Very succinct responses. 
I am always (ok, usually) interested in learning more. Can you point me to any data on the pentode v.s. triode results?
Unless you know of a suitable dual pentode I will have to find room for an additional tube and start drilling which makes it a bit more complicated than changing bias method. Of course I may have to figure out a new feedback resistor value but that should be doable.
I am always (ok, usually
) interested in learning more. Can you point me to any data on the pentode v.s. triode results?Unless you know of a suitable dual pentode I will have to find room for an additional tube and start drilling which makes it a bit more complicated than changing bias method. Of course I may have to figure out a new feedback resistor value but that should be doable.
Very succinct responses.
I am always (ok, usually
Do you mean textbooks for dummies, about Ri of tubes, parallel connections of load resistor and Ri, and about voltage dividers?
Step 1: remove LED, put resistor there. Read still the same textbook and figure out how higher will be resulting output resistance of the stage (easy to calculate!), and how it's impact on feedback ratio will be less.
Good luck!
Just to let you know that I have just uploaded another Baxandall design- a simple preamp made with one double triode. (per channel, that is)
The Wireless World Archive
The Wireless World Archive
ooops, I think we managed to miss each other's meaning. What I was looking for is either some experimental data on the differences in output signal integrity (distortion in the broadest sense from input to output) for triode drivers v.s. pentode drivers in this circuit and/or some qualitative analysis of the effects of the more stable FB ratio v.s. the inherent non-linearity of the open loop pentode's output.
As to your why question, I thought you were asking why redoing the design with pentode drivers would be more complicated than changing the bias of the triode driver design. As you state, changing the bias from LED to resistive is simple. My point was that changing from single 9 pin dual triode for drivers to a pair of 7 pin pentodes is a bit more labor intensive in modifying the chassis. Calculating the new feedback resistor for the same feedback ratio would not be too difficult but the open loop gain of the pentode will be much higher too which may mean that either a larger feedback ratio will be required or possibly a reduction of gain in the preamp. Obviously not impossible but definitely more work. I suppose I could enlarge the hole intended for the 6N1P for a 12 Pin Compactron socket and use a 6BV11 twin pentode instead of finding a way to put in the two 7 pin mini sockets in place of the 9 pin but the plate curves on that one look a little tricky to work with.
As to your why question, I thought you were asking why redoing the design with pentode drivers would be more complicated than changing the bias of the triode driver design. As you state, changing the bias from LED to resistive is simple. My point was that changing from single 9 pin dual triode for drivers to a pair of 7 pin pentodes is a bit more labor intensive in modifying the chassis. Calculating the new feedback resistor for the same feedback ratio would not be too difficult but the open loop gain of the pentode will be much higher too which may mean that either a larger feedback ratio will be required or possibly a reduction of gain in the preamp. Obviously not impossible but definitely more work. I suppose I could enlarge the hole intended for the 6N1P for a 12 Pin Compactron socket and use a 6BV11 twin pentode instead of finding a way to put in the two 7 pin mini sockets in place of the 9 pin but the plate curves on that one look a little tricky to work with.
You're kidding, right?
Expensive? Vacuum Tubes Radio Tubes lists 6BQ6GAs for $2.00 a piece. (Price doubled since I started talkin' 'em up) and 12BQ6GAs go for $1.00. Priced those new SiC JFETs from SemiSouth lately? Oy vey! How about lateral MOSFETs? The going rate for these is about $9.00 a pop. OnSemi's Thermaltrak BJTs are somewhat more expensive than that. You don't have to pay audiophool silly prices for good audio finals, and the *BQ6GAs make mighty fine audio finals.
Unreliable? Mine are still going good after five years of nearly daily use.
Noisy? Not that I can tell.
Non-linear?
Potentially dangerous? Only slightly moreso than high-powered solid state rigs that use rail voltages of 100V and up. (This would exclude such exotic designs such as 845 SETs, or amps north of 100W, of course.)
Inefficient? OK, I'll grant you that. It takes watts to light 'em up.
I started out as a solid state guy myself, and have some designs that sound better than any store bought amps I've ever heard. They're quite close, but just don't quite get there.
I'm also not a sand-o-phobe, and make use of solid state for support such as CCSs, power supplies, and source follower grid drivers. That gets you better performance than you could have gotten back in 1957. I'm also convinced that NFB is not the Spawn of Satan either.
Gonna double again methinks.
