Power Transformer Problem Tube Rectifier

[Ed_Chambley]

One of my amps had a loud hum, fried a rectifier tube and blew the fuse.  Installed a good tube and switched back on the loud hum and blown fuse.  I removed all tubes to get measurements.  The PT is the one Doug sells for Super Reverb and it is only about 3 months old.  Connected with the tube in I am getting 5.7 volts on pin 2 and 0 volts on pin 8.  I disconnected both yellow wires and I am getting 11.89 volts and on the other I get 7.64.  The red wires both connected I am getting the correct voltage.  Is there anything which could cause this other than a bad winding?

[sluckey]

I disconnected both yellow wires and I am getting 11.89 volts and on the other I get 7.64.

Are you saying that you have one meter lead on chassis and are checking for voltage on each of the yellow wires in respect to chassis? If so, those are bogus readings. The correct way to measure voltage on a winding that's totally disconnected is to connect the meter across the two leads, ie, measure voltage from one yellow lead to the other yellow lead. Bet the reading will make more sense.

Your loud hum and fuse blowing is likely a short between the B+ rail and ground, possibly a shorted filter cap. Disconnect the wire leaving the rectifier socket (should be pin 8) and heading toward the STBY switch or filter cap. Does it still blow the fuse?

Has this amp ever worked?

[Ed_Chambley]

I disconnected both yellow wires and I am getting 11.89 volts and on the other I get 7.64.

Are you saying that you have one meter lead on chassis and are checking for voltage on each of the yellow wires in respect to chassis? If so, those are bogus readings. The correct way to measure voltage on a winding that's totally disconnected is to connect the meter across the two leads, ie, measure voltage from one yellow lead to the other yellow lead. Bet the reading will make more sense.

Your loud hum and fuse blowing is likely a short between the B+ rail and ground, possibly a shorted filter cap. Disconnect the wire leaving the rectifier socket (should be pin 8) and heading toward the STBY switch or filter cap. Does it still blow the fuse?

Has this amp ever worked?

Yes since 1966 it has been working.  I was playing it last night and had a big hum and the the fuse popped.  I removed the wire going to the standby and no blown fuse.  So either a bad filter cap or a wiring problem?

EDIT by sluckey... fixed quote

[HotBluePlates]

Connected with the tube in I am getting 5.7 volts on pin 2 and 0 volts on pin 8. 

Stop and consider a hypothetical case:

The Super Reverb's 5v winding is connected only to the rectifier tube's pin 8 and 2. There is no ground reference, and no other connection to the "outside workd" except the wire from pin 8 to the filter caps. You measured (presumably from pins to chassis) 5v+ at pin 2 and 0v at pin 8.

Hypothetical: If the B+ rail is shorted to ground, pin 8 is therefore connected to ground. It would measure 0vac between it and ground; however, pin 2 would measure the full voltage of the 5v winding. A shorted B+ rail would also blow the fuse.

Your situation seems to match the hypothetical. You mention the transformer is 3 months old, so that would indicate recent tinkering. If the amp was not blowing fuses prior to tinkering, suspect anything/everything involved in the recent changes. A B+ short to ground could be a failed filter cap (if it failed short rather than open), but could also be a solder blob connecting a plate/screen pin to something else (probably ground). It could also be a bare strand of wire at a tube socket contacting the chassis, a solder blob on the filter cap board, or a fault in the wiring from the cap board to the main eyelet board.

If you first inspect everything recently tinkered, you will probably reduce your troubleshooting time.

Also, Sluckey's initial guidance to disconnect the rectifier from the "outside world" (the wire from pin 8 to the cap board) would test/eliminate a failed rectifier tube as the cause of the blown fuses.

If you inspect all wiring/solder joints and find no faults, you might try the following to isolate the failed cap. Unsolder the negative lead of all the caps from the grounded eyelets on the cap board (be VERY careful!!). Apply power. Fuse good? Re-connect the first filter cap pair, reapply power and see if the fuse blows. Proceed one at a time, reconnecting caps until you find the one that blows a fuses (or make a lightbulb limiter shine brightly). Replace the obviously failed cap.

Of course, if all the caps are the same approximate age (and known-old) you might practice preventive medicine and replace them all. This approach will not solve a shorted B+ due to solder/wire fault/error, but it might highlight that a cap is not the source of the short.

[Ed_Chambley]

Connected with the tube in I am getting 5.7 volts on pin 2 and 0 volts on pin 8. 

