Gibson GA-95 RVT
Moderators: pompeiisneaks, Colossal
Re: Gibson GA-95 RVT
I may be mistaken, but I think the shielded cable is important for the reverb to operate without noise. Anyhow, seeing you don't have the stock f/s, I think you have nothing to lose.
Re: Gibson GA-95 RVT
I figured it was worth a shot, but was going to examine the innards of my switch first.Phil_S wrote:I may be mistaken, but I think the shielded cable is important for the reverb to operate without noise. Anyhow, seeing you don't have the stock f/s, I think you have nothing to lose.
Anyway... I checked R12, and it's actually a 1.3k resistor. R13, R28, and R29 all check out fine. I'd replaced two of them earlier on. I also jumpered pins 4 and 5 on the footswitch.
I will report dramatic improvement on the trem channel, both in volume and tone. However, it's not 90 watts of power, and the normal channel is still weak. But, it's significant progress.
Current is now steady on V5, pin 8. Turning the pot only got me to .75v. The reading at pin 6 was 151v at that setting.
Across C15, I measure -2v.
Re: Gibson GA-95 RVT
I think, probably R12 is meant to be 1.5K from the Apollo schematic - near impossible to read.
C15 is leaky and needs to be replaced. If you don't have a 1uf cap, use whatever you can find up to 20uf. The higher the cap value, the more bass you'll let through. You can always change it later.
The job of the cap, by nature of it's construction is to block DC. If there is any DC across the cap, it should be replaced. What's happening there is that the cap is creating a path straight to ground and the resistor isn't doing anything, or very little. The electrons find the most efficient path to ground, and the R across the cap is less than R28. A new cap will have infinite R, forcing R28 to be the path to ground.
A cap (simplified) is two sheets of foil with a non-conductive dielectric layer in between. Roll it up and cover it. Draw a lead out from each foil. Without a direct connection, it is impossible for DC to pass from one layer to the other. AC (your instrument signal) will pass, as that is the nature of a sine wave.
So, maybe you see why that cap has to go. Even if you just clip it and don't replace it, you'll see the plate voltage drop. It's OK to do that. The tone won't be right, but circuit-wise, there's no problem doing it. I think, because the two sections of the triode draw from the same node, both are affected by the leaky cap, but I'm not really sure about that.
Good to see progress, even in small steps!
C15 is leaky and needs to be replaced. If you don't have a 1uf cap, use whatever you can find up to 20uf. The higher the cap value, the more bass you'll let through. You can always change it later.
The job of the cap, by nature of it's construction is to block DC. If there is any DC across the cap, it should be replaced. What's happening there is that the cap is creating a path straight to ground and the resistor isn't doing anything, or very little. The electrons find the most efficient path to ground, and the R across the cap is less than R28. A new cap will have infinite R, forcing R28 to be the path to ground.
A cap (simplified) is two sheets of foil with a non-conductive dielectric layer in between. Roll it up and cover it. Draw a lead out from each foil. Without a direct connection, it is impossible for DC to pass from one layer to the other. AC (your instrument signal) will pass, as that is the nature of a sine wave.
So, maybe you see why that cap has to go. Even if you just clip it and don't replace it, you'll see the plate voltage drop. It's OK to do that. The tone won't be right, but circuit-wise, there's no problem doing it. I think, because the two sections of the triode draw from the same node, both are affected by the leaky cap, but I'm not really sure about that.
Good to see progress, even in small steps!
Re: Gibson GA-95 RVT
I replaced the C15 cap with a 10uf cap. Tone is good on the effects channel, and volume is up, but it's down to about zero on the normal channel and the tone is very thin there.
Voltage at V4 pin 1 is now 212v. Voltage at pin 6 is now a whopping 252v. Voltage across C15 is now 2.1v.
Voltage at V4 pin 1 is now 212v. Voltage at pin 6 is now a whopping 252v. Voltage across C15 is now 2.1v.
Re: Gibson GA-95 RVT
I believe I made an error with respect to C15 and I'm sorry, but it did not turn out to be a fool's errand, either. Because C15 is in the circuit, clipping onto both sides of it, you are also clipped onto both sides of R28. In retrospect, I see that 2.1v (basically unchanged from the prior reading) is the cathode voltage, which the schematic says should be 4.5v. Unfortunately, this is a consequence of working with an admitted amateur. I'm not surprised a 10u cap raised the voltage.
No real harm done there and no danger to the amp. The tube can handle the voltages you report and even more. Leave that replacement cap in there. In the end, I was going to suggest you might like the effect of raising that one anyway.
We still need to figure out what is allowing plate voltage to rise. When you are able to get plate voltage down, cathode voltage will probably come up to spec.
I think, at this point, I would deploy a divide and conquer strategy. Since we have channel 2 working pretty well, let's see if we can get it fixed and ignore channel 1 for a while. We should remove channel 1 from the circuit. To do that, lift R14 (330K) where it meets C25 (.01) and R30 (68K). This will leave the amp with only channel 2 (reverb) operative.
While you are working at that solder cup (all three parts should be terminating in the same place, but with a Gibson, no bets will be taken), lift that end of R30 (68K) as well. Then, you can meter R30 out of circuit and confirm it is in spec. If out of spec, replace it. If it meters OK, reconnect it.
Let me know what happens. Though not very scientific, tell me if you think it is approaching the sort of volume you'd expect from a 75-100W amp. Please provide new plate and cathode readings for V1 (6,4), V4 (1,3), and V5 (1,3).
If you are inclined to go further at this point, lift one end of R41 (470K), coupled to C25, maybe the same cup, to confirm if it is out of spec or not. Then check R42 (1K) to see if it is out of spec.
Here you can also do a proper check for a leaky coupling cap. Put your meter across C25 to check for DC flowing across it. Because both ends are in the circuit (not one end to ground) this will be a valid check, unlike the error I made with C15 <still embarrassed about that>. If it leaks, that one is important to replace with a like value.
Replace any out of spec resistors as you find them. I would say +/-20% of spec can wait for replacement, though I'd want to be within ~5% before I closed up the amp.
Something is eating the signal on ch. 1. This could be a bad ground, a bad cap, a bad resistor or pot, or an unintended ground. I think you said the tubes are known to be good, so I'm figuring they are not the problem. I am guessing that lifting R14 to disconnect it may allow ch2 to operate properly. If this happens, that's very good news. From there, we disconnect ch 2 by lifting R30 where it meets C25, and reconnecting R14. Then we will have ch 2 out of circuit and only ch 1 active. It will be easier to track down the problem when we keep each channel fully isolated.
I had similar problems on my GA20-RVT and this is how I got to the bottom of it. I may be an amateur, but I also have the benefit of working on an amp very similar to this with a good result. I just want you to know, I'm not making this up as we go along. I think I understand what we are chasing here. The preamp sections of the two amps are strikingly similar. Please stick with this.
No real harm done there and no danger to the amp. The tube can handle the voltages you report and even more. Leave that replacement cap in there. In the end, I was going to suggest you might like the effect of raising that one anyway.
We still need to figure out what is allowing plate voltage to rise. When you are able to get plate voltage down, cathode voltage will probably come up to spec.
I think, at this point, I would deploy a divide and conquer strategy. Since we have channel 2 working pretty well, let's see if we can get it fixed and ignore channel 1 for a while. We should remove channel 1 from the circuit. To do that, lift R14 (330K) where it meets C25 (.01) and R30 (68K). This will leave the amp with only channel 2 (reverb) operative.
While you are working at that solder cup (all three parts should be terminating in the same place, but with a Gibson, no bets will be taken), lift that end of R30 (68K) as well. Then, you can meter R30 out of circuit and confirm it is in spec. If out of spec, replace it. If it meters OK, reconnect it.
Let me know what happens. Though not very scientific, tell me if you think it is approaching the sort of volume you'd expect from a 75-100W amp. Please provide new plate and cathode readings for V1 (6,4), V4 (1,3), and V5 (1,3).
If you are inclined to go further at this point, lift one end of R41 (470K), coupled to C25, maybe the same cup, to confirm if it is out of spec or not. Then check R42 (1K) to see if it is out of spec.
Here you can also do a proper check for a leaky coupling cap. Put your meter across C25 to check for DC flowing across it. Because both ends are in the circuit (not one end to ground) this will be a valid check, unlike the error I made with C15 <still embarrassed about that>. If it leaks, that one is important to replace with a like value.
Replace any out of spec resistors as you find them. I would say +/-20% of spec can wait for replacement, though I'd want to be within ~5% before I closed up the amp.
Something is eating the signal on ch. 1. This could be a bad ground, a bad cap, a bad resistor or pot, or an unintended ground. I think you said the tubes are known to be good, so I'm figuring they are not the problem. I am guessing that lifting R14 to disconnect it may allow ch2 to operate properly. If this happens, that's very good news. From there, we disconnect ch 2 by lifting R30 where it meets C25, and reconnecting R14. Then we will have ch 2 out of circuit and only ch 1 active. It will be easier to track down the problem when we keep each channel fully isolated.
I had similar problems on my GA20-RVT and this is how I got to the bottom of it. I may be an amateur, but I also have the benefit of working on an amp very similar to this with a good result. I just want you to know, I'm not making this up as we go along. I think I understand what we are chasing here. The preamp sections of the two amps are strikingly similar. Please stick with this.
Re: Gibson GA-95 RVT
Thanks for outlining the next steps. Actually, I'm far from frustrated by all this. It's given me a very valuable learning experience, which is why I picked it up in the first place. A lot of this, coupled with reading other materials, is starting to make more sense to me. It's really a lot of fun sorting through all this and stopping to read up on the stuff I don't know or don't fully understand.
Re: Gibson GA-95 RVT
I was looking at your picture of V1-V4 and there is a bit of navigation I can help with. You'll see a .0047 400v brown cap in a vertical orientation in front of the right side of the V4 socket, attached to the board (what cup#?). There is a wire that goes from that cup back to V3 pin 4. V3 pin 4 is at the left bottom of the schematic in the tremolo section. It is the cathode and goes to ground. Is that cup (#12?) part of the ground rail?
There is a difference here between the Epi schematic and the Apollo. The Apollo has a notch filter between (looking at the Epi schematic) C25 and V5 pin 2. The notch filter consists of the triangle formed by a pair of 220K resistors, the .0047 cap to ground, and a barrel cap; the triangle appears to be floating above V5. V5 pin 2 is obstructed by the white barrel cap. I assume there is a wire that runs from V5 pin 2 to cup #14(?). This is where there is a new mylar cap, probably C25 (.01) that spans #13-#14. Since this cap appears new, it is probably not leaky. Anyway, I think you'll find R30 and R14 meet C25 at cup #13(?).
Since it looks like R29, R30, R13, and R14 are all new replacements, they are probably in spec and if you verified them with your meter before you put them in, you don't need to verify them again. However, if you didn't meter them and misread the stripes (a common error) that could potentially be a problem. (I can tell you this from personal experience, always use your meter before you install a resistor, as it saves headaches and I've had a couple of real bangers.) I know this board is terrible to work on. If you have any doubts about a cold solder joint in those cups, you'll need to reflow the solder. Also, make sure none of the metal leads are making unintended contact with any other parts.
There is a difference here between the Epi schematic and the Apollo. The Apollo has a notch filter between (looking at the Epi schematic) C25 and V5 pin 2. The notch filter consists of the triangle formed by a pair of 220K resistors, the .0047 cap to ground, and a barrel cap; the triangle appears to be floating above V5. V5 pin 2 is obstructed by the white barrel cap. I assume there is a wire that runs from V5 pin 2 to cup #14(?). This is where there is a new mylar cap, probably C25 (.01) that spans #13-#14. Since this cap appears new, it is probably not leaky. Anyway, I think you'll find R30 and R14 meet C25 at cup #13(?).
