Reply to post 77: Substitute the 7k5 resistor by a penthode EL34, desolder the coupling cap. This gives a much lighter ac load to the triode connected (signal processing) EL34. Of course the current draw by both tubes should be made equal. Output power will be 5-6W rms.
6A3s - yeah, that's what the 7500 Ohms do; sink 54 mA to ground on one side of a p-p transformer, with a 405V B+. The other side of the p-p is connected to a 6CA7, biased also ~50 mA.
Mangetically, I'd think that 1/2 of a p-p winding with 50mA going through would do the very same thing as one's own winding - using whatever available space - with whatever current that arrangement would take to offset the primary current of the remaining tube.
Perhaps the tertiary winding method allows use of the full transformer primary winding, with a higher impedance for tubes and circuitry that could take advantage of it.
Using half of the existing output transformer primary winding to offset the 6CA7 bias current seems to be an easy way to use a p-p transformer in a single ended arrangement.
The schematic is the OEM version of the amp's circuitry. I've done a couple things;
1. changed pentode input to triode mode.
2. changed regulated G2 pentode outputs to triode mode. 27V "zener" from plate to G2 on each output tube.
3. When I went to triode mode on the outputs, that left two independent zener regulated 300V supply rails dangling. I connected these to the triode modified pentode input's plate resistor to give that section it's own regulated B+.
4. The feedback is connected separately to the full range speaker's terminals; two resistors support fallback. These are not shown in the OEM schematic.
The 7500 Ohm resistors are just an experiment and I think they're doing exactly "it". I'm not sure if I want to pursue it across further, say a different resistor value, one that draws 40 mA at 405V... The whole thing could change character - as it did when I was running 2 tubes at ~35mA and then went to one tube running at 50 mA.
I'm going to try the dial I've been thinking of, that cuts off the high-side drive of the cathodyne to the output tube. It should vary the plate drive signal from a little more than the cathode output, all the way to zero. This would allow you to try to balance the p-p levels exactly by ear, or turn one tube gradually off which would unbalance the 2nd harmonic cancellation in the OPT.
I dont like how limited in power the amp is when the resistors are used. I put the tubes back in listened, went back to the resistor - both sound good, the single tube is fine generally, but if I want to turn it up AT ALL it begins to really distort. Maybe a nice dial that you can spin to single ended or back to push pull, or anywhere in between. There's that hole in the front of the chassis - gotta fill it with somethin'.
Thanks!
Mangetically, I'd think that 1/2 of a p-p winding with 50mA going through would do the very same thing as one's own winding - using whatever available space - with whatever current that arrangement would take to offset the primary current of the remaining tube.
Perhaps the tertiary winding method allows use of the full transformer primary winding, with a higher impedance for tubes and circuitry that could take advantage of it.
Using half of the existing output transformer primary winding to offset the 6CA7 bias current seems to be an easy way to use a p-p transformer in a single ended arrangement.
The schematic is the OEM version of the amp's circuitry. I've done a couple things;
1. changed pentode input to triode mode.
2. changed regulated G2 pentode outputs to triode mode. 27V "zener" from plate to G2 on each output tube.
3. When I went to triode mode on the outputs, that left two independent zener regulated 300V supply rails dangling. I connected these to the triode modified pentode input's plate resistor to give that section it's own regulated B+.
4. The feedback is connected separately to the full range speaker's terminals; two resistors support fallback. These are not shown in the OEM schematic.
The 7500 Ohm resistors are just an experiment and I think they're doing exactly "it". I'm not sure if I want to pursue it across further, say a different resistor value, one that draws 40 mA at 405V... The whole thing could change character - as it did when I was running 2 tubes at ~35mA and then went to one tube running at 50 mA.
I'm going to try the dial I've been thinking of, that cuts off the high-side drive of the cathodyne to the output tube. It should vary the plate drive signal from a little more than the cathode output, all the way to zero. This would allow you to try to balance the p-p levels exactly by ear, or turn one tube gradually off which would unbalance the 2nd harmonic cancellation in the OPT.
I dont like how limited in power the amp is when the resistors are used. I put the tubes back in listened, went back to the resistor - both sound good, the single tube is fine generally, but if I want to turn it up AT ALL it begins to really distort. Maybe a nice dial that you can spin to single ended or back to push pull, or anywhere in between. There's that hole in the front of the chassis - gotta fill it with somethin'.
