807 SE ULTRALINEAL MAX +B...G2 COULD WORK WITH +B 350V ? ?
Before I read in some post, that the 807 cannot work as a triode or ultralinear, because G2 cannot work with the same voltage as the anode. Something like G2 is to close to G1 and and it can be harmful.
Santiago.
Before I read in some post, that the 807 cannot work as a triode or ultralinear, because G2 cannot work with the same voltage as the anode. Something like G2 is to close to G1 and and it can be harmful.
Santiago.
Attachments
807's G2 voltage can't be close to plate voltage if the latter is close to its max (600V). 350V is more or less fine.
Triode 807 is ok up to 400 V, as of data sheet. UL is indeed limited to 300 V, or you need output transformer with separate UL windings. The 6BG6 version of 807 has 350 V g2 rating. Some Sylvania 6BG6GA, so-called "super 6BG6" can take 400 V on g2 with no problem.
Which would be the difference between G2 with 400V in triode mode , and G2 400v in Ultralinear mode ?
The current ?
I can say I have run the 1625 (12 volt 807) in a car amp ultralinear with about 410 volts on plate and screen. No problems running on bumpy roads with that configuration.
It all comes down to actual screen dissipation which remains lower in ultralinear configuration.
It all comes down to actual screen dissipation which remains lower in ultralinear configuration.
Last edited:
I think the 400V limit only applies with cathode bias, so it is something like 350VDC relative to the cathode.
I think the difference between UL and triode mode is the g2 AC voltage swing. In UL mode g2 does not get so close to the g1, if the anode voltage swing is the same as in triode mode.
In triode connection, the screen always follows the plate. In UL, not always. At some portions of the waveform screen may become positive wrt plate and draw substantial part of cathode current.
santitrucco,
Can the 807 be used in Ultra Linear with 350V on the screen.
Yes, of course it can.
Here is the math:
Quiescent conditions, in UL:
350V on the screen & 345V on the plate (voltage drop across the DCR from UL tap to plate tap).
The plate-to-cathode voltage x plate current has to be 25 Watts or less (plate dissipation).
Signal Conditions, UL:
Let the plate voltage swing down to 50V. That is a 295V drop.
With a 40% UL tap, the screen voltage will drop by 0.4 x 295V = 118V drop.
350 - 118V = 232V on the screen. OK
Quiescent and Signal Conditions, Beam Power mode:
The 807 in Beam Power mode, it is OK to run the screen at 300V when the plate swings down to 50V (there is Lots of screen current then, but just during the peaks of the signal).
So if that is OK for Beam Power mode, then UL is OK too.
Quiescent and Signal Conditions, Triode Wired mode:
The 807 in Triode wired mode, is OK with the Plate and Screen both at 400V.
And when the plate swings down to 50V, the screen is also at only 50V. OK.
Now, you are wondering what happens when the plate swings to 2 x B+.
Well, in triode mode, and with 400V B+, the plate and screen swing to 800V.
That is OK for triode wired mode (Think of Triode Wired mode as 100% UL tap mode).
And in Ultra Linear with 400V B+, and less than 100% UL tap, the screen will not go as high as 800V.
Please do not ask me what is the proper percentage (%) of the UL tap for an 807.
But if it was only 10% (extremely unlikely), UL with the screen at 350V, it could be
crossing the line of correct voltage operation.
Plate from 345 to 50V = 295V
10% of 295V = 29.5V
Screen 350V - 29.5V = 320.5V, barely beyond the 300V screen limit (and only during transients).
Even that will work.
I suspect the correct UL tap percentage (%) is exactly the same as an old 6L6, 6L6-G, 6L6-GA, and a 6L6-GB.
The 6L6-GC, on the other hand, has double the maximum screen dissipation versus the other 6L6s
(2.5 Watts versus 5 Watts 6L6-GC).
(The 807 maximum screen dissipation is 3.5 Watts).
Go on line, and find out what the proper percent (%) tap is for a 6L6.
Use that percent UL tap for the 807.
Go for it!
Build either a single ended 807 UL amp; or build a push pull 807 UL amp.
Happy listening!
And Please, purchase a Real Insulated Plate Cap, before you bring you friends over for a listening session, and before your kids come into the room.
