• 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.

newbie question about 7591 cathode bias

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello all I first want to thank the people who have helped me in the past with all my projects I couldn't do any of this stuff with out YOUS GUYS.
OK here is my situation got my new weller soldering station station today
and started in on a eico st40 I have replaced the coupling power supply cans and the 25v caps in the bias . Have also replaced the 1.8k and 8.2k
resistors in the power supply I droped in a new matched quad of jj tubes
the 7591s version and fired it up on a variac over about an hour I only brought it up to about 110 volts and played sum cd's for awhile sounded nice no red spots on the tubes no pops no spits so I then pluged it in to the wall and powerd it up ran fine for about 45min and I started to see real faint glow on one of the output tubes OH NO lol now yall are gonna haft to keep mind I have basic understanding of this stuff I have a rca tube book of essential characteristics with this book close by I measured the voltage at the wall and found it to be 124v I called a friend across town from me and asked him to get a reading out of his wall just for comparison and he said 118v so my ac is a little hot out here. SO I put the eico back on the work bench and powerd it up and im getting 418v at
the 1st filter cap after the 1.8k resistor im getting 370v I then measured
the voltage at the plate and was getting 411v then I multiplied 411 by .05
for the dissipation and I came up with 20.55 i then looked up the data sheet in my rca manual and on jj's site for this tube and both are rated for 19watts dissipation did I do this correctly ? and if so should I change value of the 1.8k resistor in the power supply or should I change the resistance in the bias network ? The tubes that were in the amp were the original westing house 7591's I put those back in there they test marginal for emissions on my sencore and they test almost marginal on me eico 666 for conductance but they dont run hot like the jj's do those old tubes are pretty STOUT. Any ways the st40 was bought off my uncle and it is the first peice of tube gear Iv ever had I ran it on a pair of vandersteen model 1's for a few years with no problems and then got my hands on 2 pairs of quad esl57's that are double stacked and nothing else I have drives the quads like the st40 does I have a ear 834 and its so sloppy and muddy compared to the st40 when driving the esl's. The quads with there impedance swings especially 2 per channel in parallel have taken a toll on the 7591's and I knew I couldnt keep running it with all the old caps and stuff in it so this is where Im at every thing I have replaced in the amp is of the same values and voltage rateing I have not done the loudness mod but I plan to after I get these output tubes under control.

The work you've done on your Eico ST40 is correct and commendable. As i see it, your problem with a tube getting a little red is two fold. First is the 124 volt line, assuming your voltmeter is correct on AC. (some of them aren't) The second part is modern Chinese tubes.

High line voltage in this country is all to common a problem that should not be. Although there are reasons why this is so that have to do with modern demand on an outdated infrastructure. And modern tubes are not made to the strict specifications they once were. What was once a specification has morphed into only a rough guide line.

Your calculation for plate dissipation is nearly correct. You only need to take into account the voltage loss across the cathode resistor. Even so, the tubes are running at maxium which will shorten their life by running hot. Your old 7591s being a little weak, but of better original quaity, prevents them from running red.

You will need to control this either by raising the bias voltage or dropping the overall input voltage to the unit. Since raising the bias is easier, increasing the cathode resistor will do it. Try raising it 10 or 15%. I wouldn't change the values of the two resistors you mentioned (1.8K and 8.2K) as this will alter the operating points of the rest of the circuit. While this would lower dissipation by dropping plate voltage, it is a more extensive modification. You might want to use a 50 volt capacitor if the voltage across the cathode resistor is too close to 25v.

If you have the room on the chassis you could mount a filament transformer to buck the primary of the power transformer. A 12 volt transformer would give you about a 10% reduction. (a 6.3v about ½ that) You'd have to phase it right so as not to boost it. I assume you know what I mean by this. If not, it's putting the filament transformer secondary in series with the main transformer primary, and connecting the filament primary to the line out of phase to appose.

One more thing, your post was a little awkward to read. Please try using periods to end sentences. Also, breaking thoughts up into small paragraphs makes things a lot easier to read. Thanks.

edit: One last thought. Check the coupling capacitors for any DC leakage to the output tube grids. This would cause a tube to run hot by lowering the bias on that tube. Since you've done a lot of replacing, perhaps replace the coupling caps too even though they seem to be ok with a lowered line voltage.

Thank you Victor I have had a few to drink tonight sorry about the poor punctuation . I have replaced the coupling caps and I felt really stupid after making this thread all I had to do was like you said, raise the bias. I have some 160ohm at 10 watt here and put those in and it seems to have done the trick. As for the the filiment transformer I guess that would allow control over line voltage ?
As for the the filiment transformer I guess that would allow control over line voltage ?

That is correct. If the 160 ohm resistors have corrected the overheating problem, then don't use the filament bucking transformer. But if you find that the tube filaments are running at more then 10% overvoltage, you might want to reduce that too. Either with a low value resister in the filament circuit, or control it through the mains primary.

Running the 7591 output tubes at 50mA is a little too high. They want to see more like 40 to 45mA max. So you'll need to strike a happy mid point between line voltage and/or output tube current.

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.