G2 power testing and conduction angle

[Andy Le Blanc]

had an Ibanez tn120 over the bench, solid looking product, the power side
was hand dressed easy to service, much more rugged than marshal equivalents

good job ibanez nice looking product inside and out

The power side has a toroidal PT, plate V was riding about 492 with about
40ma on each 6l6 or about 19w plate dissipation.

It had 1k screen grid resistors with the screen riding around -2v to the plate
pretty close to a 6l6gc max screen voltage

The thing I noticed was the G2 voltage was not stable, it wiggled up and down
relative to plate centering around -2v under static conditions.
A half volt either way, I must have been through a dozen amps the last couple
months following G2 voltages and this is the first to show such a instability.

no regulation, and it shows up in the clean tone

[surfsup]

I might try this to see if there is a difference. El84s. Currently my screens are 6v higher than the plates. I was going to increase the screen resistors from 100r to 470r but might put another resistor in, a 1k5w, between the choke and the junction of the two 100r screen Rs, with a RC in parallel to ground.

Where did you connect to ground? At the tube on a mounting bolt, or the main ground star?

[Gaz]

Grounding depends on the overall grounding scheme, but in general, the additional cap should be grounded at the same point as the original screen supply. Hope that makes sense.

[surfsup]

What's more important here? The "regulation" using the cap? or the voltage being lower than the plates, so the 1k5W?

Sounds like the screens should stay below the plates and if I just put in the resistor, shouldn't that do it?

[pdf64]

[quote]plate V was riding about 492 ...with the screen riding around -2v to the plate
pretty close to a 6l6gc max screen voltage [quote]

Surely that's way over the max screen voltage (450V)?
http://www.mif.pg.gda.pl/homepages/fran ... /6L6GC.pdf
Pete.

[Gaz]

You're right that that is over spec, but it has nothing to do with the topic...

[azatplayer]

Hey Andy, i read and reread your post on screen grid regulation yesterday and found it answered some questions for me. I have recently built this Komet COncorde and watch the screens glowing like theres no tomorrow. I bias on the cooler side, but to no avail, them little EL34's get toasted, tho sound great in the process.
i tried a series resistor between the screen node and the screen resistors, initially a 1K then reduced it to 500R then 200R, the 1K had way too much compression, was kinda cool for about ten minutes, but not what that amps about, its much more immediate.
Then i read your post. Thing that im confused by is Gaz's diagram he put up was not how i read the post. The concorde has series filtering on the plates and screens with a pair of 220K balancers. Runs about 490V plate.
You have a diagram of the choke return splitting to the string one way then to the series resistor the other, with the screen filters after the the series resistor.
Gaz diagram had the screen filters as usual, followed by the same setup as your diagram, so 2 sets of filtering.
Its a pretty easy rewire to get the amp to your layout suggestion, was there a need for the pre screen filters?
I have some JJ KT66's in it now that sound fantastic, they have limits of 550V plate and screen.
Im real curious to hear the difference in this amp with lower screen voltage, it sounds pretty great right now, flatout sounds huge, tho i have had some issue with a bit of fizzy top.
CHeers mate, Don

[Andy Le Blanc]

Hey there. Where are the screens now?

You should be able to measure the voltage difference between the plate and screen.

This is very useful, like getting a solid bias figure, it gives you a real indicator
of where the screens are at.

I think the confusion is because of the variations of the same thing we were exploring.

I started simply by adding a resistor at tube socket from the screen grid pin
to ground, after the regular screen grid resistors.

you won't be able to get away with much smaller than a 1/2 w 220k

its a "bleeder", and you begin to develop a "bias supply" for the screens

it can be a very simple divider, or as you see in the confusing variations
you can go further with the idea...

But first try and measure the difference between the plate and G2 screen

[azatplayer]

Thanks Andy, will do

[azatplayer]

Andy i have KT66's in there right now. Got 487V plates, 486V screens , biased @ 35ma. Between plate and screen i get 0.6V positive, black lead to screen, red lead to plate.
Just tried EL34's and i get 482V plates, 478V screens biased @30ma
Difference between plate and screen is 3.9V positive measured.
This sounds ok. But the screen grids go wild with glowage.
These new JJ KT66's sound fantastic out of the box, so much more detail and chimey top, full bottom, compared to the GT 34's that ive been using. Tho they have been hammered.. the GT label is looking very poor..
I know its high volts, tho the amp is, i believe at spec. Right iron etc. And marries to charts ive seen.
This is idling.
Whats the next test
Cheers Don

[azatplayer]

Ok just had a play, pretty loud..
The plates and screens dip when driven hard maybe 40V.
The screens dip lower than the the plates tho, had the leads connected to the plate and screen, and saw maybe 30V difference when pushed hard.
This correlated with what i saw having the black to ground and red on either plate and screen, the screens went down to 415 or so, plates dipped down to maybe 430-440. all happens pretty fast.
BUT, was all in the postive range, AND them grids were burning bright.

[Andy Le Blanc]

you've explained the compression, if the screens are already negative to the
plate going too far will change the tone. The voltage dip is normal.
And you hear the tone of the kt66 with the voltage difference between the
screen and plate being much less than the el34.

you put up 487v plate and a bias of 35ma, that 17w plate dissipation
is that each tube or the total for the pair?

the old kt66 sheets put the screen V limit at 400v, and if your pushing the plate v
theres a recommendation to use a regulated screen supply keeping the voltage to less than 400v

other than that things seem ok, but the screens are day-glo...

the high screen V seems a common concern in your glowing screen grid thread.
And you said it happens (the brite screen grids) regardless of tube type?

[azatplayer]

Yeah pretty much, screens glowing on all of them.
The JJ KT's have high ratings, 550V max plate and screen, and i did have them running a little hot, tried them at 30ma each, then 35ma, 35 sounded better
I have been running the 34's at around 14-15 watts thruout the amps journey tho. I realize i keep placing KT's in with 6L6's dissipation of 30watts, my bad.
I tried the KT77's which JJ state having limits of 800V plate and screen, and they glowed the worst!
EL34's have a screen voltage max of 500V? Will likely stay with them, just gotta get some new ones, its my feeling that this amp is going to eat tubes tho, so im very interested in applying some regulation to the screens, and trying lower voltage on them, say 450V max idle, and seeing what that sounds like. It will of course be different, but might be just as nice.
Can you give me an idea on how to start that?
Cheers Andy, thanks for the discourse!

[Colossal]

Don,

Out of curiosity, did you measure your OT's primary? I know you and I discussed some mixups over that on the supplier's part so just wondering. I think I recall you saying you got it sorted. The Concorde runs at a 4k8 primary but regardless with plate voltages at 500V and a stock screen resistor of 1k, they've all gotta be running with toasty screens. Simply the cost of getting the tone. I'm sure you could bias colder but probably at some compromise in tone.

What is your grid voltage at idle?

Regards,
Dave

[azatplayer]

Hey Dave. The OT measured 4650 within 5% I guess. From memory grids have 46v in them. Yeah I'm guessing it's the price of the tone, which is finally pretty good but I'm interested in the exercise and the lessons learnt. Will confirm that voltage when I get home tonite. Cheers mate