Ok, what do you think about replacing the driver tube with something like the 6EM7 (http://www.mif.pg.gda.pl/homepages/frank/sheets/049/6/6EM7.pdf)?
Is is a dissimilar dual triode with one high gain section, and one driver section. Basically sort of like a 6sl7 and a 6sn7 in one package. See the Bottlehead SEX amp for an example of this with a 6DM7. There are schematics in the old Valve magazines if you have access to them. You could use one half for lots of gain, and the other half for a sturdy drive section.
That just leaves the phase inverter. I think the coolest solution would be to use an inexpensive transformer, like what is offered by Edcor. You would have to find room under the chassis, and be sure it doesn't pick up and hum, but a transformer is a very balanced phase inverter. My second choice would be to do a self-split output section where the output tubes are configured as the phase inveter (one grid grounded, cathodes shared). To make this arrangement more linear, you could use a current source (sink?) instead of a cathode resistor, or even better, use a choke with enough current capacity and approximately the right DC resistance to replace the cathode resistor. I have done this in my choke-tail phase inverter and really liked it. Of ocurse for this you would have to find space for yet another choke.
I would stay away fromt he cathodyne phase inverter, but that is just a personal bias of mine. I think balance is the most important thing in a phase inverter, and cathodyne just doesn't cut it. Plus they have no gain, and you want gain here if possible.
jsn
Is is a dissimilar dual triode with one high gain section, and one driver section. Basically sort of like a 6sl7 and a 6sn7 in one package. See the Bottlehead SEX amp for an example of this with a 6DM7. There are schematics in the old Valve magazines if you have access to them. You could use one half for lots of gain, and the other half for a sturdy drive section.
That just leaves the phase inverter. I think the coolest solution would be to use an inexpensive transformer, like what is offered by Edcor. You would have to find room under the chassis, and be sure it doesn't pick up and hum, but a transformer is a very balanced phase inverter. My second choice would be to do a self-split output section where the output tubes are configured as the phase inveter (one grid grounded, cathodes shared). To make this arrangement more linear, you could use a current source (sink?) instead of a cathode resistor, or even better, use a choke with enough current capacity and approximately the right DC resistance to replace the cathode resistor. I have done this in my choke-tail phase inverter and really liked it. Of ocurse for this you would have to find space for yet another choke.
I would stay away fromt he cathodyne phase inverter, but that is just a personal bias of mine. I think balance is the most important thing in a phase inverter, and cathodyne just doesn't cut it. Plus they have no gain, and you want gain here if possible.
jsn
SY said:
Dang it, I got caught making a sweeping and unsubstantiated statement of gut feeling
I just think that the operation of a cathodyne phase inverter is compromized becuase of the way that half the signal comes from the plate and half from the cathode, one with gain, one without. Sure you can tweak it to minimize the difference, but why not start with something more inherently balanced/symmetrical. The long-tail-pair is better, especially if you have a substantial "tail" which is why I like a choke there. And a trnsformer adds cost and space, but sort of eliminated the whole balance issue all together.
Does that make any sense? I realise I am discussing this in the least technical of terms, but that is sort of how I think about these things.
jsn
Here's an easy way to think about it: the plate current and cathode current of a triode MUST be equal as long as the tube isn't overloaded, right? So any swing on one must equal the swing on the other. Therefor, the balance is as perfect as the matching of the loads.
The diff amp (long tail pair with one input grounded) actually has lower gain than the same two tubes configured as a grounded cathode voltage amp and a split load inverter.
The diff amp (long tail pair with one input grounded) actually has lower gain than the same two tubes configured as a grounded cathode voltage amp and a split load inverter.
So 6SL7 is the tube to use with the current configuration.
With the difference in Cathode resistors on the driver tube, the grid resistor on the output tube, the feedback resistors and other differences, is it worth making it exactly like the Acro 6L6 schematic? Does the Acro amp design look like it should sound nice?
I know it a rat's nest of wiring, but there is no hum or hiss. Should I bother? I would hate to clean up the arrangement just to get hum afterwards...
I like the amp and how it sounds. Being the engineer I am, I can't help but want to mess with it.
Is it worth modifying to a 6SN7 driver or making the Audio Note 2PSE amp with the single 6SL7 pair of 6L6 per channel?
Also, (brace yourself for real dumb question) I have made a few preamps that have LED Bias. I have yet to see LED Cathode Bias on tube amps...could I experiment with it here on the driver tubes?
And the current thru driver (the "wrong" 6SN7) is 1.74ma - (258-171)/50k, and (258-90)/100k. Does this sound low for the 6SN7?
With the difference in Cathode resistors on the driver tube, the grid resistor on the output tube, the feedback resistors and other differences, is it worth making it exactly like the Acro 6L6 schematic? Does the Acro amp design look like it should sound nice?
