linear1 forums
[part of the linear1 network]
+ [linear1 case mods]
+ [LED Center]
+ [privacy policy]
Welcome, Guest. Please login or register.
March 21, 2010, 02:22:25 AM

Login with username, password and session length
Search:     Advanced search
* Home Help Search Login Register

linear1 forums  |  LED discussion  |  Electronics discussion  |  Topic: Where to buy capacitors. « previous next »
Pages: 1 [2] Print
Author Topic: Where to buy capacitors.  (Read 16562 times)
justDIY
Microcontroller Madman
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1652


UFOric


WWW
« Reply #15 on: January 15, 2006, 06:23:33 PM »

At the same time, I was sent to badcaps.com, from here, and in the forum, in the cap replacement guide, it says to use an iron that puts out 350C, but to make sure it's grounded to the mobo, grounding to the metal around the mouse/USB ports works. Rob posted the link, and I had already been there myself, so I figured that those guys knew what they were talking about.

They do everything with a soldering gun(electric), so point me in the right direction for grounding myself and the gun, if you could. Thanks.  Cool

any iron can get to 350C (roughly 660F), however, its the how easily can the iron maintain that temperature is what counts.   The gun you describe (100/140) sounds a lot like a weller gun I have that is used to burn wood, cut and shape plastic and do stained glass ... it contains an isolation transformer, and forms a direct short across the tip of line voltage AC current (couple of bad things there).

I say they are bad because of what I know about the theory ... whether it actual hurts anything or not, I cannot say.   If the folks on badcaps are using wood burning 'guns' to repair motherboards, then I suppose its something you can get away with.   I've taken mine apart before, the only part that is grounded is the metal frame around the transformer ... if you'd like, I can measure tip potential to ground on the iron I have.

most of the recommendations you read on the web about "be sure you're grounded!!!", in my humble opinion, are just over-amplications of the warnings that ship with every single piece of electronic gear, parts, pieces and devices ... in other words, I've never lost any semiconductors to ESD (electro static discharge), dispite taking none but the most basic precautions to prevent them (in the winter, I often wear a fleece sweater jacket in my lab (its a bit cold at times)... that thing is a regular static-generator.    Sure, you'll read accounts like "Dude, I killed my $5000 processor because I touched it without my three grounding straps attached (wrist, ankle and... elsewhere)", but I think those people are a) lying and damaged the part by some other means b) lying and didn't damage anything.

ok, enough chatter ... here's what I dont like about wood burning solder guns:

1) the tip is at a high voltage potential to ground.   Contact with a part at a different potential (ground or otherwise) will cause current to flow.   Random currents flowing through a board can damage parts

2) to a lesser degree, the gun and tip generate a powerful EM field, and can induce stray voltage (and current) into the coils and traces of the board, causing random currents to flow, damaging parts

now, the reason it people can 'get away' with it

1) semiconductor manuf aren't dumb - they know people like to try and break things (like me and my lack of ESD precautions)... They build protection into their devices to ward off things like stray currents, esd and random induction.

2) the board manuf also isn't dumb - they know their equipment will be exposed to a wide range of em radiation (look at all the electrical crap inside a running box), as well as the random visit from an esd, so, additional protections (like ground planes) are used to help protect the assembled product.

here's another recommendation for you to think about, since the paint stripper gun is out.

get the highest wattage soldering iron (not gun) you can find, probably 60-70 watts, and use something called the flooded tip method to desolder (warning, heat damage to components possible)

get your iron good and hot, and then melt as much solder as you can on the tip, let it form a little blob, if it falls off, you used too much, try again.   once you get a nice blob of molten solder on your tip, ease it onto the area you're trying to desolder.   This might take a few attempts, since it has to heat up that area too.   once you manage to get your iron all the way to the board, without the blob rehardening, the iron will conduct its heat much better into the surrounding solder - you should see a 'wave' of melting solder start to radiate away from the solder tip

the reason you'll have trouble with any method, other than the paint stripper gun, is as soon as your tip makes contact with the board, the board starts doing what one of the design intentions is, sinking heat.   there is a lot of copper and lead and other metals to absorb a lot of heat, including large electrically connected expanses called ground and supply planes (which your caps will be connected to), so you need to provide enough heat fast enough to melt the solder locally at your part, so you can insert it (or remove the old pins)

The reason the paint stripper gun works so well is, there's no direct connection between the source of heat inside the stripper, and the motherboard, so it can't get "cooled off" by the board like an iron would.   It continues to pump out super hot air regardless of whether the board is absorbing it or not..

one last tip - when you're ready to remove the old caps, dont try to desolder them intact ... carefully rip them off the board (try twisting the cap to break its pins)  .... then you only have to desolder the little pins instead of the entire cap (filled with liquid goo)

that might be an option too... break off the old caps, and solder the new caps to the pins of the old ones, not messing with the board at all.