Since it looks like R29, R30, R13, and R14 are all new replacements, they are probably in spec and if you verified them with your meter before you put them in, you don't need to verify them again. However, if you didn't meter them and misread the stripes (a common error) that could potentially be a problem. (I can tell you this from personal experience, always use your meter before you install a resistor, as it saves headaches and I've had a couple of real bangers.) I know this board is terrible to work on. If you have any doubts about a cold solder joint in those cups, you'll need to reflow the solder. Also, make sure none of the metal leads are making unintended contact with any other parts.
Re: Gibson GA-95 RVT
Phil, I lifted R14 and checked R30, R41, and R42. All are within spec. Testing the amp at this point, I noticed no increase in volume vs. last night's test. Keep in mind that last night we marked a significant improvement.
I read -193v across C25, which seems very weird.. On the tubes:
V1, p4 1.1v
V1, p6 191v
V4, p1 207v
V4, p3 2v
V5, p1 286v
V5, p3 1.9v
The .047 brown cap does connect to pot #12. R30 and R14 connect at pot #13, as you said. As to whether pot #12 is part of the ground rail-I thought the ground rail was the middle row of pots. I know it sounds stupid that I'm not sure. If that's not the case, it calls into question the orientation of C29A, which has it's positive lead on the bottom row and the negative lead on the middle row. I'm a bit confused on this topic, but I think I'm correct.
I read -193v across C25, which seems very weird.. On the tubes:
V1, p4 1.1v
V1, p6 191v
V4, p1 207v
V4, p3 2v
V5, p1 286v
V5, p3 1.9v
The .047 brown cap does connect to pot #12. R30 and R14 connect at pot #13, as you said. As to whether pot #12 is part of the ground rail-I thought the ground rail was the middle row of pots. I know it sounds stupid that I'm not sure. If that's not the case, it calls into question the orientation of C29A, which has it's positive lead on the bottom row and the negative lead on the middle row. I'm a bit confused on this topic, but I think I'm correct.
Re: Gibson GA-95 RVT
Unless you reconnected something wrong or if C25 is a straight short, you don't have 193VDC. Next time you look at it, see if you can check again, as it seems all wrong. The cap has two legs, one in each cup. Put one probe in each cup (13-14). You should see 0 or maybe mV. If you are really seeing 193V, stop here, take closeup pictures of that cap from each side of the chassis, so I can have a good look at where it is installed and what it's connected to. If you really have this sort of voltage here, it is probably due to a miswire and has to be addressed.
You are right that the ground rail runs down the middle. This is an exception (for the .0047). If you look carefully, you'll see a series of wires making a string through the middle of the board (not every cup) carrying the ground connections. That large blue e-cap hanging off the board is installed correctly. The - lead is in the last cup for the ground rail on the board. My GA20-RVT is identical in this respect.
Cathode voltage on V5 (pin 3) is too low. I am trying to get you to move towards that one. Lift the cathode bypass cap there (C26; black cap at #15?) and see what happens. Also, please confirm that R41 (470K) is grounded at one end. If unsure, try to lead me to it on your picture. I don't see R41. I want it to go from #14 to ground. Ick, Gibson made a mess of this. R41 to ground is also essential for proper operation.
Just an aside, I call things on the board "cups". Eyelets is OK. "Pots" is confusing, as a pot is a potentiometer, like a volume control. I'd avoid this reference for clarity.
Now that you mention that blue cap hanging off the board, before we finish, I'd feel better if it can be relocated and mounted more appropriately. Not for now; just leave it until we get the amp working.
You are right that the ground rail runs down the middle. This is an exception (for the .0047). If you look carefully, you'll see a series of wires making a string through the middle of the board (not every cup) carrying the ground connections. That large blue e-cap hanging off the board is installed correctly. The - lead is in the last cup for the ground rail on the board. My GA20-RVT is identical in this respect.
