Help Ampeg B-15NC B+ voltages low when pluging 6L6 Power Tubes

[agedhorse]

Ok, then, although I know I want to change the filter caps can, I would like to know how I can test it to find its fault (s). Cause, I'm really suspecting it now. I made the mistake of ruling it out as the problem...

Remove from the circuit, fully discharge and test with an LCR meter or bridge.

 

[Dom65]

I would really want to know what’s up with that can cap before doing anything else.

Troubleshooting rule #1 is to investigate and correct the damage from previous repair attempts and modifications.

The owner bought that amp used in 1998. He used used it a lot for a long time and very little in the last couple of years.
Apart from the normal jack and pots noise he said the it didn't have punch that it was kind of ducking when he pushed it.
I'd like to change the can cap anyways. The one that is on now is a 40/40/40/40MFD 500V and the schematics says 40/40/20/20MFD 500V.
Which one should I order?

 

[beans-on-toast]

The owner bought that amp used in 1998. He used used it a lot for a long time and very little in the last couple of years.
Apart from the normal jack and pots noise he said the it didn't have punch that it was kind of ducking when he pushed it.
I'd like to change the can cap anyways. The one that is on now is a 40/40/40/40MFD 500V and the schematics says 40/40/20/20MFD 500V.
Which one should I order?

The NC mostly used the 40-40-20-20. The 40-40-40-40 @525 will provide stiffer preamp stages, less sag and distortion. Either can be found in that revision. Order the insulator at the same time. If you want a cap can, I would go with CE Mfg (tubesandmore.com), it fits under the tube cage and is based on the original Mallory.

 

[agedhorse]

You don’t know what else might have been done to the amp that you haven’t seen yet.

What does the voltage do AT THAT NODE unloaded and loaded?

 

[Dom65]

The NC mostly used the 40-40-20-20. The 40-40-40-40 @525 will provide stiffer preamp stages, less sag and distortion. Either can be found in that revision. Order the insulator at the same time. If you want a cap can, I would go with CE Mfg (tubesandmore.com), it fits under the tube cage and is based on the original Mallory.

I see.
Since the owner has been playing it with the 40/40/40/40 all those years, I'm gonna stick to it then.
Thank you!

 

[Dom65]

You don’t know what else might have been done to the amp that you haven’t seen yet.

What does the voltage do AT THAT NODE unloaded and loaded?

The amp was used a lot for many years and was functional at least.
But, it doesn't mean that things were done properly... Obviously not, actually.
My understanding is that there was a 2 to 3 prong AC cable conversion, capacitor can replacement and 4 pins to 1/4 inch speaker jacks conversion (all done before the owner bought it in 1998.
Everything else was looking original.
Here is the picture of the inside when I first opened it.

The voltage at the first node of filter cap can Loaded (all tubes in) is 359VDC and Unloaded 535VDC
Taking the voltages this morning, the behavior of the fluctuation, when all the tubes are in, seems strange to me. After the normal peak of beginning it drops over a 30 second period to start climbing up slowly over a period of 5 min (say it goes down to 320V and climb up to 359V) At that point I turned off the amp as I'm afraid of stressing PT.

 

[agedhorse]

Those changes in voltage may very well be normal, the tube’s transconductance changes with temperature and this is reflected in the voltage changes as the current changes.

If you were to look at the B+ node, you might see excess ripple voltage as the load increases compared with a correct capacitor (assuming the existing cap is failing).

Looks like there were repairs and changes made around the switches, the input jack and a couple of suspect resistors that may have overheated or were changed.

Using a scope and comparing the actual waveform to what is expected from the schematic can usually help identify where things are different. IME, many problems I see when an amp is modified is due to the modifications or mistakes made during the process.

 

[Dom65]

Those changes in voltage may very well be normal, the tube’s transconductance changes with temperature and this is reflected in the voltage changes as the current changes.

If you were to look at the B+ node, you might see excess ripple voltage as the load increases compared with a correct capacitor (assuming the existing cap is failing).