Stop and consider a hypothetical case:

The Super Reverb's 5v winding is connected only to the rectifier tube's pin 8 and 2. There is no ground reference, and no other connection to the "outside workd" except the wire from pin 8 to the filter caps. You measured (presumably from pins to chassis) 5v+ at pin 2 and 0v at pin 8.

Hypothetical: If the B+ rail is shorted to ground, pin 8 is therefore connected to ground. It would measure 0vac between it and ground; however, pin 2 would measure the full voltage of the 5v winding. A shorted B+ rail would also blow the fuse.

Your situation seems to match the hypothetical. You mention the transformer is 3 months old, so that would indicate recent tinkering. If the amp was not blowing fuses prior to tinkering, suspect anything/everything involved in the recent changes. A B+ short to ground could be a failed filter cap (if it failed short rather than open), but could also be a solder blob connecting a plate/screen pin to something else (probably ground). It could also be a bare strand of wire at a tube socket contacting the chassis, a solder blob on the filter cap board, or a fault in the wiring from the cap board to the main eyelet board.

If you first inspect everything recently tinkered, you will probably reduce your troubleshooting time.

Also, Sluckey's initial guidance to disconnect the rectifier from the "outside world" (the wire from pin 8 to the cap board) would test/eliminate a failed rectifier tube as the cause of the blown fuses.

If you inspect all wiring/solder joints and find no faults, you might try the following to isolate the failed cap. Unsolder the negative lead of all the caps from the grounded eyelets on the cap board (be VERY careful!!). Apply power. Fuse good? Re-connect the first filter cap pair, reapply power and see if the fuse blows. Proceed one at a time, reconnecting caps until you find the one that blows a fuses (or make a lightbulb limiter shine brightly). Replace the obviously failed cap.

Of course, if all the caps are the same approximate age (and known-old) you might practice preventive medicine and replace them all. This approach will not solve a shorted B+ due to solder/wire fault/error, but it might highlight that a cap is not the source of the short.

Of course there is recent tinkering.  I only have 2 amps I do not tinker with.  My main amp and its back-up.  Thanks for the troubleshooting procedure.  Recent tinkering was re-voicing the normal channel.  I have checked the wiring twice and will again.  I changed caps and added shielded wire to V1, pin 2 and 7 I believe.  I am not looking at the schematic.  There is a mod in the Tube Amp Book whereas you can make the normal channel tweedish.  That is the mod.   It could easily be s solder blob.

The filter caps were changed at the same time I changed the PT, so they are also 3 months old, so are the bypass caps and a few resistors that have drifted.

Thank you for your hypothetical, it makes sense especially replacing the filter caps one at a time.  Seems though if the problem were in v1, pulling the tube would solve it.  Gotta do some more checking.

[HotBluePlates]

Seems though if the problem were in v1, pulling the tube would solve it.  Gotta do some more checking.

Not necessarily.

If (for example), the shielded wire is ground at the input jack (or volume control), and on the tube end there is a strand of shielding contacting pin 1 (or 6), then there is a low-grade short through the plate load resistor. It will still exist after the tube is pulled from the socket.

Such a partial-short probably won't blow fuses (fast anyway), because there would be 100k between B+ and ground. Seems like that would overheat the resistor, maybe not even burning it (depending on wattage rating).

But the replacement of the filter caps suggests possible solder drips or melted insulation in those wires. That's a good candidate area to inspect.

Try the test with disconnecting the wire from the rectifier's pin 8 to the filter cap board. If the fuse holds, try measuring resistance from B+ rail to ground (should be ~440k, due to bleeder resistors). If you measure 0 ohms, you know there's some solid short to the chassis. If you get ~470 ohms, you might suspect a short from screen resistors to ground. ~100k might suggest a partial short of bleeder resistors, or the error I described with a strand of shield contacting a preamp plate pin.

You just have to go looking for clues, and see what they suggest. Some might be misleading, until you really understand what happened.

[Ed_Chambley]

I did remove the filter caps negative end and before connecting any back I turned on the amp.  I still had the hum and blew the fuse.  Actually I am not blowing fuses, I have installed those breakers weber sells along with the fuse.  I do it in all my amps as it seems like a good idea.  They trip easier than a sloblow fuse and with all the tinkering I do, they are a great investment.
Next I will measure the resistance between the wire leading to standby and the chassis.  I know the breaker will hold as I have already disconnected it.

[HotBluePlates]

What about the test for the rectifier tube? The breaker held with the rectifier installed but disconnected from the rest of the B+?

