Repair Capacitors - Bulged
- Thread starter DJediMaster
- Start date
I have never done what you are looking at. I have had some serious soldering instruction in the Army courtesy of the Air Force.
Here is what I would do. I would get an old junk mainboard and practice and experiment with removing and replacing capacitors. When I thought I had a good grasp of the skills, I would do the repair.
You might want to look at getting some basic desoldering materials.
You also have to get the replacement capacitors. I think it is a conspiracy to make you get a new mainboard.
Here is what I would do. I would get an old junk mainboard and practice and experiment with removing and replacing capacitors. When I thought I had a good grasp of the skills, I would do the repair.
You might want to look at getting some basic desoldering materials.
You also have to get the replacement capacitors. I think it is a conspiracy to make you get a new mainboard.
There is a "how to" article on that in the November issue of Computer Power User magazine.
Wish I hadn't chucked out a mobo a couple months ago, would have liked to have tried that cap replacement.
Wish I hadn't chucked out a mobo a couple months ago, would have liked to have tried that cap replacement.
You need a 30w soldering iron and some 60/40 solder. A solder sucker or some wick would help too. Make sure get the right replacements.
Have been trying to figure out a stability problem w/ GVC motherboard from Andara barebones project. It had been working perfectly up until recently. Then started getting lock-ups. I tried swapping out everything, but coudn't seem to figure it out. Finally, I noticed 3 of the capacitors had a bulging top. One of them had crusty brown gunk on top. I believe this is the problem, but have never tackled replacing capacitors.
Mem
Lifer
- Apr 23, 2000
- 21,476
- 13
- 81
This guide might help,click here .Personally I think you are better going for a new motherboard if it`s out of warranty, since you may fix them and find you get some more leaking down the road.
Thanks for the link... i'll probably try to replace the capacitors....board is useless as is...my soldering skills are sorely lacking, though.
IBM had a huge problem with this on the NetVista models they had a few years ago. It's easier to replace the mobo, but hey, if the board is shot and you want to have fun, GIVER!
If you can learn something, then do it. Do replace all the capacitors of the same type that you've seen leaking, else you'll be back on the job in a couple of weeks.
This news about this problem with defective capacitors has been around for quite a while, so I have looked at my capacitors. When I checked my old (3 years?) ABIT KT7, the caps looked OK, I thought. Then one day when I was checking something else the light was just right, or something, and I could see several of these caps had somewhat bulged tops where there is an X folded into the metal. I think the X's are there exactly for the purpose of accomodating any outgassing due to heating or otherwise. Otherwise I don't see any purpose for them. Most caps don't have the X's. IAC, this mobo has never given me any problems. I almost wish it would so I would have an excuse to switch over to my new mobo (DFI NFII Lanparty) that has been operating downstairs for months. I would not get overly concerned about bulging X's unless you have functional problems. Now if they leak or split, yes get worried.
The other way of dealing with expanding gasses would be to have a tiny leak in the sealed cans.
The other way of dealing with expanding gasses would be to have a tiny leak in the sealed cans.
Have you ever seen a shotgun shell? I think those X's are from folding the sides in to make a top for the capacitor, though I could be wrong. The problem is crimped together metal like that is never that strong, so if there is any pressure, they leak.
>I think those X's are from folding the sides in to make a top for the capacitor, though
>I could be wrong. The problem is crimped together metal like that is never that strong,
>so if there is any pressure, they leak.
Sounds reasonable. But... From inspection of the real object I think there is no crimping at these X's.
Also take a look at:
pic from original post
Take a look at where the caps actually leak from that article at overclockers.com.
leaking caps that have X's on top
It's on the bottom where the plastic plug is crimped into the can.
I'll stick with my theory.
For people that don't know, electrolytic capactors are made as follows: Two very long strips of foil soaked in a slighty corrosive liquid (the electrolyte) are rolled up and sealed in a can. The reason for the long strips is that the larger the area of the conductors the larger the capacitance. The two conductors have to be insulated from one another in order to be a capacitor. The thinner the insulator, the closer the two conductors, and the higher the capacitance. To form the insulator, they run a current through the capacitor until a thin layer of non-conductive material forms on the foil. Without the electrolyte that layer could not form. This method makes a very thin insulator, which is why large value capacitors with a small physical size are generally electrolytic.
The thickness of the insulating layer also determines how high the voltage can be before it arcs over. That is the reason for the voltage rating on the capacitor. Quite often when you buy capacitors, one with a higher capacitance and lower voltage will be the same size as one with a lower capacitance and higher voltage. That's probably because the only difference is that a thicker or thinner insulator was formed during manufacture.
If you ever hook up the voltage to an electrolytic capacitor backwards, the insulator material un-forms reversing the chemical process that created it originally. It doesn't take long for the insides to heat up and probably explode.
