Hey guys,
I've become interested in this project, and hope to one day build one (after my never-ending home renovation is complete and my construction workshop is converted back to an amplifier/speaker "lab".) In the meantime I'd like to contribute by doing what I enjoy second-most. I'm an ex-technical writer and technical illustrator, so I see an opportunity here to help out with some documentation.
Something I've been looking for is a well documented project, and there doesn't seem to be too many of those here in tube world. This project looks like a great candidate to become one of those. I'm new to tubes and I've discovered a great way to gain a better understanding of new subjects is to draw and write about them (at least until said reno is just a distant memory and I can resume burning myself and buying parts.)
If you guys are game, send me all your latest schematics, BOMs, specs and plans and I'll compile them and turn it into a project that others can follow (eventually). Once the amp design is finalized and a couple amps are built and debugged, a "constructor's thread" can be started for others to get help building this.
Oh, I also think this amp needs a catchy name. Got one in mind? 🙄
... Todd
PM me for my email address.
I've become interested in this project, and hope to one day build one (after my never-ending home renovation is complete and my construction workshop is converted back to an amplifier/speaker "lab".) In the meantime I'd like to contribute by doing what I enjoy second-most. I'm an ex-technical writer and technical illustrator, so I see an opportunity here to help out with some documentation.
Something I've been looking for is a well documented project, and there doesn't seem to be too many of those here in tube world. This project looks like a great candidate to become one of those. I'm new to tubes and I've discovered a great way to gain a better understanding of new subjects is to draw and write about them (at least until said reno is just a distant memory and I can resume burning myself and buying parts.)
If you guys are game, send me all your latest schematics, BOMs, specs and plans and I'll compile them and turn it into a project that others can follow (eventually). Once the amp design is finalized and a couple amps are built and debugged, a "constructor's thread" can be started for others to get help building this.
Oh, I also think this amp needs a catchy name. Got one in mind? 🙄
... Todd
PM me for my email address.
Glad to see more come on board! I'm afraid I don't have much yet. The last posted Main, and PS schematic by me is it. The PS appears to be about done with most parts listed above by wonderboy. The main amp is also about done, but no parts have been decided on. I may need more help with this. I'm also unsure about the OT specs. I'm thinking about 100W rating, 40% UL tap, but I have not yet learned how to spec the Impeadance. I'm still unsure about the CCS, as there hasn't been any strong comfirmation on it yet. There is also the question of a possible sub of the EF86 with a 6GK5? I've been resorting to rereading my books!
Taj, document away! I see many many new guy's comming to the forum looking to built KT88 monos. Currently there is no FULL documentation to do this. Like me, many have to resort to building some variation of a ST70 due to the lack of comlpete documentation. I'd love to have a part in a change here. The great thing about this amp is that it first saw light a few years ago by another forum member, and many members here helped with it's current form. It has had a long birth, first by Mullard, then redone by Claus Byrith with yet another redesign here on this forum buy several experts. My main contribution has been the decision to break it up into monoblocks. I can send you bigger files for the schematics.
Hmmm. A name.........? I think something classy from classical music I reckon.....
Taj, document away! I see many many new guy's comming to the forum looking to built KT88 monos. Currently there is no FULL documentation to do this. Like me, many have to resort to building some variation of a ST70 due to the lack of comlpete documentation. I'd love to have a part in a change here. The great thing about this amp is that it first saw light a few years ago by another forum member, and many members here helped with it's current form. It has had a long birth, first by Mullard, then redone by Claus Byrith with yet another redesign here on this forum buy several experts. My main contribution has been the decision to break it up into monoblocks. I can send you bigger files for the schematics.
Hmmm. A name.........? I think something classy from classical music I reckon.....
Last edited:
A 100 W. rating is correct. You need to allow magnetic headroom for a low freq. NFB error correction signal.
I not expert on the primary impedance subject, but 8 K for "12" W. tubes, 6.6 K for 6L6s and 7591s, 4 K for EL34s, and a bit less still for KT88s seems to do quite well. The Edcor model CXPP100-MS-3.3K will get the job done, at something resembling reasonable cost.
Eli, While your at it, what are your thoughts on the EF86 compared to thr 6GK5 you have reccomended in the past for voltage drivers on LTP amps
I strongly favor high gm types in the small signal circuitry of amps, which employ loop NFB. High gm provides protection against HF error correction signal induced slew limiting.
