I am still in if it is t7 ltp build. It will work well. While I don't have iron on the way yet, my tubes will arrive tommorrow. So I am itching to get the iron ordered.
Yes, apology folks for a few days absence because of you do not want to know. As said before I did a number of calculations as well as tests, with which I will come forward ... should be by tomorrow-night. (It is now Friday 19th, 13:43 SA time.)
These were based on using a double triode LTP in the present schematic (or at least the one a week ago!) Addition of FETs will change matters, but that afterwards then.
I also agree 100%. The early 5-20 design with the CCS appears to be sound, as well as the early PS. I'm sure there will be a few tweaks once a few amps get breadboarded/built.
It sounds like Johan will be posting his latest findings shortly.
Also, the Hammond 374BX looks like it may be a higher voltage alternative (375-0-375@201ma) to the Edcor, although the bias tap is only 50V. I'll stick this in the model later today (hopefully).
If Chrish and Tubelab don't mind, I think we should continue the 6GK5-MOSFET-AB2 design in the http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp.html thread, as it appears to be a continuation of that discussion.
I am very interested in a FET coupled AB2 KT88 triode design, but as mentioned, we are far way from TubeMack's original design goals.
It sounds like Johan will be posting his latest findings shortly.
Also, the Hammond 374BX looks like it may be a higher voltage alternative (375-0-375@201ma) to the Edcor, although the bias tap is only 50V. I'll stick this in the model later today (hopefully).
If Chrish and Tubelab don't mind, I think we should continue the 6GK5-MOSFET-AB2 design in the http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp.html thread, as it appears to be a continuation of that discussion.
I am very interested in a FET coupled AB2 KT88 triode design, but as mentioned, we are far way from TubeMack's original design goals.
50V Bias will work fine if bridge rectified.
AB1 AB2 or even MOSFETless are really simple tweeks between them, especially in a breadboard set up. I am with boywonder in desire, but for right now we should focus on the most KISS solution per TubeMack's original intentions with the thread.
I am curious as well on Johan's future posting.
AB1 AB2 or even MOSFETless are really simple tweeks between them, especially in a breadboard set up. I am with boywonder in desire, but for right now we should focus on the most KISS solution per TubeMack's original intentions with the thread.
I am curious as well on Johan's future posting.
Just to be clear guy's, i'm not total against the 6GK5 being used in place of the EF86, but I do have reservations. One is that the somewhat rare 7 pin socket that would make future tweaks using other 9 pin triodes very hard. And, the biggie is that Johan has spend time to breadboard the amp with the EF86. If it gets changed, we lose all his findings, and good numbers. I feel having these findings is a great asset to the project.
Thoughts?
Thoughts?
Last edited:
I am interested to see what Johan says. Good data is hard to beat.
I am curious in the amount of feedback he had to use to tame the gain for the t7 ltp.
I am curious in the amount of feedback he had to use to tame the gain for the t7 ltp.
TubeMack: 7 pin sockets are a dime a dozen. Apex Jr. has NOS phenolic for .35/ea. Or you can go uptown and get ceramics for a couple bucks.
Tube Scokets
Sgregory: 50V x 1.41=70.5V........to meet the needs of the present schema we need 80V, although 70V may work....what's the max bias V that we need?
Tube Scokets
Sgregory: 50V x 1.41=70.5V........to meet the needs of the present schema we need 80V, although 70V may work....what's the max bias V that we need?
Personally, I like the EF86. However, if you go with the 6GK5 and decide you like to try a pentode, there's always the 6AU6. Tek seemed to like them.
Unfortunately you can't do that with a Hammond. The 374BX has a 375-50-0-375 winding. The bias tap is only on one side which forces simple half wave rectification. That is ok for applications where you don't need any current on the negative supply.
I don't care, but both of our amps seem settled on 6SN7's right now, but Chris hasn't built anything yet, and mine is still in the breadboard stage. The thread got started as a paper "what if" design. Eli made his usual insistance on the use of a 12AT7 in the signal chain somewhere (I happen to like the 12AT7). With any "what if" project there will be several people voicing their strong opinions.
I built a 300B based amp about 10 years ago that was loosely based on Kevin Kennedys design. His amp used an LTP (6SL7) followed by another LTP (6SN7) to drive push pull 300B's. Mine used a 5751 followed by a 6CG7. I had already started on a redesign of that amplifier using the same tubes when this thread came along. A prototype driver circuit existed, so I wired it up to some 6L6GC's. This prompted a few experiments that resulted in a running amplifier using 12AT7 - 6CG7 - 6L6GC amp that made 75 watts without violating the tube specs.
For the first 121 posts the 12AT7 was part of the picture. In post #122 Chris mentioned wanting to do an all octal amp. I have been receiving requests for an octal amp for years. In post #123 I noted that my existing design would be quite similar to Morgan Jones' Crystal Palace amplifier if 6SN7's were used. Curiosity led me to make one, then another. The result is the breadboarded amplifier that I now have.
Chris had drawn the proposed schematic of his amp as it would be built. Whether he builds it as planned, or changes the plan is up to him. I plan to make at least two complete amps using the octal driver board that I have laid out, maybe more (it sounds really good). I still have the original 12AT7 - 6CG7 board that will likely be used for at least one screen drive amp.
I built a little all pentode 6AU6 - 6AQ5 amp about 2 years ago. I never got very far with it (10 watts just didn't excite me, I guess). I bought one of Pete Milletts "engineers amp" boards last year, and I started hot rodding it (200 WPC!). I have been so impressed with it that I will probably experiment in that direction next.
Boywonder, I guess what I meant was that the 7 pin socket only leaves one with 7 pin tube options for the future, and that seemed more limiting than 9 pin options.
Chances are that any 9-pin swaps will have a different pin-out anyway, so you will likely still have to rewire the socket when you want to try a different tube. I guess it saves you having to punch/drill the chassis again. Are there no 9-pin triodes with high enough mu, like 12AT7 or (horror) 12AX7?
Anyway, it's your thread!
Anyway, it's your thread!
In my build I will probably leave larger chassis holes for the driver + LTP, and mount a plate under that. In which case I can later experiment w/Octal or whatever looks interesting.
Another plus for the EF86 would be that it is still in production. If you are designing something partially so that it can be duplicated easily by others, it would be helpful that it can be purchased easily from any tube retailer. Same is true for the 12A*7 types.
boywonder, -70V should be enough to bias the kt88. Depending on bias current and anode voltage somewhere between -50 and -60 is reasonable. However based on tubelab's comment it may be an issue with the Hammond xfmr.
Perhaps you need to rethink things. SY pointed out how non-linear an EF86 will be, when combined with the small load resistance a good overall gain structure requires. Change the LTP type to ECC99 and an EF86 may be good. However, all bets are off as to how the net HD spectrum will come out.
As I firmly believe the 12AT7 LTP to be an excellent choice, a linear triode of modest gain and respectable gm, that works reasonably well with 100 V on its plate, needs to be found. Check the 12BH7 data sheet out. CCS loading the anode at 7 mA. in combination with a single red LED for bias looks OK to me. Yes, the 12BH7 is in production. 😉 Wire 1 monoblock to use section 1 and the 2nd monoblock to use section 2. When wear becomes evident, exchange the tubes and double the service life.
No I didn't see Sy's comment. 12BH7 , Hmmmm. I haven't seen this one mentioned, but i'm familar with it. Any one have any interest in that tube for the voltage amp?
Last edited: