1967 Gibson GA5-T Amp Issues

[martin manning]

V3 pin 5 is only being used as a tie point, and since it's tied to V3 pin 2, it's a ground. If you can remove the old 10uF and the 820 ohm from the 4k7 carbon resistor you can leave it there. When the cap and the 820 ohm are removed, then you can measure the 4k7 to see if it is within tolerance (10%).

Don't forget that the negative end of the new 100uF cathode bypass cap goes to ground, V3 Pin 2.

There must be a broken wire or bad joint for the voltage on V1 pins 4-5 to be that low.

[texstrat]

There must be a broken wire or bad joint for the voltage on V1 pins 4-5 to be that low.

Probing around I do not see anything that out of the ordinary, but what do I know. One thing that is not shown on the layout V1 pins 4 and 5 are connected and the wire runs into V2 pin 5. I had it on the original hand drawn layout, but missed it when I split V1. The problem with V1 may have started when I was attempting to get a reading on V1 pin 3 and inadvertently touch pin 4 while on pin 3. Which resulted in a pop. I stated this in the beginning of the thread.

Also, the black lead from the 20/10/10 capacitor is loose at V3 pin 2. What effect will this have with the problems I am experiencing?

How should I proceed?

[martin manning]

I don't think the slip with the probe on V1 would have any serious consequences. You momentarily connected the filament to the cathode, so all low voltages involved.

RE the loose black lead from the multi-cap, if that were electrically disconnected you would have a huge hum problem. It may be contributing to the high ripple voltage you measured. If you are replacing the three main filters, then that will be taken care of. If you want you could re-solder it and see if there is any change to the power supply voltages and that large AC voltage on V3 pin 7.

The 12AX7's each contain two triodes, and each triode has its own 6-volt filament inside a tubular cathode. Internally, one end of one filament is connected to pin 4, one end of the other to pin 5, and the other ends of both to pin 9. This is so they can be wired for 6 or 12-volt operation. In your case, they are wired in parallel for 6-volt operation by connecting pins 4 and 5 together, as shown in the Gibson schematic. you should measure ~6.5 volts at each tube's filament connections. Try to measure on the socket lugs (both pin 4 and pin 5 on the 12AX7's as well as pin 9) rather than on the wire to discover any bad solder joints. You can take the tubes out and do this from the other side of the chassis to make it easier. You could also re-flow all of those connections for good measure.

Now that you have drawn in the filament wiring I can see that someone has created an artificial center tap for the filament winding, which is the two resistors marked 50K going from V2's pins 4 and 5 to the ground lug of the adjacent terminal strip. This is a better plan than grounding one side of the filament winding as shown on the schematic, and should result in reduced 60Hz hum. Those resistors should be 100-150 ohms though. If they are really 50k, I'd replace them with 100-ohm parts.

[Phil_S]

I'm only chiming in here because I think texstrat may not be clear about how to measure the filament voltage. With the tube side up, power on, tubes pulled for the 12AX7 sockets, place one probe on pin 9. Place the other probe on pin 4. Read the meter for VAC (voltage across 9 and 4). Then, leave one probe on pin 9, and place the other on pin 5 (across 9 and 5). You should get ~6.3VAC either way because pins 4 and 5 are jumpered. We want to check for a) whether there is a problem with filament voltage, and 2) whether there is a problem with the jumper or the tube pin. This is not done from the underside because what you do there may show good voltage that is not reaching the tube itself.

[martin manning]

That makes it crystal-clear; thanks!

[texstrat]

Now that you have drawn in the filament wiring I can see that someone has created an artificial center tap for the filament winding, which is the two resistors marked 50K going from V2's pins 4 and 5 to the ground lug of the adjacent terminal strip. This is a better plan than grounding one side of the filament winding as shown on the schematic, and should result in reduced 60Hz hum. Those resistors should be 100-150 ohms though. If they are really 50k, I'd replace them with 100-ohm parts.

Well I am definitely learning this the hard way. I noted all the resistor colors for the metal and carbon pieces and Martin you are correct again. The two resistors from V2's pin 4 and 5 are actually 100 ohm, not 50K.

Note to self, look at the bands for values and not try to get them by measuring in the circuit.

I have ordered three new Sprague Atom capacitors for the power supply tubesandmore.com and should have them later this week.