Logged

Want to contact me directly? gmail gordonthree
My Project Blog - http://projects.dimension-x.net

Favorite numbers:
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #16 on: January 15, 2006, 06:58:06 PM »

Thanks DIY, you've just backed up a lot of things I read in the badcaps forum. I'm always looking for an opinion other than the "one" opinion expressed in one forum. Your opinion is the same, very cool.

One thing mentioned there, which you mentioned also, was the ground plane, which sucks away the heat. It was suggested over there to keep the heat on the mobo, in the particular area you're working in, with a heat gun or at the least a hair dryer, so the ground plane/heatsink doesn't get the chance to dissipate the heat. Probably a good idea, I'll make sure to do that. Although I personally don't use my hairdryer anymore, my scalp is dry enough(what hair?), heh heh heh, it does come in handy after the dog has her bath.  Grin

Another thing mentioned by one poster was pulling the cap off the leads before desoldering, as you have also suggested. This made sense to me, but was "foo-foo'ed" by others in the forum. Seems the easiest way to do it to me, and I do believe that's how I'll go about it.

I'll be doing this next Saturday, the 21st, and I'll be sure to post pics and my results, good or bad.

Wish me luck, heh.  Grin
Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
SurJector
Smart like tractor
***
Offline Offline

Posts: 107


« Reply #17 on: January 17, 2006, 04:26:41 AM »

I'd also to know if that provided it has a 3 prong plug that it's definitely grounded.
Speaking from the other side of the Atlantic, I have a three pronged iron that looks grounded (I've measured the resistance from the tip to the ground pin and it's in the 2ohm range, while the resistance from tip to the other pins is > 20Mohm). I've not measured the potential while connected though, so don't take from granted that MY iron tip is grounded and much less that YOURS will be.
Logged
Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #18 on: January 20, 2006, 06:06:43 PM »

Well, so it begins.  Grin

First off, scored a soldering gun. I was going to get a Weller iron, but hey, only 2 prong plug, and not polarized, I know that's not grounded. Here in Ontario, Canada, we're talking $24.99 for a 25watt iron. 40watt iron is $34.99. Not grounded. I see a "no-name", as in "Mega-Power" soldering gun, 100watt, $14.99, comes with solder, paste, and a spare tip. Oh yeah, and a polarized plug, 2 prong, but polarized. I get the idea this may be grounded. I scoop it up.

So, large prong to tip, switch on, infinity. Small prong to tip, switch on, infinity. Prong to prong, switch on, 0 ohms resistance. Short the prongs, go from prongs to tip, switch on, infinity. Wow, grounded tip in a $15 soldering iron. Cool, or "hot", as the case may be. Here's my new toy, gets hot enough to make the tip glow red if I hold the trigger long enough.  Wink



If I'm correct in my assumtion that this gun is grounded, what I did will be fine, if not, woe is me. I took absolutely no grounding precautions whatsoever, I put my faith in the gun. Desoldering the caps was a breeze, a few seconds of heat, out they popped. I had read about clearing the holes, not a problem, mine were clear, I could see through them, so I figured the new caps would slip right in. A bit of heat on the backside of the board, they popped right through without any effort at all.

My soldering job leaves a wee bit to be desired, I have to clean this up before I power up, but it went well. Now, if I heat the side away from the excess solder, the solder will be drawn to the heat, right? I just don't want any solder on the traces. Here's the solder job, feel free to critique it, it's my first shot at something this small, I couldn't actually see what I had done until I took a pic in macro-mode, my eyes don't focus that small, heh heh heh.  Grin





Here be the new caps, hopefully happy in their new home:



So, all is well? Not quite, unfortunately.  Embarrassed I should have looked at the board more carefully when I first started this project. There's another bad cap. Problem is that it's 2200uf, 10volt, the caps I swapped out are 2700uf, 6.3volt. Obvious newb question here, will a 2700uf/6.3volt cap work here?



I have my mobo sitting here, I want to swap everything in my current system over, but man, if I have another bad cap, what's the point? I'm ready to swap out the bad cap for a 2700uf, I really can't see paying another $18.50 in shipping handling for a $3 part.  Angry  That works out to $25 for a freakin capacitor!

So, any ideas, or anyone definitely know, if this 2700uf/6.3V might work in place of a 2200uf/10V cap?

Thanks for all your help so far guys, much appreciated.  chug-a-lug
Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #19 on: January 21, 2006, 09:07:24 AM »

So far, so good, got the 1.26 running at 1.35, but it's not stable any higher, I do believe that other bad cap has something to do with it. I couldn't get these puppies working at all before swapping the caps on this board, now it's smokin, up to 143 on the FSB, then it locks. But, at 1.35, this sucker rocks, gonna have one happy kid when he gets this system.  oh yeah



What's the worst case scenario if I use a 6.3 volt cap in place of the 10V? It blows? This I could live with, I'll be careful removing the old one, if needed, I can put the "almost blown" cap back on.