The EF86 is a VERY nice small signal pentode, perhaps the best ever, but it's hardly high gm. OTOH, the 6GK5 is a very slick combination of high μ, high gm, and low RP. Yes, the curves are bunched up, when large negative voltages are applied to the grid, but the curves look pretty darn good in the area of a 2 VRMS I/P signal. Poindexter has taken to using 6GK5s in the small signal circuitry of his "Musical Machine". That (IMO) is a positive endorsement.
Most projects I've seen specify something between 4,5k and 5,5k. If understood correctly, within the viable range, the less impedance you use, the more power you get but the more quality you loose, too. Correct me if I'm wrong.
I'm on that. Have to finish the amps first, have them working, complete the documentation, translate it and THEN ....
...we can all state together that my ugly build doesnt work as a model. 😛
Stradivirus? 😀
I'd call it "KTulu Amp" or "KT Lord". It's not easy to make up a word play with "KT"... The KT = good thing is the only thing i can think of.
TubeMack: see the KT88 tube data from Duncan amps TDSL here:
TDSL Tube data [KT88]
Look at the application data, it gives power out, distortion, etc, for various topologies, B+ voltages and anode loads. Focus on the UL data, since this design is UL.
Ia is the anode current (which=bias current in triode, and is close to the bias current in the other topologies, except g2 current needs to be factored in), Rk is the cathode R for cathode bias; in this case we are using fixed bias, so we don't need to worry about Rk......except that since there is no (actually negligable) voltage drop across the cathode R we can lower the B+ a little to compensate.
As mentioned above, increasing the OT primary impedance lowers power out and distortion, and reducing the OT primary impedance increases power out and distortion.
Eli: I have built two iterations of Poinz' 6GK5/EL34 MM, but I am too green to have any idea about dropping in the 6GK5 in this design...any input would be greatly appreciated. I suppose that I should draw a couple of load lines, one for the EF86 and one for the 6GK5 operating in it's happy place.....
Last edited:
Classy classical names huh? I won't be of much help there, I'm a jazz/blues guy who finds most classical music annoying (not that there's anything wrong with it).
The only name I can think of in that genre is "Nocturne" which does actually have a nice evening glow to it.
..Todd
The only name I can think of in that genre is "Nocturne" which does actually have a nice evening glow to it.
..Todd
Naming amps after pathogens... nice. I never would have thought of that. How about acidophilus for us jazz folks? 😉 Or a combination; Stradivirus Acidophilus (the violin virus, yogurt bacteria "superbug" amplifier). Maybe we could throw some Saccharomyces (yeast) in with it to really start it glowing.
..Todd
Last edited:
Let's really go for the gusto and CCS load the 6GK5 with a 10M45S. A horizontal load line kicks butt. Look at the 6GK5 data sheet, again. A red LED for bias (between 1.5 and 2 V.) and IB = 8 mA. looks good. Don't worry about clipping at the grid. The NFB loop prevents it. Thank SY for that hint.
This "beast" will have plenty of gain. I would not be at all surprised to find a "unity" gain preamp is best.
Lovely... The Girl from Treponema.
I also like SY's theory that Japanese names are trendy for audiophile gear. Thus, as one who studied Aikido (the martial art, not the preamp) I can suggest that Senshusei would be a good amp name. If you've seen the documentary called "Hell Week" about Navy Seal training, then imagine someone being trained the same manner in Aikido for a solid year -- the equivalent of Hell Year. That's a Senshusei (pronounced Sen' shoo say).
Senshusei are a lot more powerful than Aikidodoka. Very well respected, as in "please don't kick my butt, sir, I am but a lowly 5th dan".
Osu!..Todd (5th kyu 🙄)
Last edited:
Now who's gonna draw this up?
What is comes down to for me is SQ. could this exercise trump the EF86 in SQ I wonder?
Please, no viral names! Depending how how it sounds, it may be called the McInturd.....
Last edited:
i'm interested. just finished a build of a vacuum stereo amp and would like to build something interesting next, this seems to fit the bill... 🙂
How about kappo as in the "kappo build". Kappo means resuscitate in japanese and specifically means to resuscitate a fighter using acupuncture techniques (applied when a fighter gets knocked out or otherwise becomes unconscious).