Looks like there were repairs and changes made around the switches, the input jack and a couple of suspect resistors that may have overheated or were changed.

Using a scope and comparing the actual waveform to what is expected from the schematic can usually help identify where things are different. IME, many problems I see when an amp is modified is due to the modifications or mistakes made during the process.

Unfortunately, I don't have a scope.
Since the amp was used extensively for more than 20 years (touring and all) without any problem, I'm assuming that if I can finally get my voltages back to normal, everything will be fine. The owner hasn't played it much in the later years and now he wants to sell it. He brought it to me for the noisy pots and jacks plus the lack of punch and volume. He just wants it functional. He said it will be for the next owner to completely restore it if that's what the person wants.
My plan is to fix the improper 2 to 3 prong conversion. Hot to fuse, then fuse to power switch, then other side of power switch to the primary of PT.
Then finally common (white) directly to the other side of primary. Right now the hot side is going to the black and white primary PT wire and the common (white from the wall) is going to the black of the primary PT. Does it have to be inverted or it doesn't matter? Should I implement a power line high frequency noise filter like here? If so, what modern capacitor do you recommend to replace the vintage one that is still in the amp?
Also, I will replace the filter can with a new one. Speaking of, both the filter can positive (one of the outer legs underneath) and the output of standby switch are going to sleeve (ground) of the main speaker output jack. Is that normal? I see on the schematics that standby switch goes to what was the pin 2 of the old 4 pin speaker connector, but is that the way to properly ground the filter capacitor can?

 

[agedhorse]

No, do not use a new death cap, it's not a filter but a device to intentionally leak one side or the other to the chassis in ungrounded systems. Terrible idea back then, no better now.

If you/he just want to make it functional (an aspiration to mediocrity IMO), then wire the primary like all new amps are wired and grounded. The filter can common (ground) gets grounded to the chassis through a non-isolated speaker jack. There shouldn't be any high voltage positive leads grounded to the chassis... perhaps you found one of the root causes?

This is why ANY previously repaired units that come through our shop for a thorough evaluation before any work is done, you can NEVER assume that previous work was done correctly (or safely).

 

[Dom65]

No, do not use a new death cap, it's not a filter but a device to intentionally leak one side or the other to the chassis in ungrounded systems. Terrible idea back then, no better now.

If you/he just want to make it functional (an aspiration to mediocrity IMO), then wire the primary like all new amps are wired and grounded. The filter can common (ground) gets grounded to the chassis through a non-isolated speaker jack. There shouldn't be any high voltage positive leads grounded to the chassis... perhaps you found one of the root causes?

This is why ANY previously repaired units that come through our shop for a thorough evaluation before any work is done, you can NEVER assume that previous work was done correctly (or safely).

Ok perfect no death cap.
It's the owner that wants to get it just properly working to sell it at an affordable price.
On my side, I would like it, at least, in the kind of shape he played it for so many years.
Nobody touched it after he bought it in 1998. All I did so far is disconnected the death cap and lift and solder again a few components to read values.

 

[Wasnex]

Unfortunately, I don't have a scope.
Since the amp was used extensively for more than 20 years (touring and all) without any problem, I'm assuming that if I can finally get my voltages back to normal, everything will be fine. The owner hasn't played it much in the later years and now he wants to sell it. He brought it to me for the noisy pots and jacks plus the lack of punch and volume. He just wants it functional. He said it will be for the next owner to completely restore it if that's what the person wants.
My plan is to fix the improper 2 to 3 prong conversion. Hot to fuse, then fuse to power switch, then other side of power switch to the primary of PT.
Then finally common (white) directly to the other side of primary. Right now the hot side is going to the black and white primary PT wire and the common (white from the wall) is going to the black of the primary PT. Does it have to be inverted or it doesn't matter? Should I implement a power line high frequency noise filter like here? If so, what modern capacitor do you recommend to replace the vintage one that is still in the amp?
Also, I will replace the filter can with a new one. Speaking of, both the filter can positive (one of the outer legs underneath) and the output of standby switch are going to sleeve (ground) of the main speaker output jack. Is that normal? I see on the schematics that standby switch goes to what was the pin 2 of the old 4 pin speaker connector, but is that the way to properly ground the filter capacitor can?

