Silvertone 1474 twin twelve intermittent output

[tibbar6]

Hi everyone, i am new to the forum and have been reading through a bunch of old posts regarding silvertone 1484 amps with output problems.
I have a general electronics background, have built amps and serviced many fender and silvertone amps before and havent really run into an issue quite like this one.

Basically the amp will run for about 5 minutes and sound completely normal. Then it will drop of in volume, and then slowly start to fade back in as you keep playing. I also noticed that the v1 and v2 preamp tubes fillaments arent lighting up at all, however their filaments read around 11v and they still make sound. Is this normal? I know this amp employs an unusual filament wiring.

I have tested every tube with a B&K tube tester. All of the molded/paper and electrolytic caps appear to have been replaced before i received the amp. I also have cleaned every tube socket with deoxit and replaced the 200ohm 5 watt resistor that attaches between the rectifier tube and the filament wire (it was crumbling apart and measured about 40ohms on a meter). Speakers have also been replaced and the reverb tank has been disconnected.

Any guesses as to what could be causing the amp to act up like this? I hear that the output transformers are notorious for going on these amps so that will probably be the next thing i test.

Thanks!

link to schematic:

http://elektrotanya.com/PREVIEWS/634...1474.pdf_1.png

[52 Bill]

Welcome to the place.

Rather than looking to solve your problem by replacing things, try and solve it by measuring things when it works and then when it stops working.

Start by trying to isolate the problem to different sections of the amp. When the signal dies is it the preamp or the power amp that kills the signal?

You mentioned the preamp filaments being odd. They are dc powered from the cathode voltage of the output tubes. Does the 11 dc volts stay steady when the signal fades in and out? Every tube is constructed differently, the filaments may not be visible on the tube that you have.

Do all of the dc supply voltages remain steady when the signal fades?

[tibbar6]

Thanks for the help bill, I followed your advice and measured the voltages and found that the 350v line coming out of the choke varied with the fading signal. The voltage before and after the choke measured rather high at 400v (spec is 170v before and 350v after). The signal at the next test point in the line vaired between 150v and 250v (spec is 335v). As I tested further down the line the voltages continued to be lower than spec and varied by about 100v. It also appears that whoever serviced the amp in the past replaced the electrolyic caps with ones that are way off spec. The caps that read 40uf, 16uf and 10uf on the schematic have all been replaced with 220uf capacitors. Could this be causing the choke or high wattage resistors to fail or act erratically? Seems like 220 is rather high. I also tested the choke and resistor values and they appear to have drifted upward in value by around 7-10%

[Enzo]

As long as those 220uf caps have enough voltage rating, it won't kill your sound. It might affect the stiffness of the bottom end or the sustain or something, but won't stop it from working. I don't care if your resistors or caps are 20% off, it will still work. it might sound different, though not a lot, but it won't come and go, just because some resistor is off value. Values are not critical, these are guitar amps, not NASA space laboratory gear.

Tube testers are just a gross measure, they are good at telling you a really bad tube is really bad. But what they can't do is tell you a tube is good. If I catch a guy robbing a bank, I know he is a bank robber. If I meet a polite gentleman on the street, I have no way to know if he robs banks or not.

Tubes work in sockets, so the best tube in the world can be plugged into a faulty socket, and so not work.

Fading in and out usually suggest heaters. Heaters when power is removed cool down over a few seconds, so the audio fades away. Then when the heater power returns, they fade back in over a few seconds. Loose connections in the circuit usually result in off/on symptoms. Of course the central adverb there is "usually."

Plug something like a CD player or a stereo receiver into the amp, so you don't need to hold a guitar. If the sound goes, try whacking the top of the amp with your fist. Or a rubber mallet if you are shy. Does that make it return? Doe whacking a working amp make it quit? That exposes loose connections. OK, you COULD use a signal generator, but who wants to listen to that?

If a tube makes sound, it has a working heater, regardless of how it looks. So look closely, even turn out the room lights if you are unsure. But once on the beam there, look closely when the sound quits. Do any of the heaters get bright and dark as the sound comes and goes?

B+ varying widely is consistent with tubes not conducting, and not conducting is a sign the heaters are off. The voltage drop across the B+ resistors comes from the tube current through them. No conduction, no current. No current, no voltage drop. So if the heaters fade out, then the B+ will fade up.

Pick a triode, like maybe the input stage. The B+ for that stage will be something like 300v, and right at the plate, will be maybe 200v. I am making up numbers. When current flows through the tube, then there will also be a voltage on the cathode, typically a volt or two. When you sound fades out, does the couple volts on the cathode fall to zero? That will tell us the tube stopped working. The cathode voltage trick works on most any tube in the amp, except grounded cathodes on power tubes.