Cathode voltage on V5 (pin 3) is too low. I am trying to get you to move towards that one. Lift the cathode bypass cap there (C26; black cap at #15?) and see what happens. Also, please confirm that R41 (470K) is grounded at one end. If unsure, try to lead me to it on your picture. I don't see R41. I want it to go from #14 to ground. Ick, Gibson made a mess of this. R41 to ground is also essential for proper operation.
Just an aside, I call things on the board "cups". Eyelets is OK. "Pots" is confusing, as a pot is a potentiometer, like a volume control. I'd avoid this reference for clarity.
Now that you mention that blue cap hanging off the board, before we finish, I'd feel better if it can be relocated and mounted more appropriately. Not for now; just leave it until we get the amp working.
Re: Gibson GA-95 RVT
I measured the voltage correctly, so I stopped there and took these pics.
You do not have the required permissions to view the files attached to this post.
Re: Gibson GA-95 RVT
Your board might be conducting. Take a good look at pic 002 and you'll see flux bridging the space between cups 13 and 14. Dip a Q-tip in 90% isopropyl alchohol and clean off the flux between cups 13 and 14. While you are at it, clean up any other suspect areas or excess flux around and between the solder cups, as well as any dirt that may have accumulated.
If you are comfortable doing it, put the alcohol in a spray bottle and "hose down" the board. You'll have to work quick with a rag on the end of a chopstick or something like that because the alcohol evaporates quickly. This can be done in small sections. There are also commercial de-fluxing solutions in aerosol cans, but they cost considerably more than alcohol.
Avoid the more ordinary 70% solution, as it has too much water in it. You can get this at any pharmacy, Target, etc. Ask if you don't see it on the shelf.
It is still possible the cap is bad. My thinking is that this sort of cap rarely goes bad and it appears to be in physically good condition. I'm not clear if it's new. After you clean off the flux, search for other possible shorts and come up with nothing else, replace the cap. If you replace the cap, I suggest stopping to clean flux from the board after removal, and again after installing the new one.
While you are in there, and only after you solve the problem of the short circuit, you might consider removing the notch filter. Only change one thing at a time at this point or we won't know what causes what. The notch filter consists of 4 parts: the pair of 220K + white barrel cap that form a triangle floating above V5 plus the .0047 cap to ground. This isn't needed and in my view it is a tone killer. It will make this part of the amp a bit simpler. In place of the notch filter, run a bit of wire from the junction of C25/R41 (cup 14?) to V5 pin 2 (where you disconnected the triangle from). This is short low voltage run -- any 300v rated wire is fine there.
If you are comfortable doing it, put the alcohol in a spray bottle and "hose down" the board. You'll have to work quick with a rag on the end of a chopstick or something like that because the alcohol evaporates quickly. This can be done in small sections. There are also commercial de-fluxing solutions in aerosol cans, but they cost considerably more than alcohol.
Avoid the more ordinary 70% solution, as it has too much water in it. You can get this at any pharmacy, Target, etc. Ask if you don't see it on the shelf.
It is still possible the cap is bad. My thinking is that this sort of cap rarely goes bad and it appears to be in physically good condition. I'm not clear if it's new. After you clean off the flux, search for other possible shorts and come up with nothing else, replace the cap. If you replace the cap, I suggest stopping to clean flux from the board after removal, and again after installing the new one.
While you are in there, and only after you solve the problem of the short circuit, you might consider removing the notch filter. Only change one thing at a time at this point or we won't know what causes what. The notch filter consists of 4 parts: the pair of 220K + white barrel cap that form a triangle floating above V5 plus the .0047 cap to ground. This isn't needed and in my view it is a tone killer. It will make this part of the amp a bit simpler. In place of the notch filter, run a bit of wire from the junction of C25/R41 (cup 14?) to V5 pin 2 (where you disconnected the triangle from). This is short low voltage run -- any 300v rated wire is fine there.