AFAIK: The ground wire from the power chord should be connected to the chassis. There should also be at least one point where the power supply/signal ground is tied to the chassis. Sometimes there is more than one power supply/signal ground. My guess is the B15NC main signal ground is at the input jack.

The wire from the capacitor can to the speaker jack is not necessarily a chassis ground point. The speaker jacks may be insulated from the chassis. So the wire may actually be to deliver ground to the speaker jack.

If there is no power supply/signal ground, the readings are potentially suspect because the amp's circuits are are floating rather than ground referenced.

 

[Dom65]

AFAIK: The ground wire from the power chord should be connected to the chassis. There should also be at least one point where the power supply/signal ground is tied to the chassis. Sometimes there is more than one power supply/signal ground. My guess is the B15NC main signal ground is at the input jack.

The wire from the capacitor can to the speaker jack is not necessarily a chassis ground point. The speaker jacks may be insulated from the chassis. So the wire may actually be to deliver ground to the speaker jack.

If there is no power supply/signal ground, the readings are potentially suspect because the amp's circuits are are floating rather than ground referenced.

Sorry, I forgot to mention that the green wire goes to chassis of course ( like it is now).
"The power supply/signal..." Do you mean the filter can? The wire that goes from the can to the sleeve of speaker jack does conduct to the chassis. From that wire to all the nodes on board to the speaker jack, it all conducts to chassis.
Also, all the tubes, apart from the 6L6 of power section are reading fine on plates voltages when all tubes are connected. It's only the 6L6s that are reading low.
As for the signal part from the input, I already tested for conductivity many spots (like the grid leaks and some cathode resistance) and they are all going to chassis.

 

[beans-on-toast]

The ground bus runs from the first power supply node ground point down to the input jacks.

Connect the black speaker return to the phase inverter ground. I fly it over the circuit board. This is close to where the OPT black wire, center tap primary is connected. This makes the amp even quieter.

The only ground bus connection is at the input jacks. Ensure that this contact is clean. The lock washers often oxidize. Change them if they are oxidized. Be careful when moving the gray shield wires. They can be brittle and break when moved.

The speaker cable is isolated from the chassis. The EXT SPEAKER jack needs to be isolated from the chassis to prevent a ground loop.

The cap can should not be grounded to the chassis, if it is, that will also create a ground loop.

I would replace the cap can, the power tube cathode bias resistor and capacitor. Mount the resistor a mm or two above the board, it can run hot and scorch the board.

Remove the death cap. all connections to the ground switch. A terminal on the ground switch can be used to join the white power cord wire and the non-fused power transformer primary.

The green power cord safety ground is often connected to a power transformer mounting screw with a ring terminal and locking nut. It should have it’s own chassis hole but this would require drilling a hole and using special mounting hardware intended for the safety ground. The green wire should be longer than the black and white wires. The green wire should break last if the power cable is pulled out of the chassis.

Add the tube rectifier diode mod that I referenced in an earlier post.

For reference, this is an original B-15NC eyelet board.

 

[agedhorse]

Guys, the amp has already been modified with a 1/4" non-isolated main speaker jack (from the photo, with the "Japan" giveaway on it), it MUST have a ground from some point, the bottom of the OT secondary IS grounded. How it's done needs to be investigated as this may have been part of the modification "CF", because the original 4 wire cable used one wire for speaker ground since there's no other connection to ground. The EXTERNAL speaker jack MUST be isolated in order to achieve a series connection as it was shipped originally.

The can's shell must be grounded, since a isolation wafer was used originally than there will be a wire bonding it to the designer's desired ground bonding point (generally for noise control). None of the 4 section's + terminals should connect with ground.