[sluckey]

Yes, he tried that back in reply #2, but it was kinda hidden in the tangled quote. I just now fixed the quote.

[Ed_Chambley]

tested wire to standby to chassis. -395ohms.  This is where I am currently.

[HotBluePlates]

If you unsolder all the filter cap negative leads, and still get 395, I'd look at the screen resistors. Specifically, if the tube socket pin 4 is shorted to ground, you'd measure 470 ohms (+/- tolerance) from B+ to ground.

[sluckey]

Well, that's likely the problem. Now you just got to sniff out the culprit. First, examine all wiring on the B+ rail to look for frayed wires, solder blobs, etc.  Also look at  OT plate wires and tube socket plate/screen connections, and screen resistors. Since the amp has been working I don't really suspect this but it only takes a minute to rule it out. Likely suspects are the first and second filter caps even though they are relatively new.

Next, disconnect the positive end of the first filter cap and recheck resistance from STBY switch to ground. If resistance reading goes way up, you probably found the culprit. Confirm by measuring resistance directly across the cap (positive end still disconnected).

Repeat for the next filter cap which should be the screen supply node.

It's not likely that a cap failure further down the B+ rail will cause a 400Ω reading at the STBY switch since there will be dropping resistors between them and the switch.

[sluckey]

BTW, which amp are we talking about? You mentioned 1966 so I assumed a factory built amp but am not clear about the model? It would be nice to refer to a specific schematic.

[Ed_Chambley]

BTW, which amp are we talking about? You mentioned 1966 so I assumed a factory built amp but am not clear about the model? It would be nice to refer to a specific schematic.

AB763

[sluckey]

AB763

But which one? There are about a dozen different ones and  the power supplies vary slightly between models.

[HotBluePlates]

He implied Super Reverb, based on the type of PT has said he has.

But you're right that it doesn't necessarily mean that is what the amp is. I assumed that's what it means, after the statement the amp was from '66.

[Ed_Chambley]

AB763

But which one? There are about a dozen different ones and  the power supplies vary slightly between models.

66 Super Reverb.  I checked the first filter cap reading -410 ohms from standby to ground.  I do not get a reading across the cap itself.  I do not have a replacement to check it, but I think this would be the problem since you guys say so.  If you would, Please explain why if positive is disconnected, resistance increases from the B+ rail to ground?

If this is the problem it is the second like new Sprague Atom filter cap that has gone south on me in less than a month.  I think it is time to change brands.  I am going to order a replacement for this cap and I have an old 5f4 I am getting ready to recap.  Is there any benefit in increasing the cap values in a 5f4 like there is in a Super Reverb or just go with the 16uf like is in there?

[sluckey]

Any progress?

[Ed_Chambley]

Any progress?

I just received the caps moments ago.  I will know later tonight.  Thanks for asking.  While I got you here, I am recapping a 5F4 soon.  Is there any benefit in increasing the filtering from 16uf to say 22uf or more in the first filter cap.  I know increasing in a Super Reverb makes the bass much tighter.  I like it.

[Ed_Chambley]

Replaced filter caps.  Checked the ones I took out as I have a farad meter.  All checked ok.  Replaced them anyway.
Lifted all caps and checked them, all are within spec.  Replaced some anyway.  Kept power tubes out.  Lost another rectifier tube.  Checked choke wire to wire unhooked and got .005 ohms.  Removed choke from chassis, no change.  Checked OT got .083ohms from center to blue and .074 from center to brown.  I did not disconnect the OT.  Disconnected wires to flip up board.  All underboard wire checks ok.  Changed power cord.  Traced every wire and checked solder joints, all ok.  PT gives me 330 at rectifier.  Bias tap -57.  Checked all tube sockets, all look good.  Heaters check fine.  What else could cause a crackle sound and then popped fuse?  I am at a loss.  Maybe something I cannot see in the rectifier socket?  I am getting very close to a complete rebuild.  One thing did notice is the rectifier tube is getting too hot.

[sluckey]

What else could cause a crackle sound and then popped fuse?  I am at a loss.  Maybe something I cannot see in the rectifier socket?  I am getting very close to a complete rebuild.  One thing did notice is the rectifier tube is getting too hot.

You probably still have that 400Ω resistance on the STBY switch. You've already eliminated the rectifier when you disconnected the wire from pin 8.

I checked the first filter cap reading -410 ohms from standby to ground.  I do not get a reading across the cap itself.  I do not have a replacement to check it, but I think this would be the problem since you guys say so.