Electrolytic capacitors leak DC slightly even when hooked up correctly. It is inevitable from the way the insulator is formed. The remaining electrolytic fluid has a tendency to dissolve the insulator. But the insulator is self repairing, due to the electrical leakage, so this is not huge problem. However if the leakage current gets too high, the fluid will get excessively hot, expand, and push itself out somehow somewhere.
Big can electrolytic capacitors - like the size of a soup can- have pressure relief valves/holes you can see in the plastic plug. I don't know about the small ones.
>I could be wrong. The problem is crimped together metal like that is never that strong,
>so if there is any pressure, they leak.
Sounds reasonable. But... From inspection of the real object I think there is no crimping at these X's.
Also take a look at:
pic from original post
Take a look at where the caps actually leak from that article at overclockers.com.
leaking caps that have X's on top
It's on the bottom where the plastic plug is crimped into the can.
I'll stick with my theory.
For people that don't know, electrolytic capactors are made as follows: Two very long strips of foil soaked in a slighty corrosive liquid (the electrolyte) are rolled up and sealed in a can. The reason for the long strips is that the larger the area of the conductors the larger the capacitance. The two conductors have to be insulated from one another in order to be a capacitor. The thinner the insulator, the closer the two conductors, and the higher the capacitance. To form the insulator, they run a current through the capacitor until a thin layer of non-conductive material forms on the foil. Without the electrolyte that layer could not form. This method makes a very thin insulator, which is why large value capacitors with a small physical size are generally electrolytic.
The thickness of the insulating layer also determines how high the voltage can be before it arcs over. That is the reason for the voltage rating on the capacitor. Quite often when you buy capacitors, one with a higher capacitance and lower voltage will be the same size as one with a lower capacitance and higher voltage. That's probably because the only difference is that a thicker or thinner insulator was formed during manufacture.
If you ever hook up the voltage to an electrolytic capacitor backwards, the insulator material un-forms reversing the chemical process that created it originally. It doesn't take long for the insides to heat up and probably explode.
Electrolytic capacitors leak DC slightly even when hooked up correctly. It is inevitable from the way the insulator is formed. The remaining electrolytic fluid has a tendency to dissolve the insulator. But the insulator is self repairing, due to the electrical leakage, so this is not huge problem. However if the leakage current gets too high, the fluid will get excessively hot, expand, and push itself out somehow somewhere.
Big can electrolytic capacitors - like the size of a soup can- have pressure relief valves/holes you can see in the plastic plug. I don't know about the small ones.
JonB
Platinum Member
I had similar problems with two MSI boards awhile back. I ordered new caps, used my soldering skills, (ex-Navy electronics technician) and it looked great.
Only problem, it still wouln't run stable. I ended up trashing both boards.
It should have worked. The solder job looked good, but something was not quite right.
I probably won't bother again. New motherboards are too cheap.
Best of luck to you, but don't be surprised if it is still unstable.
Only problem, it still wouln't run stable. I ended up trashing both boards.
It should have worked. The solder job looked good, but something was not quite right.
I probably won't bother again. New motherboards are too cheap.
Best of luck to you, but don't be surprised if it is still unstable.
Larger capacitors have that X in their cap as a controlled breaking point should they start to blow up (literally).
Any capacitor with a bulged top or obvious liquid leak is dead. There is a certain level of redundancy particularly on mainboards, but as the number of dead ones increases, stability goes down the drain first, but your expensive stuff like CPU or AGP card are at risk too. Voltage spike no good.
Any capacitor with a bulged top or obvious liquid leak is dead. There is a certain level of redundancy particularly on mainboards, but as the number of dead ones increases, stability goes down the drain first, but your expensive stuff like CPU or AGP card are at risk too. Voltage spike no good.
>Larger capacitors have that X in their cap as a controlled breaking point should they start to blow up (literally).
Any pressure sufficient to tear the metal would in fact be an explosion. So this sort of controlled breaking point would not prevent an explosion. The X's in my caps are wide indentations, not score marks or cuts. I don't think they would be weak. In fact, the capacitors in the linked pics leaked at the bottom, so that's where the weak point must be. If the engineers intended the X's to give way first, they screwed up. The X's are a little extra metal. That extra metal allows the top to expand just a bit. Therefore the top bulges before the plastic plug at the bottom pops. Maybe this helps when they are forming the insulating layer at the factory and the capacitors heat up.
Or maybe when a capacitor explodes it explodes with minutely less violence because the top bulges a bit. Maybe that's important for some reason.
>Any capacitor with a bulged top or obvious liquid leak is dead.
A capacitor should work just fine with a bulged top. Why wouldn't it? It still has capacitance. The only function of the sealed can is to keep the electrolyte from drying out or leaking out. A bulged top COULD be the result of an insulator failure. A leak does indicate something has gone wrong. Capacitors could also function after they start leaking electrolyte, and I have no doubt they often do. But who wants a capacitor leaking all over even if it keeps functioning for a while?