I'm only chiming in here because I think texstrat may not be clear about how to measure the filament voltage. With the tube side up, power on, tubes pulled for the 12AX7 sockets, place one probe on pin 9. Place the other probe on pin 4. Read the meter for VAC (voltage across 9 and 4). Then, leave one probe on pin 9, and place the other on pin 5 (across 9 and 5). You should get ~6.3VAC either way because pins 4 and 5 are jumpered. We want to check for a) whether there is a problem with filament voltage, and 2) whether there is a problem with the jumper or the tube pin. This is not done from the underside because what you do there may show good voltage that is not reaching the tube itself.

Do I need to pull both tubes from V1 and V4 and measure, tube side, pins 4, 5 and 9? Earlier I just tested V1. Should I test tube side and then underside on the lugs, clear of solder and wires, to see if there is a difference in voltage? The probe tips are pointed but I am not sure if they are making good contact with the metal that holds the pins secure. Let me know if I need to recheck the heater pins for both V1 and V4 and if I should check the underside as well.

Thanks.

[martin manning]

I'd just measure them all from the top side. Before you do that, though, looking at your photo of V1 (the one with the black electrolytic disconnected) the solder joint on pin 9 looks nasty. Re-flow that one (adding a little new solder), and then see what you get. Be aware that you have to measure from pins 4-5 to pin 9 to get the ~6.5VAC you are looking for (one probe on 4-5, the other on 9). If you measure from either 9 or 4-5 to ground, you will only get 3.3VAC or so because of the artificial center tap.

[texstrat]

Alright, got the capacitor and resistor swapped out on V1. Added the 100uf capacitor between V2 and V3. Removed the 820 ohm resistor and put a new 10uf capacitor between V3 and V4.

Now to pull the 12ax7's and check voltages between pins 4, 5 and 9. I leave one probe in pin 9 and use the other to test pins 4 and 5.

[texstrat]

Martin/Phil:

The probe tips on my DMM are too large to fit into the socket pin holes. Any ideas how I can modify the probe to get contact with the pin tensioners?

[texstrat]

Okay, it looks like I am making some head way, but losing some ground, I get no sound when I plug the guitar in.

Voltage reading from Monday:
V1
Pin 1 - 117VDC
Pin 2 - 0
Pin 3 - 1.12VDC
Pin 6 - 78VDC
Pin 7 - 49mVDC
Pin 8 - 0.6VDC

V2
Pin 2 - 10mV
Pin 3 - 8.0VDC
Pin 7 - 265VDC
Pin 9 - 262VDC

V3
Pin 1 - 262VAC
Pin 6 - 262VAC
Pin 7 - 285VDC or 86VAC

V4
Pin 1 - 50VDC
Pin 2 - 8mVAC
Pin 3 - 1.1VDC

Voltage readings post changes mentioned above:
V1
Pin 1 - 196VDC
Pin 2 - 0
Pin 3 - 5.0VDC
Pin 6 - 135VDC
Pin 7 - 0
Pin 8 - 1.26VDC

V2
Pin 2 - 10mV
Pin 3 - 8.2VDC
Pin 7 - 264VDC
Pin 9 - 264VDC

V3
Pin 1 - 260VAC
Pin 6 - 260VAC
Pin 7 - 285VDC or 117VAC

V4
Pin 1 - 106VDC
Pin 2 - 9mV
Pin 7 - 4.96VDC

I had to test the heater pins on V1 and V4 from underneath the board as my probe tips were too large to fit into the pin holes. I placed one probe on pin 9 and moved the probe between pins 4 and 5. I got 6.7VAC for both V1 and V4.

Checking voltages off V3 pin 7 284VDC, looking at the schematic voltage chart I noticed pin 1 and 3 voltages are AC and pin 7 looks to be DC like the other tube values. Voltage for V3 pin 9, after the 3.3K resistor, 264V. Voltage reading after the 100K resistor, which is on the terminal strip where V1 pin 6 and green lead from the ps multi-cap meet is 198V.

Voltages for V1, V2 and V4 pin 3 are higher than the listed voltage of 2.3V. What may have caused this?

Also, why am I getting no sound from the amp? I check the leads to the speaker and nothing is disconnected. The wire have been spliced at one point and I need to have a closer look at them, but I don't think they are suspect.

Anyway that is where things are right now.

[martin manning]

Ok, looks like the heater voltages are in line now. Are all of the filaments glowing?

The B+ voltages are also looking better (for now), but you still need to replace the filter caps.

The V1A triode (pins 6, 7, eight) looks fine, but the V1b triode (pins 1, 2, 3) is not conducting. It's biased into cut-off, and this is probably why you have no sound now. V4 is in the same condition, since it's cathode is tied to V1b's.