Safety First!
Can the 807 be used in Ultra Linear with 350V on the screen.
Yes, of course it can.
Here is the math:
Quiescent conditions, in UL:
350V on the screen & 345V on the plate (voltage drop across the DCR from UL tap to plate tap).
The plate-to-cathode voltage x plate current has to be 25 Watts or less (plate dissipation).
Signal Conditions, UL:
Let the plate voltage swing down to 50V. That is a 295V drop.
With a 40% UL tap, the screen voltage will drop by 0.4 x 295V = 118V drop.
350 - 118V = 232V on the screen. OK
Quiescent and Signal Conditions, Beam Power mode:
The 807 in Beam Power mode, it is OK to run the screen at 300V when the plate swings down to 50V (there is Lots of screen current then, but just during the peaks of the signal).
So if that is OK for Beam Power mode, then UL is OK too.
Quiescent and Signal Conditions, Triode Wired mode:
The 807 in Triode wired mode, is OK with the Plate and Screen both at 400V.
And when the plate swings down to 50V, the screen is also at only 50V. OK.
Now, you are wondering what happens when the plate swings to 2 x B+.
Well, in triode mode, and with 400V B+, the plate and screen swing to 800V.
That is OK for triode wired mode (Think of Triode Wired mode as 100% UL tap mode).
And in Ultra Linear with 400V B+, and less than 100% UL tap, the screen will not go as high as 800V.
Please do not ask me what is the proper percentage (%) of the UL tap for an 807.
But if it was only 10% (extremely unlikely), UL with the screen at 350V, it could be
crossing the line of correct voltage operation.
Plate from 345 to 50V = 295V
10% of 295V = 29.5V
Screen 350V - 29.5V = 320.5V, barely beyond the 300V screen limit (and only during transients).
Even that will work.
I suspect the correct UL tap percentage (%) is exactly the same as an old 6L6, 6L6-G, 6L6-GA, and a 6L6-GB.
The 6L6-GC, on the other hand, has double the maximum screen dissipation versus the other 6L6s
(2.5 Watts versus 5 Watts 6L6-GC).
(The 807 maximum screen dissipation is 3.5 Watts).
Go on line, and find out what the proper percent (%) tap is for a 6L6.
Use that percent UL tap for the 807.
Go for it!
Build either a single ended 807 UL amp; or build a push pull 807 UL amp.
Happy listening!
And Please, purchase a Real Insulated Plate Cap, before you bring you friends over for a listening session, and before your kids come into the room.
Safety First!
Last edited:
The 807 never got officially rated for UL operation afaik. It did get its triode mode operation rating increased in circa 1950 to 400V as a follow on to its general usage in audio amps.
The 807 may well be more prone to screen damage in UL operation, as there is no official rating, and as there does seem to be a subtle difference in internal screen to grid-1 spacing compared to other common output stage tubes.
I can suggest that common-sense protection against HF oscillation (anode and screen stoppers) is augmented where possible with a view to protect any precious/expensive parts in the amp should an 807 screen fail (eg. worst-case short to grid-1 or cathode). Protection options could include screen to B+ over-voltage, screen over-current, B+ supply over-current, cathode over-current.
807's are relatively cheap, so it is probably the output and power transformers that one might be concerned for, but if not, then just go for it.
A quick look through the 6L6 family datasheets only shows a 6L6GC note saying up to 500V B+ for UL mode, where the B+ limit is 500V. The UL rating would not have typically been applied to earlier versions of 6L6 as that would have required the manufacturer to re-rate the 6L6 and 6L6G and GB.
The 807 may well be more prone to screen damage in UL operation, as there is no official rating, and as there does seem to be a subtle difference in internal screen to grid-1 spacing compared to other common output stage tubes.
I can suggest that common-sense protection against HF oscillation (anode and screen stoppers) is augmented where possible with a view to protect any precious/expensive parts in the amp should an 807 screen fail (eg. worst-case short to grid-1 or cathode). Protection options could include screen to B+ over-voltage, screen over-current, B+ supply over-current, cathode over-current.
807's are relatively cheap, so it is probably the output and power transformers that one might be concerned for, but if not, then just go for it.