Now the reason I checked back. Something I learned when I was fixing a lot circuit boards at a factory is that adding a little solder to solder joints made removing the solder (with a solder sucker) go quicker. Evidently the electrical solder used for repairs melts at lower temperature than what they use to make the circuit boards. So the solder you add melts quickly. It surrounds the tip of the soldering iron making heat condution much more efficient, which makes the rest of the solder melt faster. The difference is amazing.
Another thing you can do if you don't need the component you are removing is to crush and cut it up until only the separate leads are left. Removing the leads individually takes less skill and technique, and is less likely to damage the circuit board, than trying to get a whole component out that is attached in several places.
Something people often did to get rid of the solder left in holes, so you can get the new component in, is to hold the circuit board in one hand while melting the solder with the solding iron in the other hand, and tap the circuit board sharply against the table. The solder woud come flying out and splash harmlessly on the table (or floor). If you have good technique it usually gets the hole cleaner than using a solder sucker. This sometimes will get the cut off leads out of the hole simultaneously in one operation. For that, it is better to have the cut leads long. The extra mass means more inertial force, which is what pulls it out of the hole.
One thing novices might not realize is that having the soldering tip shiny with solder makes heat transfer go much faster. Assuming your tip is properly tinned (has a permanent film of solder) just use a slighly wet rag or sponge to wipe the tip just before you use it on a solder joint.
Don't use plumbing solder. Get solder from the electrical section of Home Depot or an electronics parts store. I don't think they even allow you to use lead-containg solder in plumbing any more. 60/40 solder is supposed to be the minimum melting point solder, which makes it the easiest to work with. It is also on the expensive side, which why it is only used for electronic repairs. For plumbing, when you use a torch, you don't have to worry about temperature too much. I suppose the non-lead solders for plumbing are even more expensive though.
Any pressure sufficient to tear the metal would in fact be an explosion. So this sort of controlled breaking point would not prevent an explosion. The X's in my caps are wide indentations, not score marks or cuts. I don't think they would be weak. In fact, the capacitors in the linked pics leaked at the bottom, so that's where the weak point must be. If the engineers intended the X's to give way first, they screwed up. The X's are a little extra metal. That extra metal allows the top to expand just a bit. Therefore the top bulges before the plastic plug at the bottom pops. Maybe this helps when they are forming the insulating layer at the factory and the capacitors heat up.
Or maybe when a capacitor explodes it explodes with minutely less violence because the top bulges a bit. Maybe that's important for some reason.
>Any capacitor with a bulged top or obvious liquid leak is dead.
A capacitor should work just fine with a bulged top. Why wouldn't it? It still has capacitance. The only function of the sealed can is to keep the electrolyte from drying out or leaking out. A bulged top COULD be the result of an insulator failure. A leak does indicate something has gone wrong. Capacitors could also function after they start leaking electrolyte, and I have no doubt they often do. But who wants a capacitor leaking all over even if it keeps functioning for a while?
Now the reason I checked back. Something I learned when I was fixing a lot circuit boards at a factory is that adding a little solder to solder joints made removing the solder (with a solder sucker) go quicker. Evidently the electrical solder used for repairs melts at lower temperature than what they use to make the circuit boards. So the solder you add melts quickly. It surrounds the tip of the soldering iron making heat condution much more efficient, which makes the rest of the solder melt faster. The difference is amazing.
Another thing you can do if you don't need the component you are removing is to crush and cut it up until only the separate leads are left. Removing the leads individually takes less skill and technique, and is less likely to damage the circuit board, than trying to get a whole component out that is attached in several places.
Something people often did to get rid of the solder left in holes, so you can get the new component in, is to hold the circuit board in one hand while melting the solder with the solding iron in the other hand, and tap the circuit board sharply against the table. The solder woud come flying out and splash harmlessly on the table (or floor). If you have good technique it usually gets the hole cleaner than using a solder sucker. This sometimes will get the cut off leads out of the hole simultaneously in one operation. For that, it is better to have the cut leads long. The extra mass means more inertial force, which is what pulls it out of the hole.
One thing novices might not realize is that having the soldering tip shiny with solder makes heat transfer go much faster. Assuming your tip is properly tinned (has a permanent film of solder) just use a slighly wet rag or sponge to wipe the tip just before you use it on a solder joint.
Don't use plumbing solder. Get solder from the electrical section of Home Depot or an electronics parts store. I don't think they even allow you to use lead-containg solder in plumbing any more. 60/40 solder is supposed to be the minimum melting point solder, which makes it the easiest to work with. It is also on the expensive side, which why it is only used for electronic repairs. For plumbing, when you use a torch, you don't have to worry about temperature too much. I suppose the non-lead solders for plumbing are even more expensive though.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter