the tube models you need, and attempt to design the mu stage you are looking for. It will be an educational process and that way you will end up with something very close to what you envisage.
There is a lot of information about modeling tubes here on diyaudio, just search for it..
Triode models are generally quite good, pentode models of anything but the most common types can be quite poor approximations.
Very few mu stages are going to have enough gain to allow the direct drive of an 845 without an additional stage ahead of it. I think a gain in excess of 100 (40dB) is called for in such an instance, particularly if a passive line stage is to be used.
Musings: You might want to consider a cascode mu stage. (cascoded triodes with a pentode cf sitting on top of the whole thing.) It works and provides both lots of gain and the voltage swings required. The design will be very challenging.
I was working on such a variant with 2 x 6J5 (8mA iP) (a pair of triode connected C3A ought to work reasonably well too at 3mA or so) and EL34 (30 - 40mA) for use with an 845 - it simulated well and seemed likely to work in practice. DO NOT exceed maximum Vp or filament insulation ratings.
A single 6J5 with the EL34 in a mu stage will also do the job, but will definitely need another stage ahead of it.
Note that I unfortunately never got around to building the actual amplifier. YMMV, etc.
Get yourself LT spice (free) and google for the tube models you need, and attempt to design the mu stage you are looking for. It will be an educational process and that way you will end up with something very close to what you envisage.
There is a lot of information about modeling tubes here on diyaudio, just search for it..
Triode models are generally quite good, pentode models of anything but the most common types can be quite poor approximations. Very few mu stages are going to have enough gain to allow the direct drive of an 845 without an additional stage ahead of it. I think a gain in excess of 100 (40dB) is called for in such an instance, particularly if a passive line stage is to be used.
Musings: You might want to consider a cascode mu stage. (cascoded triodes with a pentode cf sitting on top of the whole thing.) It works and provides both lots of gain and the voltage swings required. The design will be very challenging.
I was working on such a variant with 2 x 6J5 (8mA iP) (a pair of triode connected C3A ought to work reasonably well too at 3mA or so) and EL34 (30 - 40mA) for use with an 845 - it simulated well and seemed likely to work in practice. DO NOT exceed maximum Vp or filament insulation ratings. A single 6J5 with the EL34 in a mu stage will also do the job, but will definitely need another stage ahead of it.
Note that I unfortunately never got around to building the actual amplifier. YMMV, etc.
The Mu-stage is built on top of a pentode based CF, the mu-follower is built on top of a similar/identical triode CF.
I'm not quite sure whether Allen Kimmel discovered or rediscovered the idea. I believe he was probably the first to recognize and discuss its merits for audio usage. A lot has been written about the merits of pentode CFs in old college engineering text books and elsewhere.
Using a pentode as a CF has some advantages, the generally much higher transconductance results in much lower output impedances which results in nearly ideal ac behavior of the ccs aspect of the mu follower mode of operation. (Improved bootstrapping of the plate load resistor due to CF gain much more closely approaching unity and the lower losses due to lower source Z.) In simulations and in some limited bench testing this seemed to result in a significant improvement particularly in large signal linearity.
Here is the original article by Allan Kimmel, well worth the time to read imho.. http://www7.taosnet.com/f10/mustage.html
JMTCW... YMMV
I'm not quite sure whether Allen Kimmel discovered or rediscovered the idea. I believe he was probably the first to recognize and discuss its merits for audio usage. A lot has been written about the merits of pentode CFs in old college engineering text books and elsewhere.
Using a pentode as a CF has some advantages, the generally much higher transconductance results in much lower output impedances which results in nearly ideal ac behavior of the ccs aspect of the mu follower mode of operation. (Improved bootstrapping of the plate load resistor due to CF gain much more closely approaching unity and the lower losses due to lower source Z.) In simulations and in some limited bench testing this seemed to result in a significant improvement particularly in large signal linearity.
Here is the original article by Allan Kimmel, well worth the time to read imho.. http://www7.taosnet.com/f10/mustage.html
JMTCW... YMMV
Hi Kevin,
I didn't use a D3a but a instead of that a 7788. The D3a couldn't handle the needed voltage swing on the cathode.
I made and tested this design myself because I wanted to rebuild a Chinese 845 amp to a mu-stage design. The reason I don't want to put this schematic on this forum in public, is that I never really heard it play. It stayed by measuring. I don't want people to build this thing and get dissapointed. Someone with experience can value the circuit and decide to build or not building the thing....
211audio
I didn't use a D3a but a instead of that a 7788. The D3a couldn't handle the needed voltage swing on the cathode.
I made and tested this design myself because I wanted to rebuild a Chinese 845 amp to a mu-stage design. The reason I don't want to put this schematic on this forum in public, is that I never really heard it play. It stayed by measuring. I don't want people to build this thing and get dissapointed. Someone with experience can value the circuit and decide to build or not building the thing....
211audio
Apology to dr.strangelove3 - tone of 1st paragraph of earlier my post was not really what I intended and now too late to edit. Duh! 
Hi 211audio,
Almost fixed my earlier post, oops, thanks for catching that. I have never listened to mine either, but with sufficient simulation and follow up bench work there is no reason why it couldn't work and sound good.
Yes the D3A voltage rating would be rather inadequate for the kind of swings required. 7788 seems like a good alternative.
You could post the schematic here with the warning that you are not sure how well it works in practice, and for some potential critique as well.
Of course maybe the best thing for dr.strangelove3 to do would be to use tubelabs powerdrive circuit and be done with it. I know that will drive an 845 right into A2 with good results.
Hi 211audio,
Almost fixed my earlier post, oops, thanks for catching that.
I have never listened to mine either, but with sufficient simulation and follow up bench work there is no reason why it couldn't work and sound good.Yes the D3A voltage rating would be rather inadequate for the kind of swings required. 7788 seems like a good alternative.
You could post the schematic here with the warning that you are not sure how well it works in practice, and for some potential critique as well.
Of course maybe the best thing for dr.strangelove3 to do would be to use tubelabs powerdrive circuit and be done with it. I know that will drive an 845 right into A2 with good results.
@ 211audio,
Thanks for you emails and the clarification regarding mu-stages. I just replied back to your email.
@ Kevin,
Also thanks for clarifying. Thanks for the apologies. I already thought you probably meant it another way
A cascode mu-stage sounds VERY interesting...but I'm afraid for me it is a bit out of my capabillities. I am indeed also considering the tubelab approach. It is very buildable and not very expensive.
At the moment I'm ónly in the process of collecting schematics. I hope to start building something this year.
Thanks for you emails and the clarification regarding mu-stages. I just replied back to your email.
@ Kevin,
Also thanks for clarifying. Thanks for the apologies. I already thought you probably meant it another way
A cascode mu-stage sounds VERY interesting...but I'm afraid for me it is a bit out of my capabillities. I am indeed also considering the tubelab approach. It is very buildable and not very expensive.
At the moment I'm ónly in the process of collecting schematics. I hope to start building something this year.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.