V2 pins 7 and 9 should not be the same. You say you have 284 and 264 on the B+ nodes 1 and 2, so pin 7 should be close to 284 and 9 close to 264. Are you sure you have that right?

I am really wondering about the value of the 4k7 common cathode resistor on V1b and V4. That's what is shown on the schematic, and that's what the carbon comp resistor in the amp is, but it seemed odd to me from the beginning. A value of ~0.8k would make more sense. My first thought would be that the value on the schematic is an error, but the existing carbon resistor argues against that. You could tack the 820 ohm back across the 4k7 to see if it starts working again, but get the new filters in before you sort this out for good.

A comment about your lead dress, as it's called: Keep component leads short. With point-to-point wiring such as you have, the shortest distance is as straight a line as you can manage. When you parallel one component with another, keep the two close together.

[texstrat]

Ok, looks like the heater voltages are in line now. Are all of the filaments glowing?

Yes, V1 is now heating both at the top and bottom of the tube.

The B+ voltages are also looking better (for now), but you still need to replace the filter caps.

Those are on there way.

The V1A triode (pins 6, 7, eight) looks fine, but the V1b triode (pins 1, 2, 3) is not conducting. It's biased into cut-off, and this is probably why you have no sound now. V4 is in the same condition, since it's cathode is tied to V1b's.

V2 pins 7 and 9 should not be the same. You say you have 284 and 264 on the B+ nodes 1 and 2, so pin 7 should be close to 284 and 9 close to 264. Are you sure you have that right?

V2 pins 7 and 9 read 264V and 264V respectively. V3 pin 7 read 284VDC and 117VAC. One thing I would like to understand better is what each pin represents, cathode, B+, heater, etc.

I am really wondering about the value of the 4k7 common cathode resistor on V1b and V4. That's what is shown on the schematic, and that's what the carbon comp resistor in the amp is, but it seemed odd to me from the beginning. A value of ~0.8k would make more sense. My first thought would be that the value on the schematic is an error, but the existing carbon resistor argues against that. You could tack the 820 ohm back across the 4k7 to see if it starts working again, but get the new filters in before you sort this out for good.

I measured the resistor with one lead free and it is at 5.17 ohms, right at the high side of the +10% for it value. Is it on its way out, should I go ahead and replace it? I will add the 0.8K back on and see if I get sound again.

A comment about your lead dress, as it's called: Keep component leads short. With point-to-point wiring such as you have, the shortest distance is as straight a line as you can manage. When you parallel one component with another, keep the two close together.

I will definitely keep that in mind.

[texstrat]

Put the 0.8K resistor on the 10uf capacitor and I got sound. Remeasured voltages and here is what I got. In parentheses are the voltage values from last night after replacing components.

V1
Pin 1 - 127VDC (196VDC)
Pin 2 - 0 (0)
Pin 3 - 1.15VDC (5.0VDC)
Pin 6 - 112VDC (135VDC)
Pin 7 - 0.3mV (0)
Pin 8 - 1.05VDC (1.26VDC)

V2
Pin 2 - 9.8mV (10mV)
Pin 3 - 8.03VDC (8.2VDC)
Pin 7 - 264VDC (264VDC)
Pin 9 - 258VDC (264VDC)

V3
Pin 1 - 260VAC (260VAC)
Pin 6 - 260VAC (260VAC)
Pin 7 - 284VDC or 129VAC (285VDC or 117VAC)

V4
Pin 1 - 51VDC (106VDC)
Pin 2 - 8mV (9mV)
Pin 7 - 1.17VDC (4.96VDC)

Checking voltages off V3 pin 7 284VDC/129VAC. Voltage for V3 pin 9, after the 3.3K resistor, 258V. Voltage reading after the 100K resistor, which is on the terminal strip where V1 pin 6 and green lead from the ps multi-cap meet is 165V.

[martin manning]

You are probably at a point now where you should wait until you get the new filters in before trying to go further. You can get the tremolo working and find the best value for the V1b/V4 cathode resistor then.

Re understanding what each tube pin represents, the schematic makes it clear which pin connects to which element; or, do you mean what is the function of the various elements in a vacuum tube? For that you can find lots of info on the web... Google is your friend.

[Phil_S]

Just a note of encouragement here. It is passing sound, voltages are shaping up. This is great news. Congratulations to texstrat and Martin for a job well done!

You want to know more about the tube structures? I suggest you look for a copy of RC-30 in pdf to download. I am posting a link because it helps to know where to look and this is a gold mine. The first few pages of RC-30 should answer your question.
http://www.tubebooks.org/tube_data.htm