..todd
Better that than they decide that it's not worth the storage space, and they all get dumped into a landfill somewhere. Still, even $4.00 a pop is onehelluva bargain, as compared to the prices of new release 45s, 300Bs, KT88s, etc (and let's not even discuss NOS) or even the really decent transistor finals, like those SiC JFETs, or lateral MOSFETs. (These days, the premium is on switching/digital/quasi-digital performance, not linear performance. Vertical MOSFETs vs. laterals, for example. Or those IGBTs that were never intended for linear amplification, and whose audio performance is anyone's guess.)
Thank you Douglas; but today tubes are used mostly for High-End. High-end and piezoelectric pickups, high-end and huge R in screen grids, high-end and equalization in feedback, are not compatible. Of course, such (or even better) results is better to get using something like TI 1612 chips.
That's all quite true, but Bax wasn't designing for the high end. He wanted a functional consumer item designed for a general public which didn't necessarily listen closely. If you're the type to put on some music while you're doing something else, then it's good enough.
If you're not that type, then you're probably here. Otherwise, you're probably not.
Although you do learn something new. I wasn't aware that Gramophones had outputs: you just wound 'em up and the sound came out the big wooden horn.
For this purpose $350 Denon receiver with Audissey digital processor serves me like a charm!
Why should I waste tubes for such quality? I plan to marry Audissey with tubes, it will be a killing toy! Planning and drooling... 5 x 12J1L+12P17L drivers and Semisouth JFET outputs arranged as augmented source followers loaded on counter-modulated CCS... 6S17K-V phono preamp...
Well let's try to revive it.
I have just uploaded another vintage Wireless World article, on a 400W valve amplifier using KT88s.
The Wireless World Archive
Enjoy
I have just uploaded another vintage Wireless World article, on a 400W valve amplifier using KT88s.
The Wireless World Archive
Enjoy
Interesting- why would he directly couple the cathode follower to the driver, then capacitively couple to the finals? Typically, when an amp clips, the recovery is much faster with the opposite connection- in this design, half the virtue of the CF is going unrealized.
Not that one ought to be clipping a 400W amp...
Not that one ought to be clipping a 400W amp...
"why would he directly couple the cathode follower to the driver, then capacitively couple to the finals? "
Agree. But one could use the 6AL5 baseline restorer circuit on the final grids that Bob Carver came up with.
--------------------
"we here in Tube forum" "...don't build 400W amps"
George has been up to 250 Watts on the little Red Board with just two Sweep tubes. And 500 W with both channels paralleled. I wouldn't count on it for an Apollo Moon mission though. But using Sweeps, instead of a barrel full of KT88s, make a lot more sense as far as getting a workable low Z OT without NASA funding.
Using 6 surplus 250 Watt Variac cores ($20 each, just add the 50 turn secondary) wired in series, can get you a high performance circlotron OT for 500 Watts. Two EvilBay Lambda 1000W 360 V power factor converters ($25 each) and a couple of industrial 230V isolation power xfmrs ($25 each) and you are nearly there. Total cost around $250 (tubes and sockets) to $300 (& chassis) for 500 Watts. Cheaper than most beginnner 5 Watt tube kits.
Agree. But one could use the 6AL5 baseline restorer circuit on the final grids that Bob Carver came up with.
--------------------
"we here in Tube forum" "...don't build 400W amps"
George has been up to 250 Watts on the little Red Board with just two Sweep tubes. And 500 W with both channels paralleled. I wouldn't count on it for an Apollo Moon mission though. But using Sweeps, instead of a barrel full of KT88s, make a lot more sense as far as getting a workable low Z OT without NASA funding.
Using 6 surplus 250 Watt Variac cores ($20 each, just add the 50 turn secondary) wired in series, can get you a high performance circlotron OT for 500 Watts. Two EvilBay Lambda 1000W 360 V power factor converters ($25 each) and a couple of industrial 230V isolation power xfmrs ($25 each) and you are nearly there. Total cost around $250 (tubes and sockets) to $300 (& chassis) for 500 Watts. Cheaper than most beginnner 5 Watt tube kits.
"why would he directly couple the cathode follower to the driver, then capacitively couple to the finals? "
Expedience.
The circuit obviously works.
To achieve the optimal as suggested, you would need extra supplies of around plus and minus 200/250 volts which would add to complexity and cost.
Remember at the time Britain was paying for a war it didn't start.
Regards
Henry
Expedience.
The circuit obviously works.
To achieve the optimal as suggested, you would need extra supplies of around plus and minus 200/250 volts which would add to complexity and cost.
Remember at the time Britain was paying for a war it didn't start.
Regards
Henry
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.