Thanks!
Last edited:
Hello Joe,
The problem with using half the primary means you get a quarter of the output impedance of the OT. Usually the output impedance of a pp EL34 is somewhere from 3k6-4k. This means the one EL 34 sees 900-1000 ohms. Hence the distortion you mentioned.
A triode-connected single ended EL 34 at 400V at 50ma likes to see around 5k (this is a guess but it won't be far off). Some nfb will be helpful here as it lowers the output impedance and also connecting your speaker (when it's 8 ohms) to the 4 ohms pin of your OT (if that's possible).
Other possibilities are
- paralleling the two EL34's while the unused half of the OT is earthed through a mosfet current source. In that case the PT needs to deliver double the current however so this seems unrealistic.
- lower Vb to about 300V as this lowers the output impedance considerably. Although the price to pay is a lower output but at a lower distortion. Maybe this plus some nfb will produce some good results. Success!
The problem with using half the primary means you get a quarter of the output impedance of the OT. Usually the output impedance of a pp EL34 is somewhere from 3k6-4k. This means the one EL 34 sees 900-1000 ohms. Hence the distortion you mentioned.
A triode-connected single ended EL 34 at 400V at 50ma likes to see around 5k (this is a guess but it won't be far off). Some nfb will be helpful here as it lowers the output impedance and also connecting your speaker (when it's 8 ohms) to the 4 ohms pin of your OT (if that's possible).
Other possibilities are
- paralleling the two EL34's while the unused half of the OT is earthed through a mosfet current source. In that case the PT needs to deliver double the current however so this seems unrealistic.
- lower Vb to about 300V as this lowers the output impedance considerably. Although the price to pay is a lower output but at a lower distortion. Maybe this plus some nfb will produce some good results. Success!
jjasniew,
Using a 6CA7 or EL34 in Pentode mode as a current sink for the other half of the primary, you have somewhere between 15 to 24k of plate resistance. That does not rob much power from the other 6CA7 or EL34 that is Triode wired driving the other half of the primary. The Triode wired 6CA7 or EL34 has a plate resistance of about 1500 to 2000. Ohms.
Compared to that Pentode wired 6CA7 or EL34, the 7500 Ohm resistor robs lots more power from the Triode wired tube on the other end of the primary.
But one thing going for the 7500 Ohm resistor is, it is always 7500 Ohms. The 6CA7 or EL34 Pentode wired may not be the same plate resistance as the Triode wired tube voltage swings with signal. This is especially as the Pentode wired tube plate to cathode voltage drops to a lower voltage, say 100, 75, or 50V.
In order to make the Pentode wired tube have a more constant current (and higher plate impedance), the self bias resistor of that tube should not be bypassed. Then to see how good that configuration can get, short the grid stopper resistor to ground after you disconnect the coupling cap. If you do not have individual self bias resistors on each tube, then use two resistors, and double the resistance of what the common self bias resistor was. Be sure to bypass the self bias resistor of the tube that is driven from the signal, and do not bypass the other tube that has the function of being a current sink.
The typical push pull transformer has lots more inductance than a typical single ended transformer (has an air gap). So, with the push pull transformer, you can put the 8 Ohm speaker on the 4 Ohm tap. That increases the load impedance the primary has on the Triode wired tube. (see KeesB post #83)
Tradeoffs.
Using a 6CA7 or EL34 in Pentode mode as a current sink for the other half of the primary, you have somewhere between 15 to 24k of plate resistance. That does not rob much power from the other 6CA7 or EL34 that is Triode wired driving the other half of the primary. The Triode wired 6CA7 or EL34 has a plate resistance of about 1500 to 2000. Ohms.
Compared to that Pentode wired 6CA7 or EL34, the 7500 Ohm resistor robs lots more power from the Triode wired tube on the other end of the primary.
But one thing going for the 7500 Ohm resistor is, it is always 7500 Ohms. The 6CA7 or EL34 Pentode wired may not be the same plate resistance as the Triode wired tube voltage swings with signal. This is especially as the Pentode wired tube plate to cathode voltage drops to a lower voltage, say 100, 75, or 50V.
In order to make the Pentode wired tube have a more constant current (and higher plate impedance), the self bias resistor of that tube should not be bypassed. Then to see how good that configuration can get, short the grid stopper resistor to ground after you disconnect the coupling cap. If you do not have individual self bias resistors on each tube, then use two resistors, and double the resistance of what the common self bias resistor was. Be sure to bypass the self bias resistor of the tube that is driven from the signal, and do not bypass the other tube that has the function of being a current sink.
The typical push pull transformer has lots more inductance than a typical single ended transformer (has an air gap). So, with the push pull transformer, you can put the 8 Ohm speaker on the 4 Ohm tap. That increases the load impedance the primary has on the Triode wired tube. (see KeesB post #83)
Tradeoffs.
Last edited:
I like the idea of floating CCS if it was powered by a floating 910v as written on schematic , but it is just 91v . The supply of CCS is 91v+ac output , if the output triodes pull more current the voltage of CCS becomes several hundred volts ,OK . But when the triodes release their currents the CCS supply becomes negative , the PL519 goes to cutoff and the bias current now must pass through the primary hence increasing instead of decreasing. As you confirm the circuit is functional, how it behaves in real ?
How about canceling the dc bias current at the secondary by a low voltage SS CSC ?
How about canceling the dc bias current at the secondary by a low voltage SS CSC ?
Last edited:
As I said before a ccs on either an extra winding or instead of that the 0-4 ohms part of the secundary seems to be the easiest way to undo the magnitizing effect of the anode current through the primary. Besides all of the primary can be used; it's possible to parallel the two EL34's. The anode current times the turns ratio primary/secundary gives the neceesary current of the ccs. Example: Suppose you connect 100v ac on the whole primary and you get 3,3v on the 4 ohms tap the ratio is 30. If the current through the primary is 0.1 A the ccs must be 3A. A LM317 plus a 150w power tor , a low voltage transformer and a few electrolytics are all you need to make it work. When connected the wrong way the OT becomes a magnet . Easy to detect by holding a screwdriver near the core . In that case reverse the connections you put your ccs on.
To answer your question, my proposal, with the floating 519 works only for low output although I had enough power in my living room. As I said before the ccs as just described is by far the easiest and cheapist way to counteract the anode current thtrough the primary.
To answer your question, my proposal, with the floating 519 works only for low output although I had enough power in my living room. As I said before the ccs as just described is by far the easiest and cheapist way to counteract the anode current thtrough the primary.
One could also say that the easiest and cheapest way to counteract the anode current through the primary is to use an identical valve with a center-tapped primary, driven in opposite polarity. This has the added advantages of increasing peak output by 3dB and lowering distortion at all levels.
All good fortune,
Chris
All good fortune,
Chris
Last edited:
You use only low power , but what is the maximum peak-peak swing voltage possible on the primary?
Last edited:
Chris H. you're suggesting to change the circuit into a pp output again. The reason I started this thread was to find an (affordable) alternative to the required high output impedance transformer of a pp VT4C . So what you propose is exactly the opposite of what this thread is about.
An other option could be to change it to a "parafeed" amp. But your idea is much more interesting as it is original.
I'm suggesting that you're running up against Second Law restraints (TANSTAAFL). If the actual goal is to best utilize a given transformer in a role for which it's poorly designed, the discussion should be opened up to other designs that could perhaps better suit.
Aesthetically, well, the type 211/VT-4C deserves better than the misuse it's always been given. But I'm an old crank, so that must be weighed in.
All good fortune,
Chris
I 'll split up your reaction. First: "The actual goal is to best utlize a given transformer in a role for which it is poorly designed".
In my situation a 6k6 pp output transformer to be used as a single ended one because I had this at hand; therefore the ccs as counter current source. The alternative ccs as proposed by several panel members is greatly appreciated because it's in line with what I proposed and which is both easy and cheap. The idea was many amateurs are in the same situation and that they could try something like this too.
Now part two: "the discussion should be opened up to other designs that could perhaps better suit"
This is true only regarding the ccs. This is in line with my proposition. But suggesting a VT4C pp application with a 6k6 PP OT is not and wrong at that.