A quick look through the 6L6 family datasheets only shows a 6L6GC note saying up to 500V B+ for UL mode, where the B+ limit is 500V. The UL rating would not have typically been applied to earlier versions of 6L6 as that would have required the manufacturer to re-rate the 6L6 and 6L6G and GB.
Last edited:
Running 807's in UL is fine, as long as G2 max power dissipation is not exceeded. I've built an amp that has 420v HT/B+ UL parallel push pull, just make sure you use a suitable grid stopper.
There is an amp design that runs 807's in Extended Class A, two 807's are strapped as triodes, there is a big 1k resistor between anode and screen grid.
It would be a good idea to fit a protection circuit whereby a voltage is sensed across a 10r or similar Rk, this trips a thyristor that energises a relay that cuts the neutral to your mains transformer, see Patrick Turners sire, EL34 UL PP amp 2004.
That said I did kill an 807 running them in UL AB2, this resulted in a cathode heater short.
Andy.
There is an amp design that runs 807's in Extended Class A, two 807's are strapped as triodes, there is a big 1k resistor between anode and screen grid.
It would be a good idea to fit a protection circuit whereby a voltage is sensed across a 10r or similar Rk, this trips a thyristor that energises a relay that cuts the neutral to your mains transformer, see Patrick Turners sire, EL34 UL PP amp 2004.
That said I did kill an 807 running them in UL AB2, this resulted in a cathode heater short.
Andy.
Last edited:
I guess nobody read all that math I did in my post # 10.
I am too long winded for most readers.
Calculating UL screen dissipation is easy.
I did the math there.
To get an estimate, take a look at the Beam Power curves on a 6L6 for example. It shows the screen current versus plate voltage, with the screen at a constant 300V. Look at the very large screen current when the plate swings down to 50V (like when the 6L6 G1 to cathode goes to Zero V.
Just multiply the extremely large screen current times 300V (that = the peak dissipation during the short time that the plate is at 50V).
Anybody that runs an 807 in Class A2 for a single ended UL amp may be asking for tube failure.
But even more so, running an 807 in Beam Power mode, in Class A2 for a single ended amp may have tube failure.
Running an 807 in class C with G1 grid current is more like what the 807 was made to do.
A typical 807 in class C runs lots of G1 grid current for much much much less than 180 degrees.
That is the point of Class C RF: efficiency.
Class C is more akin to switching mode operation.
The point of Class C is Not low harmonic distortion.
Instead, the low harmonic distortion in a Class C RF amp comes from the very effective pi filter (a very good low pass filter).
Sorry, you can not use Class C for audio, and can not use a pi filter for audio (unless it is something like a David Berning amp, or similar).
Just like any other tube, over dissipate at your own risk. Your mileage may vary.
The 6L6-GC is a completely different animal. it came out decades later than the 6L6 "family", and decades later than the 807.
I am too long winded for most readers.
Calculating UL screen dissipation is easy.
I did the math there.
To get an estimate, take a look at the Beam Power curves on a 6L6 for example. It shows the screen current versus plate voltage, with the screen at a constant 300V. Look at the very large screen current when the plate swings down to 50V (like when the 6L6 G1 to cathode goes to Zero V.
Just multiply the extremely large screen current times 300V (that = the peak dissipation during the short time that the plate is at 50V).
Anybody that runs an 807 in Class A2 for a single ended UL amp may be asking for tube failure.
But even more so, running an 807 in Beam Power mode, in Class A2 for a single ended amp may have tube failure.
Running an 807 in class C with G1 grid current is more like what the 807 was made to do.
A typical 807 in class C runs lots of G1 grid current for much much much less than 180 degrees.
That is the point of Class C RF: efficiency.
Class C is more akin to switching mode operation.
The point of Class C is Not low harmonic distortion.
Instead, the low harmonic distortion in a Class C RF amp comes from the very effective pi filter (a very good low pass filter).
Sorry, you can not use Class C for audio, and can not use a pi filter for audio (unless it is something like a David Berning amp, or similar).
Just like any other tube, over dissipate at your own risk. Your mileage may vary.
The 6L6-GC is a completely different animal. it came out decades later than the 6L6 "family", and decades later than the 807.
Last edited: