Is anyone familiar with Bob Cordell's C70 KT-88 amp design? (link: https://linearaudio.net/sites/linearaudio.net/files/v2 bc.pdf)
I've been in to tube guitar/bass amps for while, but looking to build a HiFi amp.
I like the KT-88, although could also run it singled ended and still get enough power. I'd look to add some solid state input switching and volume control.
Any feedback on the above design, or suggestions for alternates that I could look into would be appreciated.
Cheers
I've been in to tube guitar/bass amps for while, but looking to build a HiFi amp.
I like the KT-88, although could also run it singled ended and still get enough power. I'd look to add some solid state input switching and volume control.
Any feedback on the above design, or suggestions for alternates that I could look into would be appreciated.
Cheers
What else could Bob design, but one more OpAmp-like amp, this time with tubes? And he designed it to compare tube sound to solid state sound in listening tests? Oh my God! 😀😀😀
That driver stage topology usually sounds good, though. Input LTP driving diff-amp driving push-pull output stage. The Class AB2 PP 6L6GC circuit that Tubelab (George) developed is similar, but without the extra feedback compensation Zobel networks.
One nice part of Bob C's plans is that he gives values for using Dyna A470 OPTs, which are readily available, either the new reproduction or from old Dyna Stereo 70 amps.
I wonder if using a 12AT7 as the input LTP would be an improvement... (Higher gm there to better drive feedback loop?)
One nice part of Bob C's plans is that he gives values for using Dyna A470 OPTs, which are readily available, either the new reproduction or from old Dyna Stereo 70 amps.
I wonder if using a 12AT7 as the input LTP would be an improvement... (Higher gm there to better drive feedback loop?)
@Wavebourn - sounds like there's some history there that I need to get caught up on...
@rongon - thanks for the comments - a lot in there for me to unpack 🙂 Guitar/Bass instrument amps are A LOT simpler in that you're looking for the circuit to colour the sound, rather than trying to faithfully reproduce the input signal. Can you clarify what you mean by "LPT"? Also re: OPT's - I've been reading that toroidal designs offer better audio performance - is this something I should consider? or better to stick with the tried/tested A470s?
Thanks both for the responses.
@rongon - thanks for the comments - a lot in there for me to unpack 🙂 Guitar/Bass instrument amps are A LOT simpler in that you're looking for the circuit to colour the sound, rather than trying to faithfully reproduce the input signal. Can you clarify what you mean by "LPT"? Also re: OPT's - I've been reading that toroidal designs offer better audio performance - is this something I should consider? or better to stick with the tried/tested A470s?
Thanks both for the responses.
Not that much simpler: remove the right half of V1, and you have a Mullard circuit; remove both halves of V2, and you have the classic Fender circuit, like the Bassman 5F6.
I like this topology a lot: it lends itself to many variations, and the LTP's (long tail pair, or differential amplifier, V1 and V2 in the circuit you linked) recover well from overload when used as an instrument amp. Bob Cordell's contribution to this classic topology is to add constant current sources (Q1 & Q2), DC & AC balance, some local feedback in V2, and partially stabilized power supply. I'd question the use of bipolar transistors in Q1 & Q2, but otherwise it's all good.
Toroidal OPT's have much better high frequency response, but they are very sensitive to the DC balance of the output stage. Maybe not for a first build. There're many fine OPT's like Sowter, Lundahl, Sansui, etc, it all depends on how you're getting along with your banker.
You can read Bob's book to get what he is up to. He is a very good designer, one of the best, but he is not onto vacuum tubes, he does not value them properly in his writings, his passion is solid state "opamps" with FET outputs. He is a member of this forum, but participates in solid state discussions, mostly about his book.
I totally agree with his view that he expressed during BAF about the better the amp behaves, the less it sounds, but disagree about what means "misbehaving", and what means "sounding". I go by a different direction, for better results, as I understand them, and his attempt to design the best tube amp to compare to his best solid state amps I understand, but the only thing that he compared, was his tube opamp-like power amp to his solid state opamp-like power amp. It can not be generalized to all tube amps and solid state amps.
I did not mean that the amp was poorly executed. It is designed very professionally, using modern schemo-technic approach, by one of the best designers on the planet. But still, if you make a food processor using Kalashnikov machine gun blueprints, you would get a machine gun, right?
It already behaves pretty well, but not as an amp can behave fooling imagination. It does not matter which active components you use; what matters, is the end result. The amp has to fool imagination during listening, not during measuring according to specs.
I did come up with something quite similar a little over ten years ago. The entire amp was designed by myself and chrish in Australia over a period of about a year. All the details of how and why for most everything are laid out in one long thread, and worthwhile reading for someone wanting to understand the design. At the time chrish wanted to use 6SN7's in his design. I built several prototypes, and found that a 5751 for the first tube, and a 12BH7 or 5965 for the second tube were my favorites. It is here:
6L6GC AB2 Amp
Work, life, and other minor details left my amp breadboard on the shelf in a closet for 6 or 8 years. Fast forward to 2014....my 41 year career as an engineer at Motorola has ended and I find myself in temporary housing while my "retirement home" is being built. I take the info from that old thread, and my notes and lay out a new version of the driver board, and get some professionally made since I have no way to DIY boards. The package from the board house sits on a shelf, unopened for a year or so.
During that time, I am busy clearing land and dealing with the details of getting a house built. Then I set out to build a new lab in the basement of the new house.....then I find the box with the boards.....and the fun starts.
I decided that my breadboard from 2008 was junk, so I make a new one. it's much like the old one, but real plywood instead of Masonite so it can hold some real OPT's. Several test runs happen with everything from 50C5's to 4D32's, and power levels from 10 watts to 200+.
Someone sends me an email asking if he can buy a pair of the driver boards.....before long the entire package of boards are gone, so I get some more.
I don't know of Bob Cordell, or his designs. There are some strong similarities, between the two amps, and between the amps and an opamp. My design originated in 2007. I am assuming his is from later. It is fully differential. It does use solid state CCS chips in the tails of the LTP's much like an opamp would. I also use a mosfet follower between the last LTP and the grids of the output tubes. This is to eliminate blocking distortion and allow the tube's grids to be driven positive if needed. There is the usual pair of differential inputs, and it IS possible to connect a GNFB loop from the speaker output to the unused input. I do not do this on most of my amps. I prefer to run the output tubes in triode mode, so that the entire amp can run open loop....no more opamp!
Not too long ago I made a version with KT88's in triode. It can be seen here....but this video was shot in one take with a $179 cell phone. Not exactly HiFi, nor was it meant to be, but still not a bad video.
YouTube
I started a thread detailing these boards and their use here:
Tubelab Universal Driver Board, 2015 version
The boards will be added to my web site with build-up info and full instructions once the TSE-II redesign is finished.......soon.
This morning I grabbed a pair of Edcor OPT's from the swap meet section of this forum, giving me the final push (and parts) needed to turn the prototype seen on YouTube into a complete amplifier.
A well designed Fully Differential OTL amplifier can be considered a serious all tube fully differential power opamp with serious frequency response, only one gain stage, under 1% TDH, low output impedance if you use 4x 6C33C or similar tubes and if you use enough gNFB you can get under half an ohm output impedance.
Just check out Atma-Sphere OTL amps. OK... one drawback... it generates lots of heat.
But hey.. it's like a fireplace.... gives good heat, some glow and nice sound 😛
Just check out Atma-Sphere OTL amps. OK... one drawback... it generates lots of heat.
But hey.. it's like a fireplace.... gives good heat, some glow and nice sound 😛
Single ended amps are better for fireplaces.
And for sound, of course. If I need some opamp, I would take some IC designed by Scott Wurcer, for example.
It's a rather involved project, with performance higher than you would expect from a tube amp, partly because of the well executed power supply system. If you decide to build, I have the layout files for a power supply and driver stages PCB. PM me.
Jan
Last edited:
I assume that you meant LTP. An LTP is a pair of active gain devices, tubes, Bipolar Junction Transistors, or mosfets with a common lead (cathode, emitter, or source) tied together and usually to a Constant Current Source or large valued resistor. This forms a differential pair where the two devices operate out of phase and, if properly configured cause much of the even ordered distortion products to cancel. Two such LTP circuits are cascaded in this design. Several are usually used in an opamp to maximize the open loop gain.
A 12AT7 and many other TV tuner tubes (6BQ7, 6BK7....) that often generate some even ordered distortion work great in LTP configuration. I have used the 12AT7 and the 6BQ7 in the first LTP in my boards.
I'm not sure if you are still curious about this design but I'm in the process of building it. I already tested the boards and now I'm assembling it into an enclosure. The entire process can be seen in this thread: https://www.diyaudio.com/community/threads/kt-88-amplifier.381883/page-4#post-7118538
I'll update the post when everything is assembled and running. I hope this will be sometime during this week or at least next week.