1967 Gibson GA5-T Amp Issues

[texstrat]

I do hear a hum when I fire it up, but not sure what Hz it is.

I agree, get it up and running like it is supposed to and then modify later.

Will advise.

[texstrat]

A separate note, I was looking at the new Mesa Boogie 12ax7 tube that I got to test in V1 and I noticed along the bottom of the tube that the glass is rough, as if though it had rubbed against something abrasive. I did not notice it before I installed it and it from the box to the socket. Will this cause the tube to fail prematurely?

[Phil_S]

FWIW, I fully agree with Martin's advice to get the amp working in stock form. This particular amp, as I understand it, has killer tone. I wouldn't be so quick to assume it will sound thin. I also agree with Martin that you can try to fiddle with the tone caps later. Right now, you've got basic operating issues. Let's not introduce unnecessary variables. Reasonably strict scientific method will aid in drawing correct conclusions.

I'd note that the Gibson tone stack (C1, C2, R1, R2) is a weird configuration that acts as a notch filter that modern day guitar players tend not to like and that is something we can focus on after other problems are solved. It is not today's agenda.

I also support Martin's recommendation to R&R the three filter caps 20/10/10 at original values or as close as you can reasonably get. There are many ways to skin this cat. That 10/10 Martin is showing you is very nice and will be easy to work with. If you want to spend less money, you can look at the radial or axial single caps which should run around $5 for all three. Get caps with a voltage rating of 300V or more. 350V and 450V are common ratings. Don't go below 350V, as you want a margin of protection. You have plenty of room in the chassis. You can hot glue (sparingly) the three caps together (or use a zip tie) and to the chassis to stabilize them. If the leads are too short, you can add a bit of wire to make them work. You may want to sleeve the + wire -- just strip a bit of insulation from 300V or higher rated wire or get some heat shrink tubing. I'd remove that large 47u cap, too, and I'd replace all the other electrolytic caps with similar values. Cathode bypass caps rated 25V are more than adequate for the other caps.

Now, speaking with a forked tongue, the one mod I'd probably do without hesitation is the change the cathode bypass cap on V1 pin 8 from 1u to something between 10u and 25u. This will allow more bass to pass and, as long as you are in there fixing it, I'd just do this now. This is one place where Martin and I disagree and I suggest this based on my experience with another Gibson V1A setup for an amp I own. If you have any hesitation, order a replacement for the 1u cap, too.

I think I might also look to split the shared 10u bypass cap on the 2nd half of V1 pin 3 and V2 pin 3. Give each cathode its own 10u cap. These are tiny, modern ones are maybe .5" long x .25" dia.

I mention this so that you can order the parts, otherwise you will spend $10 for shipping $5 worth of parts later. Try to get everything you might need in one order, as that will be cheaper in the long run.

You asked about hum. The A note below the low E string is 55Hz. The open A string is 110Hz. So, A# is approximately 60/120. 120Hz is ripple in the power supply. 60Hz is usually bleed over from the filament supply due to poor lead dress. (Lead dress: the art of cutting and placing wires so interference is minimized.) BTW, did you do anything about that heater problem on V1?

The tremolo tube (V4) has voltage that looks wrong to me. I'm not saying you measured wrong, I'm saying it shouldn't be 50V. I don't believe the tremolo will oscillate with 50V. I didn't see your reading for B+2 (250V on the schematic), so let's assume it is something in that range. This means 200v is dropped across the 150,000 ohm resistor. Use Ohm's law (V=I*R) to solve for current = 1.3mA. I think this tube should be drawing more like 2-3mA, which would make for a voltage drop of more like 50V. (Note, similar situation on both halves of V1, where you use 50K as R value because there are two 100K in parallel.)

This leads me to wonder, what is going on in your B+ supply ladder (the R/C network with the 20/10/10 paper cap). I return to thinking the first order of business is to change the filter caps there and get your power supply operating as it should.

You asked about rough glass at the base of the Mesa 12AX7. When the tube loses vacuum, the inside turns white. You'll know it when you see this. If the tube has vacuum, the rough glass is probably only a cosmetic concern, though it should be smooth and polished. If you recently purchased this tube you might try to exchange it.

I have some trouble staying on the side lines. I want to stay out of Martin's way and it can be confusing to get coaching from more than one person. I hope the comments are helpful.

[texstrat]

FWIW, I fully agree with Martin's advice to get the amp working in stock form. This particular amp, as I understand it, has killer tone. I wouldn't be so quick to assume it will sound thin. I also agree with Martin that you can try to fiddle with the tone caps later. Right now, you've got basic operating issues. Let's not introduce unnecessary variables. Reasonably strict scientific method will aid in drawing correct conclusions.

I'd note that the Gibson tone stack (C1, C2, R1, R2) is a weird configuration that acts as a notch filter that modern day guitar players tend not to like and that is something we can focus on after other problems are solved. It is not today's agenda.

I also support Martin's recommendation to R&R the three filter caps 20/10/10 at original values or as close as you can reasonably get. There are many ways to skin this cat. That 10/10 Martin is showing you is very nice and will be easy to work with. If you want to spend less money, you can look at the radial or axial single caps which should run around $5 for all three. Get caps with a voltage rating of 300V or more. 350V and 450V are common ratings. Don't go below 350V, as you want a margin of protection. You have plenty of room in the chassis. You can hot glue (sparingly) the three caps together (or use a zip tie) and to the chassis to stabilize them. If the leads are too short, you can add a bit of wire to make them work. You may want to sleeve the + wire -- just strip a bit of insulation from 300V or higher rated wire or get some heat shrink tubing. I'd remove that large 47u cap, too, and I'd replace all the other electrolytic caps with similar values. Cathode bypass caps rated 25V are more than adequate for the other caps.

Vibroworld has a 20/10/10 multi-cap that I am ordering. Martin suggested cutting the red lead from the multi-cap and check the voltage and everything remained the same with the exception of pin 7 V3 the VAC dropped to 78 from 86 last night.

Now, speaking with a forked tongue, the one mod I'd probably do without hesitation is the change the cathode bypass cap on V1 pin 8 from 1u to something between 10u and 25u. This will allow more bass to pass and, as long as you are in there fixing it, I'd just do this now. This is one place where Martin and I disagree and I suggest this based on my experience with another Gibson V1A setup for an amp I own. If you have any hesitation, order a replacement for the 1u cap, too.

I think vibroworld has the larger value caps. I have not found a supplier that carries everything to place one order. Vibroworld has the multi-cap, but not a good selection of caps.

I think I might also look to split the shared 10u bypass cap on the 2nd half of V1 pin 3 and V2 pin 3. Give each cathode its own 10u cap. These are tiny, modern ones are maybe .5" long x .25" dia.

Do you mean first half of V1 pin3 and V4 pin3?

You asked about hum. The A note below the low E string is 55Hz. The open A string is 110Hz. So, A# is approximately 60/120. 120Hz is ripple in the power supply. 60Hz is usually bleed over from the filament supply due to poor lead dress. (Lead dress: the art of cutting and placing wires so interference is minimized.) BTW, did you do anything about that heater problem on V1?

I am not sure but I can post some additional pictures of the wiring layout. I don't think we addressed the heater problem on V1, what do I need to do?

The tremolo tube (V4) has voltage that looks wrong to me. I'm not saying you measured wrong, I'm saying it shouldn't be 50V. I don't believe the tremolo will oscillate with 50V. I didn't see your reading for B+2 (250V on the schematic), so let's assume it is something in that range. This means 200v is dropped across the 150,000 ohm resistor. Use Ohm's law (V=I*R) to solve for current = 1.3mA. I think this tube should be drawing more like 2-3mA, which would make for a voltage drop of more like 50V. (Note, similar situation on both halves of V1, where you use 50K as R value because there are two 100K in parallel.)

This leads me to wonder, what is going on in your B+ supply ladder (the R/C network with the 20/10/10 paper cap). I return to thinking the first order of business is to change the filter caps there and get your power supply operating as it should.

To make sure that I understand, and based on the Gibson schematic, the voltages off V3 pin 7 should be the same at V2 pin 7, V4 pin 1 , and V1 pin 1 and pin 6? I measured the voltage for V1 pin 6 after the 100K resistor at the terminal strip and got 155V versus the 200V showing on the schematic. Does that tell us anything? Also, what is the R/C network for my edification?

You asked about rough glass at the base of the Mesa 12AX7. When the tube loses vacuum, the inside turns white. You'll know it when you see this. If the tube has vacuum, the rough glass is probably only a cosmetic concern, though it should be smooth and polished. If you recently purchased this tube you might try to exchange it.

I have some trouble staying on the side lines. I want to stay out of Martin's way and it can be confusing to get coaching from more than one person. I hope the comments are helpful.

I will take the tube back. I do not mind the additional insight and I certainly appreciate everything Martin has done so far to help me understand the internal workings of this amplifier. Any and all help is appreciated.

[Phil_S]

It's not necessary to get a multi section cap. If this were my amp, I'd just get three singles. It is cheaper and more flexible to do that. The 20u cap will be about 1"x.75" and the 10u cap will be about .75" x .625". Just figure a way to stabilize them when mounted. Of course, the choice is yours and there's no wrong answer.

If you can't find the 25V caps at Vibroworld, look to place your order at another supplier. Don't get hung up on getting a multi cap. This is simply bad economics at work because of the shipping costs and I'd avoid it.

Reliable suppliers that ship quickly and should have everything you need include Tube Depot and AES (tubesandmore.com). There are others, too.

I suggested splitting the cathodes attached to a 10u cap. On the schematic, look directly under the words "TREMOLO V4", bottom of the page, slightly to the right of center. That 10u cap is shared by 1/2 of V1 and the half of V4 that's in use. Anyway, this one is a low priority.

On the heater problem, it's my recollection that you have one socket that is only lighting half of the filament inside the tube. If you have a dark filament, the tube won't operate well. Check V1 to see if you are getting 6.3VAC (or close to it). In this case, pull the tube and put your probes in the tube socket from the tube side. We want to know if voltage is reaching the tube. Put one probe in pin 9 (light touch, you only need contact, don't ruin the socket tension) and the other in 4 or 5. Actually, check between 4 and 9, then between 5 and 9. You should get the same exact result.

The voltage is 155VDC where you expect 200VDC. This is too low. The problem needs to be found and corrected. We don't know if this is anything more than the filter cap being so old that it is not doing it's job. Let's see what this is after you change the filter caps.

[texstrat]

Ok, for now I will pick-up two 10uf 25V or better for the power supply. I did get a couple of 10uf 35V Nichicon caps from RS. Will these be ok, or should get a specific brand for the cathode biasing and a specific cap brand for the power supply. The reason I ask is I have read so much information about different brands and how each has a different affect on the tone, etc. the last couple of days, that I would prefer a recommendation from a experienced amp builder/restorer. Plus I believe you have a similar Gibson.

I see know what you are talking about the 10 uf cap being shared by V4 and first half of V1. My layout shows it going by way of V2 and V3. I have post a new layout as Martin noted some errors that I have since corrected. The highlighted areas are the changes I plan to make tomorrow after work a 4.7K resistor and 10uf cap in series between V3 pin5 and V4 pin 3. A 2.2K resistor in series with 10uf cap from V1 pin 8 to the middle terminal strip or ground.

Addressing the heater problem, pull the tube from V1, power up the amp, and then, with the DMM set to AC, insert one probe in pin 9 and with the other probe test pins 4 and 5? does it matter which probe goes where?

Thanks again.

[Phil_S]

Ok, for now I will pick-up two 10uf 25V or better for the power supply.

NO! The power supply caps need to be rated for at least the voltage they carry. You need 300V or higher rated caps. I think you will find 350, 450, and 500V are common ratings.

[martin manning]

I'll be glad to sit back and watch for a while now. Two points of view is never a bad thing!

Regarding this:

I suggested splitting the cathodes attached to a 10u cap. On the schematic, look directly under the words "TREMOLO V4", bottom of the page, slightly to the right of center. That 10u cap is shared by 1/2 of V1 and the half of V4 that's in use. Anyway, this one is a low priority.

These two have to share, as that is how the trem effect is injected into the signal path.

[Phil_S]

I did get a couple of 10uf 35V Nichicon caps from RS. Will these be ok, or should get a specific brand for the cathode biasing and a specific cap brand for the power supply. The reason I ask is I have read so much information about different brands and how each has a different affect on the tone, etc. the last couple of days, that I would prefer a recommendation from a experienced amp builder/restorer. Plus I believe you have a similar Gibson.