Since you don't get a reading across the disconnected cap the 400Ω is not due to that cap. Did the 400Ω reading go away when you disconnected that first cap?

Exactly which caps did you replace? What value and voltage ratings?

[Ed_Chambley]

What else could cause a crackle sound and then popped fuse?  I am at a loss.  Maybe something I cannot see in the rectifier socket?  I am getting very close to a complete rebuild.  One thing did notice is the rectifier tube is getting too hot.

You probably still have that 400Ω resistance on the STBY switch. You've already eliminated the rectifier when you disconnected the wire from pin 8.

I checked the first filter cap reading -410 ohms from standby to ground.  I do not get a reading across the cap itself.  I do not have a replacement to check it, but I think this would be the problem since you guys say so.

Since you don't get a reading across the disconnected cap the 400Ω is not due to that cap. Did the 400Ω reading go away when you disconnected that first cap?

Exactly which caps did you replace? What value and voltage ratings?

I replaced the 3- 20uf 500VDC Sprague Atoms, 2- 80uf 450VDC F&T in the doghouse, also the resistors, 1- 100uf 100VDC on the tagboard.  I did not replace any of the 25/25's.  I did replace 2 of the .022's since I had some.  Last time I did a cap job, I did not have any new ones, so I left the old ones in there.

I have not checked the 400ohm reading, I connected back to the rectifier.  This is why I lost another tube.  It is strange.  I push in the reset, flip off the standby.  Hum, crackle and within 2 to 3 seconds it trips the breaker.  I did not check for the ground short again, I guess I should have.  I guess that is the next thing to do as I need to get a 0ohm reading from the disconnected B+ to the chassis.
I have a lot of trouble with naming caps in an amp, also parts.  Things like cathode follower, bypass cap.......Is there a good resource of reading for this.  I can look at them and know what they do, but I have trouble communicating.  I have read the hip book of Gerald Weber and have his videos.  I also have the Tube Amp Book, Pittman from Groove tubes.  I have various other pdf downloads I have read.  I would really like to learn to communicate tube language better.  Do you have any recommendations?

Also, I use Mac computers and have found a program called Eagle, but many speak of Express. Would it be best to change over to a windows machine for this type of work?

[sluckey]

I did not check for the ground short again, I guess I should have.

Yes, check for that! No need to turn on any power again until that is fixed.

I guess that is the next thing to do as I need to get a 0ohm reading from the disconnected B+ to the chassis

No! you don't want 0Ω reading. That would be a dead short! You want a high reading, hopefully 100K or higher.

I would really like to learn to communicate tube language better.  Do you have any recommendations?

Yes. Hang out in this forum. Read all the messages. Ask lots of questions.

[Ed_Chambley]

I did not check for the ground short again, I guess I should have.

Yes, check for that! No need to turn on any power again until that is fixed.

I guess that is the next thing to do as I need to get a 0ohm reading from the disconnected B+ to the chassis

No! you don't want 0Ω reading. That would be a dead short! You want a high reading, hopefully 100K or higher.

I would really like to learn to communicate tube language better.  Do you have any recommendations?

Yes. Hang out in this forum. Read all the messages. Ask lots of questions.

So in the beginning I was getting a -410, so now I am looking to get a positive reading >100K?
I have been reading the geofex debugging page.  That is where I got the information on checking the choke and PT.  Why is it not necessary to unhook a PT to check to see if it is open and inversly why is it necessary to unhook the choke?

Also, I did not mention this yet, but one of my 5f4's did the same thing last night while playing it.  I have not even looked at it, I assume after I get through this I will be able to fix the other.  I simply switched over to my Stuckey moded JTM-45/plexi.  That thing turned out really nice and I forgot to thank you for that.  I installed a precision dual pot 500k master and that thing will really cover lots of tones.

[HotBluePlates]

...
I have been reading the geofex debugging page.  That is where I got the information on checking the choke and PT.  Why is it not necessary to unhook a PT to check to see if it is open and inversly why is it necessary to unhook the choke? ...

The way the choke is positioned in the circuit makes it easy to get misleading readings unless you unhook at least one of the choke's wires.

More sources of info are better, but they can cause you to go in circles. The problem is without a specific repair case to describe, most of these sources are just a list of "things to check." The trick is you can spend 5 days checking/measuring everything in the amp, and not actually look in the area where the trouble lies.