Re: Gibson GA-95 RVT
Great stuff, Phil. I'll take care of cleaning up tonight. I wasn't aware of the flux issue. 
The cap in question is new, but I'll replace it if necessary.
I've also heard that the notch filters are a tone killer. After you pointed it out last night, I looked it over and read up a bit on the subject. I'll plan to remove it at some point after I get the amp working properly.
The cap in question is new, but I'll replace it if necessary.I've also heard that the notch filters are a tone killer. After you pointed it out last night, I looked it over and read up a bit on the subject. I'll plan to remove it at some point after I get the amp working properly.
Re: Gibson GA-95 RVT
I like using pure alcohol for cleaning flux.
I use a toothbrush after letting the alcohol sit on the board for a bit.
Then Q tips. Although the fibers can get caught on stuff.
Not sure if that type of board can absorb moisture like the old Fender fish paper boards and become conductive but cleaning the flux off can only be a good thing.
There is much cleaner burning solder nowadays as well.
So if you like you can try to find some Kester # 245 "No Clean" 63/37 solder.
I like the .031" wire.
I use a toothbrush after letting the alcohol sit on the board for a bit.
Then Q tips. Although the fibers can get caught on stuff.
Not sure if that type of board can absorb moisture like the old Fender fish paper boards and become conductive but cleaning the flux off can only be a good thing.
There is much cleaner burning solder nowadays as well.
So if you like you can try to find some Kester # 245 "No Clean" 63/37 solder.
I like the .031" wire.
Tom
Don't let that smoke out!
Don't let that smoke out!
Re: Gibson GA-95 RVT
Hey Tom! Welcome to this thread. Misery loves company. 
From what I can tell, the upper board is phenolic and not particularly permeable. There is an insulating board underneath that is of softer material, possibly fish paper.
You've sparked a worthy comment here, which is that the OP (Travst) should be careful not to push component leads too far into those solder cups, lest he go through the insulating board to the chassis. Admittedly, this would be difficult to do, but it brings up a note of caution.
Toothbrush is a great idea. I agree that the Q-tip fibers are a little problematic. I pick them off with a tweaser and what won't come off I torch with one of those butane lighter sticks (very carefully).
When I had the board removed from my GA20-RVT, I held it vertical and hosed it with alcohol, which really cleaned it up. I am not recommending it here, as he'll be days reconnecting everything and is too likely to miswire something.
I'm attaching pictures of a clean board I've got (one of those projects that has no end date in sight). The cups are about 3/8" deep. You can see they are V shaped and the bottom has an X sliced in it that goes nearly all the way up to the top.
From what I can tell, the upper board is phenolic and not particularly permeable. There is an insulating board underneath that is of softer material, possibly fish paper.
You've sparked a worthy comment here, which is that the OP (Travst) should be careful not to push component leads too far into those solder cups, lest he go through the insulating board to the chassis. Admittedly, this would be difficult to do, but it brings up a note of caution.
Toothbrush is a great idea. I agree that the Q-tip fibers are a little problematic. I pick them off with a tweaser and what won't come off I torch with one of those butane lighter sticks (very carefully).
When I had the board removed from my GA20-RVT, I held it vertical and hosed it with alcohol, which really cleaned it up. I am not recommending it here, as he'll be days reconnecting everything and is too likely to miswire something.
I'm attaching pictures of a clean board I've got (one of those projects that has no end date in sight). The cups are about 3/8" deep. You can see they are V shaped and the bottom has an X sliced in it that goes nearly all the way up to the top.
You do not have the required permissions to view the files attached to this post.
Re: Gibson GA-95 RVT
I wondered about putting leads in too far. I went back and trimmed all the component parts that I'd replaced for that reason. I also wonder if hot solder could have flowed beneath the board and bridged a connection there. I don't mean flowed along the underside of the board, but rather if a stray lead were beneath the eyelet.