Here's my theory about blown caps: if the cap is blown, it's doing nothing, as in it's basically useless. If I put a lower voltage cap on, at least there is a "working" cap on the board. It may not last as long as a 10V, but at least it's there, the 10V is toast.

Opinions? Suggestions? I don't mind risking a cap, I do however mind risking the mobo. Anyone tried using a lower than spec capacitor?

Help?  Huh
Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
Rob
LED guy
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1970


The constant-current gardener


WWW
« Reply #20 on: January 22, 2006, 01:53:40 AM »

What's the worst case scenario if I use a 6.3 volt cap in place of the 10V?

  Shocked Shocked Shocked

worst case if you do a lot of work on the board is that you lift the traces, and that would suck.

Great pics BTW and nice job on the soldering.
Logged

Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #21 on: January 22, 2006, 10:34:18 AM »

Heh heh heh, nice pic Rob, I hope I don't get the same results.  Grin

I have 3 caps left, so I'll be swapping them out this morning, with any luck, I'll get this board clocking like I want it to. Even though I want to do all them at once, I have to stop myself and let the board cool between caps. I don't have a lot of patience, drives me nuts waiting.  Roll Eyes

BTW, thanks, but I do believe my soldering skills need just a wee bit more work.  Wink
Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
justDIY
Microcontroller Madman
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1652


UFOric


WWW
« Reply #22 on: January 22, 2006, 11:06:46 AM »

you can't replace a 10v rated cap with a 6.3v rated cap and expect it to just work...  if your board is working well, just leave well enough alone - a few extra mHz won't result in a percievable improvement in processing ability - the only place you'll see it is benchmarks.

exceeding a capacitors rated voltage results in dielectric breakdown, which will quickly ruin the capacitor.  depending on the current used to charge the capacitor, you could have a big bang like robs comical picture, but more likely you'll just have the capacitor get hot, ooze goo and generally smell bad

this may not happen instantly either, without knowing more details, like the surge rating on your caps versus the WVDC rating, and how much current and voltage the motherboard is putting in that capacitor... maybe 10v was too small as well and they should have used 16, and thats why the cap failed and you're replacing it.

Logged

Want to contact me directly? gmail gordonthree
My Project Blog - http://projects.dimension-x.net

Favorite numbers:
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
Rob
LED guy
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1970


The constant-current gardener


WWW
« Reply #23 on: January 22, 2006, 11:39:07 AM »

I wasn't clear on why you suspect this other cap is bad, myself.
Logged

Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #24 on: January 22, 2006, 06:20:38 PM »

Second time attempting to make this post, should have turned off "Dynamic Bursting" in the BIOS, locked me up tight. Things you forget after tweaking, heh heh heh.  Embarrassed

Anyhow, I'm back, 2 new caps installed. Why, check the last pic I posted, you can see how the cap in the foreground is bulging, if not, I know it's hard to see, check this pic, the one on the right, in person, it's obvious.



The original cap that was bulging, was the only one. By the time I decided to change it, another was bulging. As I was told in the badcaps forum, if one cap of any particuler "flavour" is bad, change all the caps of that flavour, they're probably all from the same batch. On my mobo, this turned out to be true with the 2700uf/6.3V caps, as the other started to bulge later on. Because the one 2200uf/10V was bulging, I figured it would be a good idea to swap out the other beside it, while I was there, no point having to rip my system apart again.

Also, at the badcaps forum, I was told that there was no problem dropping down to a 6.3V cap, the board itself only uses 5V or 12V on the board. 6.3V is fine on the 5V circuit, but obviously, not an option on the 12V circuit. I was also told that the 2700uf rating, compared to the 200uf of the original, could only be beneficial.

So far, so good, I'm running on the board as I type, so it seems to have worked.  oh yeah

These systems are to keep my boyz happy until they buy their own, 1.35mhz will do very well, good enough for NFSU/HL2/Far Cry/Flash Point, @1024X768 with lotza eye-candy and great frame rates, with an eVGA 6600GT AGP. Both kids(14 and 16) work at restaurants. My oldest, Josh, bought an XBox after working with me last summer. My youngest, Aaron, is going to buy a PS2 next payday. They don't game a lot on the computer, but when they do, I want it to be at least bearable for them. After this, it's up to them to upgrade their systems, the Tualies will be used for folding@home.

Quote
you can't replace a 10v rated cap with a 6.3v rated cap and expect it to just work...  if your board is working well, just leave well enough alone - a few extra mHz won't result in a percievable improvement in processing ability - the only place you'll see it is benchmarks.

justDIY, I'm an avid overclocker, every single mhz counts to me, it's a disease, heh heh heh.  Grin  Saying that one single mhz doesn't matter is like saying that one vote doesn't matter in the next election. Everything matters. The diff between 1.26mhz running NFSU and 1.35mhz means that I can crank up ALL the eye-candy, and still get excellent frame rates.