Generally, the safest and quietest bonding point for both the cap can and (main) speaker output (both high current nodes) will be the power supply ground which also happens to be the can ground. I suspect that there were variations throughout the production though. The input jacks are bonded to the chassis for noise purposes as the original chassis was not grounded or bonded to the power source protective ground and this proved quiet(er) when the chassis was not grounded. This is a different case than when the amp was built.

 

[beans-on-toast]

If a switchcraft jack is used in place of the main 4-wire speaker cable, it can be isolated from the chassis with a Seitchcraft S-1028 and S-1029. Speaker [+] goes to the 8 ohm opt tap, speaker return [-] best goes to the phase inverter ground point for lowest noise. The opt secondary ground (black wire) should go to the ground bus at the phase inverter ground. %his is how it was done originally.

If there is no attached speaker cable, the safety loop circuit needs to modified. Pin-2 to pin-3 should be shorted as indicated. Since a jack is in place, I assume that this has been done.

The 4-pin speaker cable can be restored to original.

 

[beans-on-toast]

Connect the black speaker return to the phase inverter ground. I fly it over the circuit board. This is close to where the OPT black wire, center tap primary is connected. This makes the amp even quieter.

CORRECTION: In my post above, it should read ..OPT black wire secondary ground is connected close to the phase inverter circuit ground. All vintage B-15’s are wired this way.

 

[agedhorse]

Man are you guys making it harder than it is.

 

[byacey]

How do I measure AC ripple?

Measure AC voltage across the cap when loaded, while the amp is at idle.

 

[byacey]

Also, all the tubes, apart from the 6L6 of power section are reading fine on plates voltages when all tubes are connected. It's only the 6L6s that are reading low.

This further points to the first capacitor section being bad.l

 

[Dom65]

The ground bus runs from the first power supply node ground point down to the input jacks.

First power supply node being the capacitor tab connecting to ground bus (blue wire in 1rst picture)?

Connect the black speaker return to the phase inverter ground. I fly it over the circuit board. This is close to where the OPT black wire, center tap primary is connected. This makes the amp even quieter.

That spot on 5th photo?

The only ground bus connection is at the input jacks. Ensure that this contact is clean. The lock washers often oxidize. Change them if they are oxidized. Be careful when moving the gray shield wires. They can be brittle and break when moved.

Here, I can see that the washers were previously replaced and the contact to the chassis is very good. Actually, there was a noise problem due to an oxidized shunt. I fixed it by deoxidizing it, but knowing, now, the importance of it, I will change it.

The speaker cable is isolated from the chassis. The EXT SPEAKER jack needs to be isolated from the chassis to prevent a ground loop.

I put a photo of the output jacks. Right now, if I plug to the main speaker out jack (same with nothing plugged) both tips (main and EXT) are conducting to ground bus. If I'm plugged in EXT Speaker, both tips stop conducting to ground bus.

The cap can should not be grounded to the chassis, if it is, that will also create a ground loop.

See 1rst picture. 1 tab is connected to the ground bus.

I would replace the cap can, the power tube cathode bias resistor and capacitor. Mount the resistor a mm or two above the board, it can run hot and scorch the board.

Sure, I'm having a hard time finding the capacitor can here in Canada. I have to see if the owner is willing to pay the extra charges for ordering from the US.

Remove the death cap.
all connections to the ground switch.

Already did.

A terminal on the ground switch can be used to join the white power cord wire and the non-fused power transformer primary.

I connected the white power cord wire directly to one of the primary. I will revise that section since the wall ac hot (black) doesn't pass by the fuse first.

The green power cord safety ground is often connected to a power transformer mounting screw with a ring terminal and locking nut. It should have it’s own chassis hole but this would require drilling a hole and using special mounting hardware intended for the safety ground. The green wire should be longer than the black and white wires. The green wire should break last if the power cable is pulled out of the chassis.

Right now, it is under a PT mounting screw but it doesn't have an eyelet. So I will revise that.

Add the tube rectifier diode mod that I referenced in an earlier post.

OK, would that help reducing the loud pop it has when using standby switch?

For reference, this is an original B-15NC eyelet board.

Thanks for your time.

View attachment 7091218


View attachment 7091213