[fields 100]

I just finished recapping my 5S319. After turning it "on" the set went to operation right away with good reception even before alignment. Then the following started happening (before, during and after alignment): it would work good for some time, then the signal would fade away completely, then it comes back, over and over again.The longer I keep the set "on", the shorter the time between the fading. Fed the signal from my signal generator, and during the fading phase the only tube not letting the signal through was 6A8G.When I wiggle the tube and touch the grid the sound comes back after few moments. The voltages are: plate 200dc, pin#4 -200dc and pin #6- some crazy 800dc, and that is during the normal operation, but when it fades away the needle goes over 1000dc(?). Prior to that there was a break in the oscillator coil, I had to rewind it, but I think it would not work at all if I messed  it up. I do not have an extra tube laying around.
Does it sound like it is?  Thanks.

http://www.nostalgiaair.org/PagesByModel...001491.pdf

[BrendaAnnD]

The link you gave is for an A-K set.

As for your problem, and your voltages:

You've either got a gassy tube, or a bad grid leak resistor. You're not seeing 800V or 1000V on that pin, those voltages don't exist in that set. What you're most likely seeing is 800-1000mV (I sometimes make that mistake with DMM's, too). Try swapping out the tube first, then start looking for resistors connected to it that have radically changed value.

[fields 100]

Sorry for the link. Something wrong with a computer. Will fix it.

I did not have my DVM on MV, just checked, it was on DC, 1000V range...Very bizarre.

[fields 100]

http://www.nostalgiaair.org/PagesByModel...025008.pdf

Here is the right one.

[Radioroslyn]

You actually trust those new fangeled DVM???? Grab a Tripplet 610 or a Simpson 260. Then give your set some BEAN-O cause it's gassy!!
Terry 
ps  Sounds like your DVM is lieing to you. I've had it happen when work around RF and OSC circuits. The semiconductors in it pickup rf voltages from the circuit being tested and rectify them. They can cause a DVM to give erroneous readings.

[Mondial]

Fields, try connecting a 100K resistor in series with the positive DVM probe and make the measurements again. This will filter any RF from the DVM input and hopefully give you a more reasonable reading. The series resistance will only affect the reading accuracy slightly, making it read about 1% lower with a 10 Megohm input DVM.

As Terry says, it appears you  have a gassy tube or an open grid resistor somewhere.

[fields 100]

Thanks guys,

Out for the new tube. All resistors checked out good.

[fields 100]

Update. While waiting, out of nothing to do removed the metal case from the tube, which was somewhat loose.
All 8 pins were unsoldered (!), and the case came off quite freely. Re-soldered all 8 of them. All voltages are good and the set plays. The problem solved.