In my situation a 6k6 pp output transformer to be used as a single ended one because I had this at hand; therefore the ccs as counter current source. The alternative ccs as proposed by several panel members is greatly appreciated because it's in line with what I proposed and which is both easy and cheap. The idea was many amateurs are in the same situation and that they could try something like this too.
Now part two: "the discussion should be opened up to other designs that could perhaps better suit"
This is true only regarding the ccs. This is in line with my proposition. But suggesting a VT4C pp application with a 6k6 PP OT is not and wrong at that.
Lampie519,
Take a look at my post # 8 near the beginning of this thread.
I discussed the Parafeed option, and some of its advantages.
Of course the original post (# 1)was a different idea, I just wanted those who were not familiar with Parafeed to at least be aware of that classical circuit.
Take a look at my post # 8 near the beginning of this thread.
I discussed the Parafeed option, and some of its advantages.
Of course the original post (# 1)was a different idea, I just wanted those who were not familiar with Parafeed to at least be aware of that classical circuit.
Indeed you mentioned it before ! I like out of the box thinking and trying new ideas as so much more we can get out of the tubes then in the old days. We are not limited to the ideas of the “dark ages” as we now can mix old and new technologies plus all is available to us now. In the old days building a VT4 amp was limited to the technology at hand. Now we get crazy and build amps without limitations. We add as much stuff as we think is required to get the results we like to achieve. I have build a preamp with 15 tubes, who cares?
It sound just amazing, as if there is no amp at all in the signal chain.
It sound just amazing, as if there is no amp at all in the signal chain.
Attachments
Last edited:
I filled the hole, with a 100K dual potentiometer, audio taper. It works. The wiper and one side replace the 100K plate resistor of the cathodyne, such that rotated all the way open, same 100K as stock. Rotated all the way closed, plate connects directly to ~400V DC. At either end of this range, cathode is ~60V above ground.
When closed, the change in gain is perceptible. A change in character of the amplifier also. I havent played with listening enough to come up with any superlatives, except to say that the amp seems more open, less constrained as the OEM designed it and more stressed when I have one tube clamped off.
But then there's the middle area, where the cathode connected tube is operating normally, the plate connected tube is operating "some", resulting in an unbalanced drive and certainly effecting the 2nd order harmonic cancellation in the perfect push-pull situation. Gives me something to play with, which I like to have. Enough to order some linear taper dual 100K pots that'll fit in the spot. Thanks!
When closed, the change in gain is perceptible. A change in character of the amplifier also. I havent played with listening enough to come up with any superlatives, except to say that the amp seems more open, less constrained as the OEM designed it and more stressed when I have one tube clamped off.
But then there's the middle area, where the cathode connected tube is operating normally, the plate connected tube is operating "some", resulting in an unbalanced drive and certainly effecting the 2nd order harmonic cancellation in the perfect push-pull situation. Gives me something to play with, which I like to have. Enough to order some linear taper dual 100K pots that'll fit in the spot. Thanks!
Last edited:
The way I view this circuit is that it IS parafeed.
The last cap of the power supply simply becomes the parafeed cap and the only difference (added complexity) is the transformer is referenced to B+ rather than ground so additional concern must be taken to set DC voltages to prevent current flow.
If you draw out the HV power supply in any of the schematics it becomes clear that all that is needed is the last cap from the power supply to complete the AC loop through the output transformer and no DC connection to the power supply is actually needed to the primary of the output transformer.
dave
Definitively not. Much of the confusion about this circuit in this forum has arisen from PRR redrawing of my circuit so better look at the original post I started this thread with.
Is the attached drawing representative of your circuit? If yes what purpose does the red connection serve beyond adding the requirement of nulling the DC through the output?
Going back to my original schematic I now see a mistake in the drawing made by a friend of mine (He had a drawing programm, I didn't). I now understand a number of the reactions that I thought strange. I hadn't taken the time to look over it assuming all was ok. So the red line is correct. The powersupply of the 519 should be 90V under load; the current through the 519 is 0.1 A as is the anode current of the VT4C's. The used pp output requires nulling of the current through the primary as the pp output transformer doesn't have an airgap ; it serves no other purpose.
- Status
- Not open for further replies.