• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

LS50/GU50 Both Grid Drive

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Oddly enough there seems to be an actual G1=G2 graph for the LS50.
I wonder how applicable it might be for the Ruskie equivalent GU50?
Looks to me about a 7K5 loadline with a reasonable 300V B+

Anyways, Zob has one (stuck in a military socket and not coming out).
And asked if I could come up with any ideas for it. He also has a 5687
burning a hole for want of a good use.

The objective is to power a Karlson K15 type, with a vintage coaxial.
Its only gonna be a monoblock.

If there are better ways to run this tube, I am not stuck on G1=G2.
But I will present the idea for comments anyway.
 

Attachments

  • zobgu50.gif
    zobgu50.gif
    38.3 KB · Views: 1,311
G1 = G2 makes it act like an 811, or similar tubes.. almost no plate current at VG1=VG2=0, and pulls grid current above of course...

pain in the *** to drive vs. normal triode or pentode mode... If I wanted a tube that was a pain to drive, I'd go with an 811, because it looks nice at least ;)
 
Seems likely to me that g1=g2 drive will quickly start drawing heavy current on g1 before the plate current gets very high. (Thus explaining George's smoked Mosfet driver experiment) Since g2 won't have enough voltage on it to draw the little e buggers thru g1 and it also shields out the plate field too. I would suggest driving g2 directly with a resistive divider down to g1, so that each grid has drive voltage scaled to 1/gm. Net result would then be a doubling of the g2 gm effectively.

Don
 
I have seen threads about this paralleling of grids before and just shook my head and kept silent. This is not the way a tube was designed to be used. To make a triode out of a tetrode or pentode, the screen gets tied to the plate. That is the proper way. And in so doing, the gain characteristics of the tube are changed such that the screen can operate at a higher voltage because it is in league with the plate and has less controlling effect. Screen and control grid do not get connected together. This is the way I was always taught.

Yet every so often people will go against the grain and try making a triode by tying them together. Has no one told them that is incorrect? That the usefull characteristics of a tube are destroyed? Perhaps I am too set in my ways and too old to learn new tricks. But I think not. IMHO using a tube in this maner is about as cockamamie as inverted tube operation. Maybe it's the fun of doing something one shouldn't or the joy of an exercise in near futility. I just don't know.

Ok, end of rant.
Victor
 
I am going to try a double-drive such a way:
1. Differential phase splitter,
2. 2 asymmetrical Concertinas (higher value of resistors in anodes) coupled to each anode of a diff stage directly;
3. Source followers from their anodes to screen grids of output pentodes,
4. Capacitive coupled cathodes of Concertinas to control grids of output pentodes.
The result should be kind of opposite to UL mode.
 
Perhaps I am too set in my ways and too old to learn new tricks. But I think not. IMHO using a tube in this manner is about as cockamamie as inverted tube operation. Maybe it's the fun of doing something one shouldn't or the joy of an exercise in near futility. I just don't know.

"You are old, Father William," the young man said,
and your grids are uncommonly driven.
Yet you ran your load lines to positive twelve,
despite the data that's given?"

"In my youth," said the sage, as he soldered the joint,
"I chose to run conventionally;
but with these new options I find that I can
increase power sensitivity."
 
AX tech editor
Joined 2002
Paid Member
SY said:


"You are old, Father William," the young man said,
and your grids are uncommonly driven.
Yet you ran your load lines to positive twelve,
despite the data that's given?"

"In my youth," said the sage, as he soldered the joint,
"I chose to run conventionally;
but with these new options I find that I can
increase power sensitivity."

Brilliant! Now on solid state, topic "running in class B".

Jan Didden
 
I had considered driving G1 from the lower 533R, but I didn't have
a published set of curves to say exactly what might happen. It was
not my intent to hand Zob a design I couldn't be reasonably sure
would work on the very first try. There is room for later experiment.

Nobody has complained yet that I use too many capacitors?
 
Wavebourn said:
I am going to try a double-drive such a way:
1. Differential phase splitter,
2. 2 asymmetrical Concertinas (higher value of resistors in anodes) coupled to each anode of a diff stage directly;
3. Source followers from their anodes to screen grids of output pentodes,
4. Capacitive coupled cathodes of Concertinas to control grids of output pentodes.
The result should be kind of opposite to UL mode.

If you find yourself needing to adjust the overall amount of UL,
resistive plate to plate (output plates back to the concertinas)
feedback might be handy.
 
HollowState said:
I have seen threads about this paralleling of grids before and just shook my head and kept silent. This is not the way a tube was designed to be used. To make a triode out of a tetrode or pentode, the screen gets tied to the plate. That is the proper way. And in so doing, the gain characteristics of the tube are changed such that the screen can operate at a higher voltage because it is in league with the plate and has less controlling effect. Screen and control grid do not get connected together. This is the way I was always taught.

Yet every so often people will go against the grain and try making a triode by tying them together. Has no one told them that is incorrect? That the usefull characteristics of a tube are destroyed? Perhaps I am too set in my ways and too old to learn new tricks. But I think not. IMHO using a tube in this maner is about as cockamamie as inverted tube operation. Maybe it's the fun of doing something one shouldn't or the joy of an exercise in near futility. I just don't know.

Ok, end of rant.
Victor


the Tung-Sol 3D21W delivers g1=g2 curves on the data sheet. Admittedly not too useful looking, but if you trace them, they will come.
cheers,
Douglas
 
This is not the way a tube was designed to be used......Perhaps I am too set in my ways and too old to learn new tricks.

As I have been told before, every possible connection that could be done with tubes has been done already. I remember seing a schematic a long time ago that ran 807's with G1 and G2 tied together in "class B" mode. This is where I got the idea. I can't find the schematic on the web but I have a paper copy somewhere.

I am going to try a double-drive such a way.....I would suggest driving g2 directly with a resistive divider down to g1, so that each grid has drive voltage scaled to 1/gm.

OK, why not one typical P-P driver with each output feeding a pair of mosfet drivers (4 total). Each mosfet has independent gain and DC bias adjustments.
 
Hi Tubelab,

I remember seing a schematic a long time ago that ran 807's with G1 and G2 tied together in "class B" mode.

Maybe you mean the schematic as attached below.

The "current trickler" configuration you get by connecting g1 and g2 together makes pretty much sense in such a zero bias class B design. Probably just decent enough for train platform public adress systems... at least such a design has its merits regarding power-efficiency, when no drive signal is applied...

Regards,

Tom Schlangen
 

Attachments

  • 807pp_6l6g.jpg
    807pp_6l6g.jpg
    33 KB · Views: 825
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.