I like the poetic beauty of resuscitating a tube designs from old and kappo seems to be applicable..
How about kappo as in the "kappo build". Kappo means resuscitate in japanese and specifically means to resuscitate a fighter using acupuncture techniques (applied when a fighter gets knocked out or otherwise becomes unconscious).
I like the poetic beauty of resuscitating a tube designs from old and kappo seems to be applicable..
Just curious, I know this is designed to run UL but would it be that difficult to wire it with a switch between triode and UL? Also, what kind of power output can be expected in both triode and UL mode with this design?
Now who's gonna draw this up?
I think if I'm going to draw this up, someone had better give me a scratch version to follow. Not much of this insane-voltage stuff makes much sense to an old SS guy (at least in terms of execution, anyway). I wouldn't know how to wire up a10M45S CSS. (In fact I'm off now to Google this part number). 😎
..Todd
BigJP: A UL/triode switch would be do-able.
As a semi-newbie SWAG, KT88 triode is good for 25-30 watts and UL around 50W assuming a B+ of 460V. It looks like 4K OT primary would be the way to go for this. Duncan's TDSL for the 4K UL topology shows a B+ of 560V yielding 70W, so this would be scaled down a bit for the reduced B+.
TDSL Tube data [KT88]
Taj/TubeMack: For the 6GK5 with a 10M45S CCS, check out Poinz' schematic in posts #6 & #7 here:
http://www.diyaudio.com/forums/tubes-valves/136562-pp-el34-6l6-circuits.html
Regarding sensitivity to load impedance, generally the decrease in output and increase in distortion occurs quite sharper going lower than optimum, than when going higher.
There is a good explanatory graph for the KT88, of how several parameters change when the screen tap goes from 0% - 100%, in the book: "An approach to Audio Frequency Amplifier Design", originally from the G.E.C. of Englend, fig. 1-1, p3; reprinted by Audio Amateur Press, New Hampshire. (The book also contains further seldom found information for the use of the KT** tubes.) I also recall seeing the graph in an article by D R Grimwood on Optimal Ultra Linear Design, which I am unable to find now - memory serving it is the second figure in that article. (The writer reveals several other finer points about the mode but, respectfully, goes off track further down in his epistle.)
To summarise, the KT88 assumes almost triode behaviour from a tap 40% and higher, while still exhibiting pentode efficiency and output up to about 43% tap. This graph is the only one that I recall seeing showing such information; a pity. But it is typical of the behaviour of most beam tubes when moving the screen tap. Pentodes (e.g. EL34) behave roughly similar, but do not show the increase in available output with the lower screen taps. The µ[G1 - G2] plays a prominant role.
Regarding providing a switch for UL/triode; in my view the advantage of that is limited. (At least one well-known amplifier brand admitted that they installed that purely for user-attraction, not worthwhile sonic advantage.) [Incidentally, such a switch should not be used when the amplifier is powered up. Depending on the design (especially in amplifiers with NFB), high enough transients could be generated to destroy the output transformer.]
There is a good explanatory graph for the KT88, of how several parameters change when the screen tap goes from 0% - 100%, in the book: "An approach to Audio Frequency Amplifier Design", originally from the G.E.C. of Englend, fig. 1-1, p3; reprinted by Audio Amateur Press, New Hampshire. (The book also contains further seldom found information for the use of the KT** tubes.) I also recall seeing the graph in an article by D R Grimwood on Optimal Ultra Linear Design, which I am unable to find now - memory serving it is the second figure in that article. (The writer reveals several other finer points about the mode but, respectfully, goes off track further down in his epistle.)
To summarise, the KT88 assumes almost triode behaviour from a tap 40% and higher, while still exhibiting pentode efficiency and output up to about 43% tap. This graph is the only one that I recall seeing showing such information; a pity. But it is typical of the behaviour of most beam tubes when moving the screen tap. Pentodes (e.g. EL34) behave roughly similar, but do not show the increase in available output with the lower screen taps. The µ[G1 - G2] plays a prominant role.
Regarding providing a switch for UL/triode; in my view the advantage of that is limited. (At least one well-known amplifier brand admitted that they installed that purely for user-attraction, not worthwhile sonic advantage.) [Incidentally, such a switch should not be used when the amplifier is powered up. Depending on the design (especially in amplifiers with NFB), high enough transients could be generated to destroy the output transformer.]