Sluckey and I are trying to lead you through a logical troubleshooting procedure, with specific questions (or required readings), and specific next-steps based on your answers. We might save you some time and money replacing parts that don't need replacing.

[Ed_Chambley]

Thanks for all the assistance.  The problem is solved and I am not sure what the problem was.  I did find where a large solder pool on the chassis that had broken loose.  Since I did not power it up, I am not sure it was the problem.  I did not have an iron that wuld get hot enough to reattach it, so I replaced it with a bolt, lock-nut and solder eyelets for each of the wires.   Knowing there was a short somewhere that I could not find, I began looking at wires and solder joints.  Since the amp is a 66, the wires are/were old and some brittle.  I heated and removed solder and wires until I could remove the board. I changed wires and removed excess solder blobs where there was buildup over the years of repairs, recapping and whatnot.  While this was out of the way I removed all the pots and the brass plate.  Scrubbed and cleaned and made all new solder connections.  This also gave me the opportunity to install a pot/switch on the tremelo that I have been wanting to do so I can turn it on and off.  I had a couple of pots that had drifted and a few resistors that had drifted as well.  Those got replaced.  Since I had a fairly clean eyelet board, I brushed on a little rosin to the eyelets.  I found a cracked tube socket, v1 between pin 4 and 5.  I replaced.  I put the filter caps back in that I had replaced as they were only about 3 months old.  I did replace much of the wire.  Did the same with the tube socket wires.  I probably did way more than necessary, but it is a very very nice amp so I felt the effort was worth it.  Hooked up my MM and the ohms began climbing.  Plugged the amp into my variac and began bringing up the voltage.  No noise, no hum.

I really appreciate all the information and directions you guys gave me.  All it cost was some wire and time and this actually caused me to service the amp the way it should have been done long ago.  It looks very nice on the inside.  Since I could not locate the exact problem and had ruled out any bad major parts, I did the only thing I knew to do.  This amp is the one I use the most so it was probably in worse condition than the amps I tinker with.

It sounds better than I ever remember it sounding.

[Ed_Chambley]

...
I have been reading the geofex debugging page.  That is where I got the information on checking the choke and PT.  Why is it not necessary to unhook a PT to check to see if it is open and inversly why is it necessary to unhook the choke? ...

The way the choke is positioned in the circuit makes it easy to get misleading readings unless you unhook at least one of the choke's wires.

More sources of info are better, but they can cause you to go in circles. The problem is without a specific repair case to describe, most of these sources are just a list of "things to check." The trick is you can spend 5 days checking/measuring everything in the amp, and not actually look in the area where the trouble lies.

Sluckey and I are trying to lead you through a logical troubleshooting procedure, with specific questions (or required readings), and specific next-steps based on your answers. We might save you some time and money replacing parts that don't need replacing.

Yes, I know you guys were going through a troubleshooting procedure and don't think for one minute I do not appreciate it.  You guys are great and I have found a home as I have the tube amp fever.  Isolating this problem to my understanding was a short somewhere.  I just wish I was better at locating these problems.  I read a lot of information and I understand you comment of running in circles, that is why I report back to you guys my findings.  Like Stuckey told me the other day, we have already ruled out the rectifier, so I did not waste time checking the socket.

[sluckey]

The problem is solved and I am not sure what the problem was.

I hate when that happens.

[Ed_Chambley]

The problem is solved and I am not sure what the problem was.

I hate when that happens.

Me too, but I love a nice sounding amp.  Thanks again!

[HotBluePlates]

No worries... It's fixed and that's the important thing!

I think sluckey and I feel like there was a missed learning opportunity.

[Ed_Chambley]

No worries... It's fixed and that's the important thing!

I think sluckey and I feel like there was a missed learning opportunity.

Even though I did not find the exact problem, I still learned a lot.  For instance I did not know about taking an ohms reading from the standby to chassis ground and that being able to point to certain components.  Also, the procedural methods of troubleshooting like disconnecting one end of filter caps.  I do believe the problem was simply sloppy workmanship over the years finally adding up to complete failure.  I would be wiling to bet this is what caused my original PT to fail 3 moths ago.  I kept it and pulled it out to test it again as I was wondering if I replaced a good part.  It is toast.  I did learn a lot, thanks!   I would have liked to pinpoint the exact problem myself, but I think it was a multiple of small problems adding up to a big one.

[HotBluePlates]

... but I think it was a multiple of small problems adding up to a big one.

That is generally the rule when finding problems in old amps. It can also be the problem in a new build, but the specific causes are often different.