Ampeg SVT-CL zero Ig2 in one tube causes funny readings all over

[tubeswell]

Or is it just that I haven't trouble shot everything yet

Having put humpty dumpty back together for the umpteenth time (amp came in deadski after owner tried to fit new JJ6550s and bias them up himself- I found a fault in the bias sensor supply which I fixed and have also soldered the 6V leads from the filament x-former permanently to the output tube board and the bias), I got around to checking the output tube condition today.

Amp is really hummy and dialing down the bias on that side quietens it down a bit. The red and green bias light come on on that side, but only the red light on the other side.

(Note - Dialing the bias voltage down on the l.h. side quietened the hum down a bit , hence the following g1 measurements are different for each side.)

Voltages are as follows (I didn't bother taking photos of them all, but the readings are accurate enough)

L.H. side (closest to PT)
Tube 1 (front next to PT) Plate = 13V, g2 = 698V, g1 = -68.3V, tube current = 95mA
Tube 2 (middle front) Plate = 13V, g2 = 692V, g1 = -68.3V, tube current = 99mA
Tube 3 (rear next to PT) Plate = 13V, g2 = 656V, g1 = -68.3V, tube current 109mA


R.H. side (closest to OT)
Tube 1 (front next to OT) Plate = 293V, g2 = 663V, g1 = -53.5V Tube current = 210mA
Tube 2 (middle rear) Plate 293V, g2 = 677V, g1 = -53.5V, tube current = 201mA
Tube 3 (the culprit - rear next to OT) Plate 293V, g2 = 0V, g1 -53.5V, tube current = 0mA

So that looks like an open screen resistor (or faulty Rg2 rail?) for Tube 3 (R.H. side). I'll have to pull it apart again to hunt for that problem.

Would that be the cause of the large discrepancy between plate voltage readings between both sides?

Are people using flameproof Rg2 resistors in these, or can one get away with ordinary metal film Rg2s?

[Enzo]

That looks all wrong all around.

What voltages are on pins 3,4,5 on each socket with NO tubes installed? Don't install any power tubes until the voltages are right.

Yes, I'd wager that screen resistor is open. But I might expect 600v on the plates and 300 on the grid 2, not the other way around as you report.

ANy chance a ribbon cable is on the wrong pins inside? Look at the power supply on the schematic here:
http://bmamps.com/Schematics/ampeg/A..._Schematic.pdf

See the main B+ is 660-some volts to the OT CT, while the screen supplies are 340v-ish.

Note the small tubes, V1 is the PI and input buffer, while V2,3 are the driver tubes. V2,3 are 12AU7, while V1 is 12AX7. It really matters. I often find them in the wrong sockets, and sometimes I just find three 12AX7s. That would not cause your plate/screen voltage inversions, but might be involved with the difference bias voltages.

Hum is often caused by imbalance in the output stage. A dead tube would cause that as would anything else making a substantial difference side to side.

And verify the OT wiring. On the power tube board, blue wire to J11 and blue-white wire to J12. The red-wwhite center tap on the OT primary connects to J13 on the main power amp board.

Make sure the main power transformer secondary wires are as follows. two red wires to J14, J15 on the power amp board, and the two blue-yellow wires to J16, J17. If you swapped the red pair with the blu-yel pair that might explain the 300v being 600v and the 600v being under 300v.

WHy your plate voltage is completely missing on the one side (13v instead of 660v counts as missing) is likely more a matter of either a missing connection or a burnt open trace between the transformer wire and the tube sockets or a shorted set of flyback diodes.

[tubeswell]

Thanks Enzo - I had J13 and J18 wrong way around. Duh!

[Enzo]

Cool.

[tubeswell]

So all the flyback diodes test okay on my fancy pants Cat IV diode tester - They each measure continuity in one direction only and open in the other direction, and the continuity measurements all sum for each set of 3 x flyback diodes in series. Not sure why the plate voltages on either side were so far out of balance at this stage (apart from having J13 and J18 cross-wired). All the plate connections on the output tube board test for DC continuity between each socket Pin #3 and the relevant PT primary spade connector on the output tube board. The OT primary measures 46.5R between each plate end, and 23.8R and 22.7R respectively between each end and the CT. THere is no DC connectivity between the OT primary and secondary windings.


I'll replace the blown Rg2 (R6) and test it again.

[TimmyP1955]

IIRC there are 10 Ohm plate resistors, and they are on the main board, not the socket board.

[tubeswell]

...
See the main B+ is 660-some volts to the OT CT, while the screen supplies are 340v-ish...

What B+ might one expect without any 6550s installed? (out of mild curiosity over the filter caps' voltage ratings)

This is what I get with the tubes installed (averaged of 2 sets of readings running the amp for about 5 mins for each reading)

Grid voltage -46V all round

Voltages Side 1:
Plate A: 670, Plate B: 672, Plate C: 678
Screen A: 380, Screen B: 396, Screen C: 390
Tube current A: 22mA, B: 20mA, C: 21mA

Voltages Side 2:
Plate D: 674, Plate E: 683, Plate F: 676
Screen D: 382, Screen E: 397, Screen F: 392
Tube current D: 23mA, E: 20mA, F: 20mA

All the bias LEDs are showing green, and not red

By my reckoning, the Ia:Ig2 ratio (from the info on the JJ6550 data sheet) is 11.7:1.

This would mean Tube A is idling at 13W, Tube B at 12W, Tube C at 12.6W, Tube D at 13.7W, Tube E at 12W, and Tube F at 12W.

The amp is nice and quiet with almost no hum and no unwanted noise.

Only thing I'm a bit concerned about is the highish readings of the screens. The tubes are JJ6550s

[Enzo]

I wouldn't worry that sides 1 and 2 differ in idle current by a couple milliamps.

I don't recall, on this unit have we removed the 22 ohm resistor and diode in the screen resistor positions? They should be replaced with 220 ohm resistors. (No diodes)

With tubes installed. Is that ALL the tubes? the screen node does seem high. I'd not worry too much about the caps, as long as their operating voltages are within ratings with tubes. Nothing says you can't up the resistor in the 345v node to drop the voltage some.

[tubeswell]

Yeah those readings are with all the tubes in at the same time. I guess the screens would drop a bit more by biasing the tubes hotter. The 22R have been replaced by 220R 0.5W fusible. The Rg2 bypass diodes have been removed.

It's interesting that the supply voltage for the screens (going to J18 - via D26 - on the main board) is listed as '380V' on the power amp schematic, but the screen supply voltage on the output tube schematic is shown as 345V.