[drewl]

I've seen many of these with intermittent tube sockets, usually the preamp sockets.

[tibbar6]

Good tip with testing the voltage from the cathode. When the volume cut out, the voltage dropped from 1v to about .15v or even zero on the v1 and v2 preamp tubes. If its a socket issue on v2 that would explain why the filaments arent lighting up/ heating up. Ive tried give the amp a good whack and ive tapped just about every joint with a chopstick too, but that didnt reveal anything abnormal.

[Randall]

When current flows through the tube, then there will also be a voltage on the cathode, typically a volt or two. When you sound fades out, does the couple volts on the cathode fall to zero? That will tell us the tube stopped working. The cathode voltage trick works on most any tube in the amp, except grounded cathodes on power tubes.

This, I like! My 'learn something every day' quota is now satisfied for today.

[g1]

My 'learn something every day' quota is now satisfied for today.

And another one here that I've never heard of before:

You mentioned the preamp filaments being odd. They are dc powered from the cathode voltage of the output tubes. Does the 11 dc volts stay steady when the signal fades in and out?

I'd guess with the wide variance of modern tubes, every time you changed power tubes there would be a bit different heater voltage at V1 & V2.

[tibbar6]

I went ahead and replaced the V2 socket and the amp is still behaving in the same manner. The Heater lines on V1 reads about 5volts V2 reads about 11volts and each 6l6 reads about 15volts. These all begin to drop by a few volts as the volume fades out. The B+ voltage along the amp is reading at about half the voltage shown on the schematic when measured from after the first resistor. Any thoughts?

[DRH1958]

When the B+ drops after the first resistor, what is the raw B+ coming off pin 8 of the 5U4? Does it drop too, or stay close to the 370V? If it drops too, I wonder about a partial short in the power tranny. The 6L6s have their own 6 volt supply, so it seems as though as the tranny heats up, a short or partial short develops. This would explain the 6L6s heater increasing to 15 volts and the B+ decreasing. The only common element between the B+ and 6 volt heater supply is the PT. For reading the 6L6 heater voltage, are you right on pins 2 and 7?

When this happens, what does the 5 volt DC reading on pins 2 an 8 of the 5u4 do? Also, the 360 VAC on pins 4 and 6? 360v to ground and 720v between pins?

I do hope there is a 3-wire grounding cord replacing the old 2-wire power cord.

[g1]

On the 6L6's, is that 15V at the cathodes or the heaters?
The V1 & V2 heaters are connected to the 6L6 cathode circuits.

[DRH1958]

On the 6L6's, is that 15V at the cathodes or the heaters?
The V1 & V2 heaters are connected to the 6L6 cathode circuits.

Exactly. That's why I asked if he measured right at pins 2 and 7 of the 6L6s. If the B+ drops, that would be why the heater voltage drops on the 12AX7s. Cathode current drops, cathode voltage drops. Otherwise, the only thing could be the PT.

[tibbar6]

sorry, when I said there is 15v on the heater, i meant the cathode (pin 8) of the 6l6 tubes that goes out to the heaters of v1 and v2 as well as the 6au6 tremolo tube.

Pin 8 on the 5U4 is staying steady at 433 volts.
Pins 2 and 8 of the 5U4 are also steady at 5.1 volts.
Pins 4 and 6 measure 730 volts
Pin 4 to ground measures 360 and so does pin 6
The heater measured between pin 2-7 on each 6L6 tube is about 6.4 volts.

[DRH1958]

Next might be to check pin 3 plate voltage on the 6L6s and then check the screen voltages at pin 4. We need to figure out what is causing the cathode current to drop. I'm wondering about that first dropping resistor, the 1.5k 1 watt. It must be bad since you have such low voltage after it. You said it's half of what's on the schematic. That's even worse because you have 433v at pin 8 and the schematic calls for 370v, so your B+ voltages will be higher if the B+ supply is functioning normally. Also, do you have about 420v on the load side of the choke?

Really when something like this is happening, it's good to check all the B+ voltages and see if they are consistent with the corresponding point on the sockets. It's good that Dano is generous with his voltages on the schemo.

[Enzo]

If we think the tubes are losing heater current, we are ONLY concerned over the voltage across the heater, that is one end to the other, as in pins 2 and 7 of a 6L6. We are not interested in voltage to ground. heater elevation is a hum abatement circuit, if it has issues, we get hum, not loss of signal.