I know it a rat's nest of wiring, but there is no hum or hiss. Should I bother? I would hate to clean up the arrangement just to get hum afterwards...
I like the amp and how it sounds. Being the engineer I am, I can't help but want to mess with it.
Is it worth modifying to a 6SN7 driver or making the Audio Note 2PSE amp with the single 6SL7 pair of 6L6 per channel?
Also, (brace yourself for real dumb question) I have made a few preamps that have LED Bias. I have yet to see LED Cathode Bias on tube amps...could I experiment with it here on the driver tubes?
And the current thru driver (the "wrong" 6SN7) is 1.74ma - (258-171)/50k, and (258-90)/100k. Does this sound low for the 6SN7?
I took more voltages of the 6L6GC driven by the 6SN7GTA and the with the 6SL7GT in the drivers seat..
6L6GC
Plate 371V
Cathode 28V
Grid 1 0V
Grid 2 370V
6SN7
Plate 1 88V
Cathode 1 2V
Grid 1 0V
Plate 2 171V
Cathode 2 88V
Grid 2 64V
Voltage to 6SN7 Plate Resistor is 263V
Plate current across 100k plate resistor is 1.75ma
6SL7
Plate 1 201V
Cathode 1 1V
Grid 1 0V
Plate 2 251V
Cathode 2 55V
Grid 2 33V
Voltage to 6SL7 Plate Resistor is 310V
Plate Current across 100k plate resistor is 1.1ma
Aren't both of these these plate currents on drivers a bit low? Modify the cathode resistors on drivers to raise to the proper current?
I guess having separate Dual Mono power supply design I can put both driver tubes in each channel and listen/measure differences....
6L6GC
Plate 371V
Cathode 28V
Grid 1 0V
Grid 2 370V
6SN7
Plate 1 88V
Cathode 1 2V
Grid 1 0V
Plate 2 171V
Cathode 2 88V
Grid 2 64V
Voltage to 6SN7 Plate Resistor is 263V
Plate current across 100k plate resistor is 1.75ma
6SL7
Plate 1 201V
Cathode 1 1V
Grid 1 0V
Plate 2 251V
Cathode 2 55V
Grid 2 33V
Voltage to 6SL7 Plate Resistor is 310V
Plate Current across 100k plate resistor is 1.1ma
Aren't both of these these plate currents on drivers a bit low? Modify the cathode resistors on drivers to raise to the proper current?
I guess having separate Dual Mono power supply design I can put both driver tubes in each channel and listen/measure differences....
6SN7 is much lower gain than 6SL7. something like 20 vs 70 times.
If you want gain and don't mind metalwork here is an idea: use a variation of the EICO EF87 using 6SN7 LTP driver with a single triode section in front of them as an input stage. A 6SN7 driver will sound a lot beefier in my experience.
Here is a schematic: http://www.triodeel.com/eicohf87.gif
This is a variation of a classic mullard circuit which sounds very good IMHO. If you want still more gain you can have a pentode front end using a pair of EF86, like the mullard 5-20 circuit.
Another random idea: convert the output section to EL34, they have more gain and need less drive than 6L6.
If you want gain and don't mind metalwork here is an idea: use a variation of the EICO EF87 using 6SN7 LTP driver with a single triode section in front of them as an input stage. A 6SN7 driver will sound a lot beefier in my experience.
Here is a schematic: http://www.triodeel.com/eicohf87.gif
This is a variation of a classic mullard circuit which sounds very good IMHO. If you want still more gain you can have a pentode front end using a pair of EF86, like the mullard 5-20 circuit.
Another random idea: convert the output section to EL34, they have more gain and need less drive than 6L6.
Also, if I get rid of the tube rectifiers and go with solid state (1N4007?), I could open up two tube octal sockets for me to use 6SL7 and 6SN7 (ala Williamson...)
Those two monster transformers are 440V- 0V -440V, hence why those 820 ohm 25 watt resistors are there - to drop it from 490V out of the 5R4 tube rectifiers to around 385V to the PS caps.
Will SS rectifiers produce higher voltage than the 5R4 I have now? I will try modeling it in Duncan Amps PS Designer to see what the increase in voltage will be.
Is going from Tube rectification to SS recommended (I trust that I am not opening a can of worms on this question!)
Those two monster transformers are 440V- 0V -440V, hence why those 820 ohm 25 watt resistors are there - to drop it from 490V out of the 5R4 tube rectifiers to around 385V to the PS caps.
Will SS rectifiers produce higher voltage than the 5R4 I have now? I will try modeling it in Duncan Amps PS Designer to see what the increase in voltage will be.
Is going from Tube rectification to SS recommended (I trust that I am not opening a can of worms on this question!)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.