A few updates & 10 ohm cathode resistors - now no audio.

GuyNoir

Hallowed State
Good Evening ~

I'm working on a Sherwood S-5000. Per forum recommendations, I placed individual 10 ohm resistors between cathode to ground on the outputs for the purpose of measuring current. Initially, I mistakenly placed 10K resistors, reading the dull red band as black. Audio was garbled. I realized the error and ordered 10 ohm resistors. In the meantime, I had also removed the flimsy phenolic speaker terminal strip, and replaced it with a stout terminal barrier strip. I can screw in 8-32 threaded banana jacks, and they are rock-solid.

In the process of soldering up the new terminal block, I didn't put the DF (damping factor) ground terminal back in. I wired the respective OT black leads to the common terminals. In addition, there were two leads coming from the 7199 tube socket center pin, which I also soldered to the common terminals. The 4 and 8 ohm taps were straight forward, but the 16 ohm taps had leads coming from both the OT and smallish green wires coming from an area near the 7199s. There is just one black lead that isn't connected to anything now, and that's a single lead that went from a can cap negative terminal to the "DF" terminal.

Also, I replaced the 4.7K dropping resistor on the bias supply with a 3K, which gives me nearly -20 volts at the output tubes, as opposed to -17.5.

Now, I have no audio. The bias balance pot is working, and in test mode I get solid hum that can be nulled with the pot. I'm showing about -19.5 bias voltage to each of the four outputs. For current, I am reading 3.2 on the ammeter scale's second to lowest range, except for one tube that is nearly double that. I've ruled out anything on the amp's front panel - I know it by heart now, and everything is set as it should be.

One other thing is different now; the output tubes are seeing 460v on pins 7 and 9, whereas it was 438v not that long ago, both measured at line voltage.

Any ideas? (yeah, I know...stop mucking around) but other than that?

TIA,

Cary
 
With -20V on the grids of the output tubes, the output stage might now be severely over biased. That could explain the higher than expected voltage on the plates and screens. First thing I'd try is to get the bias supply voltage back to what it was before. That may or may not correct the no sound issue, but it should at least get the plate and screen voltage back where they were before.

What is the voltage reading across each of the 10 ohm cathode resistors you installed? For this amp, I think you want a reading between say 200 to 240 millivolts or 0.20 to 0.24 volts.
 
It looks like I do in fact have 200 to 220mv on three tubes, but one of the tube sockets is intermittent and reads 0, unless I rock the tube in which case kit temporarily spikes. Last week that pin literally broke off while desoldering. I replaced it with an unused pin, but it's just not a solid connection now. TubeDepot has some NOS 9 pin Noval sockets that appear to be an exact match, so I'll replace that before proceeding. Regarding the bias, I have read on a Sherwood restoration forum that bias should be around -20 to increase tube longevity and for the tubes to run a little cooler. I thought the implication was that with lower negative voltage, the tubes ran hotter. Perhaps I am mistaken.

From the article:

"....Ok, now the amp fires up without any drama. As I expected before, my bias voltage has now risen up and is now -24.85v, about where I expect it to be with silicon diodes and no dropping resistor. Yes, you can get a couple more watts out of this thing with a lower bias, but the wear and tear on the output tubes and the heat generated makes it scarcely worth it, IMHO.

The DC filament voltages are now 26.1 and 12.7 volts at V2 and V1 respectively, up from 11.06 and 22.74. These might be just a touch high, but they're within the realm of sane. I may in fact try to put a slightly higher value dropping resistor in front of these."

Also:

"There are no real test points listed in the bias supply other than the final ouptut voltage, which is supposed to be -20v. This amp is currently putting out -21.8, which is probably right on target considering the plate voltage is running a tad high - the extra bias voltage is welcome. In fact, I'd like to see it a little higher, and once the selenium rectifier is swapped out for silicon diodes, I'll expect to see closer to -23v. A real indication of the health of this circuit will be the DC voltages at the phono tubes, since these actually have some current draw to them "
 
I know it is too late, but when makiing changes to a circuit it is best to do one side of the amp and leave the other side intact. Especially if you are more of a novice. It is very easy to make mistakes and in this way you will always have the untouched as a reference, as long proper operation was ascertained before modding. We have all been where you are now, and it is best to check the work that you did recently extremely carefully for errors. If you don't have any audio it must go to a point where all the audio signal is affected, maybe the damping factor circuit.
 
Ah okay. If you are indeed getting 200 to 220 mV readings it means bias voltage is about right. Yes "lower" bias voltage, meaning a less negative value, or in other words a bias voltage closer to zero, will make the tubes idle at higher current. The no sound issue must be related to other recent changes then.
 
I know it is too late, but when makiing changes to a circuit it is best to do one side of the amp and leave the other side intact.
[SNIP] If you don't have any audio it must go to a point where all the audio signal is affected, maybe the damping factor circuit.

You are so right! I wish I'd left well enough alone. It's just that the flimsy phenolic board didn't play nice with modern speakers cables. Oh, I could have used some simple lamp cord instead of very heavy, doubled up Kimber quad cable, but the audiophool in me won out. Luckily I have pictures and I can restore the DF circuit and have enough terminals on the barrier strip to do so.

The no sound issue must be related to other recent changes then.

Hopefully, it involves the DF circuit.


I'll put it back. It went to the DF terminal with the two OPT common wires. From the DF terminal, a .47 ohm resistor was soldered to each of the speaker common terminals. Other than the resistors, the common terminals had only wires coming from the 7199 center pin grounds. With the jumper across all three, the whole mess would be tied together, negating the resistors.

Conor mcgregor khabib

Yes, I'm getting my ass kicked. :rflmao:

eljOsEGw.jpg


Thanks, gentlemen!
 
BINGO! 6DZ7 for the win! It wasn't tied to the chassis - it was an isolated ground (I think). I didn't actually try an VOM between the two points. But I've jumpered that one wire to the common terminals, and I'm hearing beautiful music! Thank you sir.
 
Prarie Home Companion? That'd be great! I jumpered the com just before showering and heading to work. Of course, it's all I can think about now. Peace of mind knowing that it's working though. I actually have two wonky output sockets I want to replace before any serious listening.
 
Prarie Home Companion? That'd be great! I jumpered the com just before showering and heading to work. Of course, it's all I can think about now. Peace of mind knowing that it's working though. I actually have two wonky output sockets I want to replace before any serious listening.
Hey Guy, listening to that Steely Dan GH album right now! I always use them for amp testing. They always have good production.
 
I once read that someone asked Donald how they get that sound. He replied "lots of EQ and compression." Can you imagine if they went straight to board? I suppose it would sound pretty lifeless. There's a great documentary on YouTube about the making of Aja. If you search for 'Steely Dan' and set the length filter for long, it's a 45 minute video. Very enlightening and enjoyable.
 
Back
Top Bottom