When this amp was manufactured, they sourced the caps from the least expensive supplier. Manufacturing standards were lower tolerance. I'll bet, if you look on that cap it says +20%/-50% or something like that. Today's caps, whether made in Asia or US, are of far higher quality and tolerance. I personally, have no problem at all with off brand caps. I think Sprague are overpriced. Xicon, Illinois, and Rubycon are all good. Just get them from a reliable vendor, which should protect you from bad quality counterfeit caps. I use whatever I can get for a reasonable price. Those Nichicon 10u/35v are great for anywhere but the power supply, where you need voltage ratings above what you expect to find there.

I see know what you are talking about the 10 uf cap being shared by V4 and first half of V1. My layout shows it going by way of V2 and V3. I have post a new layout as Martin noted some errors that I have since corrected. The highlighted areas are the changes I plan to make tomorrow after work a 4.7K resistor and 10uf cap in series between V3 pin5 and V4 pin 3. A 2.2K resistor in series with 10uf cap from V1 pin 8 to the middle terminal strip or ground.

There is a lot of sorting out to do here. The Gibson schematic calls for a 4.7K on the V4 cathode (pin 3), that goes to ground. Next to that is a 10u/25V cap in parallel -- it also goes from pin 3 to ground. There is a wire indicated from V1B cathode (pin 3) to V4 pin 3. By tying the cathodes together like this, they share the R/C 4.7K/10u. That is what is on the schematic. V1 and V4 are both 12AX7. I suggested giving each cathode its own R/C. That need not be done now. Please change the 10u cap and put it where it belongs.

V2 is the power tube, 6BQ5 or EL84 (same thing). It also has its cathode on pin 3. According to the schematic, there is a 150 ohm 1W resistor on pin 3 that goes to ground. This resistor establishes the bias for the power tube. Typically this is bypassed by a 10-25u cap (25V rating) and you will benefit tonally from having it there.

I see the pictures of V3 + V4. I can't make any sense of it. It looks wrong. Are you able to give a picture from directly above those two sockets including the cap. Please indicate which socket is to the left or right.

Addressing the heater problem, pull the tube from V1, power up the amp, and then, with the DMM set to AC, insert one probe in pin 9 and with the other probe test pins 4 and 5? does it matter which probe goes where?

AC doesn't have + or -. Probes go either way. Just be careful what you touch with the probes.

[texstrat]

Ok, for now I will pick-up two 10uf 25V or better for the power supply.

NO! The power supply caps need to be rated for at least the voltage they carry. You need 300V or higher rated caps. I think you will find 350, 450, and 500V are common ratings.

Sorry, slip of the fingers. I meant to say I would get the 10uf 25V+ for the cathode biasing. For the power supply, 10uf 350V+.

[texstrat]

here is a lot of sorting out to do here. The Gibson schematic calls for a 4.7K on the V4 cathode (pin 3), that goes to ground. Next to that is a 10u/25V cap in parallel -- it also goes from pin 3 to ground. There is a wire indicated from V1B cathode (pin 3) to V4 pin 3. By tying the cathodes together like this, they share the R/C 4.7K/10u. That is what is on the schematic. V1 and V4 are both 12AX7. I suggested giving each cathode its own R/C. That need not be done now. Please change the 10u cap and put it where it belongs.

So instead the way the r/c wired, 10uf plus two resistors (695ohms each) in parallel, between V3 pin 5 and V4 pin 3. I need to cut that out and solder in parallel a 4K7 resistor and 10uf cap from V4 pin 3 to V1 pin 3? Solder the resistor and cap to the wire on V1 pin 3.

A couple of things I need to mention that Martin caught yesterday.

1. Multistripe wire from TF-102 goes to V3 pin 2, not pin 3.
2. The two green leads off of TF-102, not originally shown, tie into V3 pins 3 and 4.
3. There is a jumper on V3 between pin 2 and pin 5 that ties into the current r/c going to V4 pin 3.

So I don't know if this changes any of the above recommendations from you or not.

V2 is the power tube, 6BQ5 or EL84 (same thing). It also has its cathode on pin 3. According to the schematic, there is a 150 ohm 1W resistor on pin 3 that goes to ground. This resistor establishes the bias for the power tube. Typically this is bypassed by a 10-25u cap (25V rating) and you will benefit tonally from having it there.

The 150 ohm resistor ties into V2 pin3 and V3 pin 2. Remove the resistor and in its place a 10-25uf 25v+ cap?

I see the pictures of V3 + V4. I can't make any sense of it. It looks wrong. Are you able to give a picture from directly above those two sockets including the cap. Please indicate which socket is to the left or right.

I will get some better pictures posted tomorrow morning along with the revised hand drawn layout that I failed to upload today.

[martin manning]

V2 is the power tube, 6BQ5 or EL84 (same thing). It also has its cathode on pin 3. According to the schematic, there is a 150 ohm 1W resistor on pin 3 that goes to ground. This resistor establishes the bias for the power tube. Typically this is bypassed by a 10-25u cap (25V rating) and you will benefit tonally from having it there.

The 150 ohm resistor ties into V2 pin3 and V3 pin 2. Remove the resistor and in its place a 10-25uf 25v+ cap?

No, "bypass" here means to wire the cap in parallel with the resistor. In other words, leave the 150-ohm resistor in place and solder the cap across it. This cap will allow AC (created by the guitar signal) to bypass the resistor so that the bias voltage (due to the DC current flowing through the resistor) is maintained. I agree this will help fatten up the tone.

The value of this bypass cap should be 100uF, though, so that the full effect of the improvement extends all the way down to the lowest audible frequencies.

[martin manning]

There is a lot of sorting out to do here. The Gibson schematic calls for a 4.7K on the V4 cathode (pin 3), that goes to ground. Next to that is a 10u/25V cap in parallel -- it also goes from pin 3 to ground. There is a wire indicated from V1B cathode (pin 3) to V4 pin 3. By tying the cathodes together like this, they share the R/C 4.7K/10u. That is what is on the schematic. V1 and V4 are both 12AX7. I suggested giving each cathode its own R/C. That need not be done now. Please change the 10u cap and put it where it belongs.

So instead the way the r/c wired, 10uf plus two resistors (695ohms each) in parallel, between V3 pin 5 and V4 pin 3. I need to cut that out and solder in parallel a 4K7 resistor and 10uf cap from V4 pin 3 to V1 pin 3? Solder the resistor and cap to the wire on V1 pin 3.

Yes do that, except you should de-solder the old components and remove the leads rather than just cutting them out. Get some de-soldering braid and/or a solder sucking thing from the local Radio Shack if you don't have those. BTW, I'd bet a '59 'burst that those existing resistors are not 695 ohms... you can't measure them in-circuit. Learning to read the color codes is very useful.

Per my comment above, you do not want to use a separate cathode resistor and bypass cap for V4 and V1b. The trem oscillator (V4) needs to be connected to V1b via this common network so that it can modulate V1b's cathode voltage and produce the tremolo effect.

[Phil_S]

I'll be glad to sit back and watch for a while now. Two points of view is never a bad thing!

Regarding this:

I suggested splitting the cathodes attached to a 10u cap. On the schematic, look directly under the words "TREMOLO V4", bottom of the page, slightly to the right of center. That 10u cap is shared by 1/2 of V1 and the half of V4 that's in use. Anyway, this one is a low priority.

These two have to share, as that is how the trem effect is injected into the signal path.

Opps! Sorry, I wasn't thnking. That's the only place where the tremolo is connected to the rest of the circuit. Of course, that has to be it. Interesting way to do it.

OK, I messed up. Your turn again, Martin.

[texstrat]

I got 0.6VAC when I probed pins 4 and 5 on V1. Not close to the 6.3V it should be. What could be causing the problem?

Martin, the resistor values between V3 and V4 are 4.7K carbon and 820 ohm metal. The Carbon resistor lead starts at V3 pin 2 across to pin 5 then over to V4 pin 3. The 10uf cap is soldered directly to V3 pin 5 and V4 pin 2. The 820 ohm resistor is in parallel with the cap.

I have included a revised layout with proposed changes highlighted. Changing the 47uf/808 ohm to 10uf/4.7K (V1 pin 8 to ground). Adding a 100uf cap in parallel with the 150 ohm resistor (V2 pin 3 to V3 pin 2). Changing the 10uf/4.7K/820 ohm set up to 10uf/4.7K (V3 pin 5 to V4 pin 3).

Phil, I have added some new photos of V3 and V4.