Never underestimate the advantage that overclocking gives you, it makes one hell of a huge difference, especially in gaming. If you're not a gamer, you'd probably never notice it. I AM a gamer, I notice. This is all about getting more bang for your buck. Personally, I don't run banchmarks, 3DMark benches are a joke in my opinion, they are synthetic, they don't bench "realworld" gaming, they don't even use a "real world" game engine, so I don't waste my time, or HDD space, using them. They are for "bragging rights", nothing more, and those bragging rights are all about..... nothing, it's freaking synthetic!   Roll Eyes

Every single mhz I squeeze from my CPU/GPU, every wee bit of speed I can score out of my system/vidcard RAM, that all makes a huge diff to me, it makes my gaming experience better.

These wee Tualatin CPUs, a couple of the few "unlocked" Intel puppies out there, run @100mhz FSB/6.5 multiplier, stock. That's 650mhz. Because they're "unlocked", they wil run @133mhz FSB/9.5 multiplier. Intel did not release a whole lot of these. These suckers are capable of 190mhz FSB/9.5 multiplier, stable, with the right mobo/chipset. That's an OC of 1155mhz up from stock 6.5/100mhz, on air-cooling.

Now I must try running a few games, see if this puppy is stable, not just with the new caps, but the different CPU, the same, but totally different.   Cheesy



Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
justDIY
Microcontroller Madman
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1652


UFOric


WWW
« Reply #25 on: January 22, 2006, 07:13:58 PM »

well, I'm glad you feel you've made a difference in your systems performance.

as for the folks on the badcaps forum, maybe their dimension operates with alternate laws of physics than the world I live in does.
dielectric breakdown is a real problem, it will happen.   I could explain why the engineering team (yes, team, as in a room full of people) designed your motherboard they way they did, but I think it would be a waste of my time to do so.

thats twice now you've asked for input and then disregarded it, quoting the "experts" at the Holy Bad Caps forum.

maybe that's your writing style or maybe you like to just get people riled up.   Know this, I am done giving you any input on anything.

Logged

Want to contact me directly? gmail gordonthree
My Project Blog - http://projects.dimension-x.net

Favorite numbers:
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
rod_vdb
Strong like bull
*
Offline Offline

Posts: 7


« Reply #26 on: January 23, 2006, 05:12:41 AM »

justDIY is talking sense...

You should remember that a larger cap costs more money, especially when you are building 100k boards so they would not have overspeced the components just for kicks.
When the PC starts to become unstable I would look there for your first culprit.
Logged
Brundle Fly
Smart like tractor
***
Offline Offline

Posts: 102



« Reply #27 on: January 23, 2006, 06:21:25 AM »

well, I'm glad you feel you've made a difference in your systems performance.

as for the folks on the badcaps forum, maybe their dimension operates with alternate laws of physics than the world I live in does.
dielectric breakdown is a real problem, it will happen.   I could explain why the engineering team (yes, team, as in a room full of people) designed your motherboard they way they did, but I think it would be a waste of my time to do so.

thats twice now you've asked for input and then disregarded it, quoting the "experts" at the Holy Bad Caps forum.

maybe that's your writing style or maybe you like to just get people riled up.   Know this, I am done giving you any input on anything.



Overclocking does improve the performance of my system, that is not "perceived", it is a fact.

I did take your advice and made sure my soldering gun had a grounded tip before using it.

The team that designed my mobo obviously didn't take the time to test the components used on it, otherwise I wouldn't have bad caps on it in the first place. New mobos come out, then people try to work around all the bugs until a BIOS update fixes things. The team is not perfect.

The "experts" at the Holy Bad Caps forum have a lot of experience changing bad caps. Why you feel you have to mock them is beyond me. I don't try to rile people up, but you certainly seem to be attempting to do just that.

The whole problem with my board was instability due to bad caps. I have replaced those caps, and my board is now stable. If the caps blow, I'll let you know about it. If they don't, I'll just keep my mouth shut.

Thanks for your help in the past, and I guess I won't be looking forward to any help from you in the future. Unfortunate.
Logged

"The secret of success is sincerity. Once you can fake that, you've got it made."

AMD Athlon64 3700+ @2.64ghz 240/11 (stock air, lapped) 1.475V
DFI LanParty Ultra-D PCI-E (Modded to SLI)
eVGA 7600GT CO @610/840
OCZ Platinum EL 2X512 2-3-3-6 1T Dual Channel @220mhz
WD Caviar SE 80G 720
Rob
LED guy
Administrator
Smart like tractor
*****
Offline Offline

Posts: 1970


The constant-current gardener


WWW
« Reply #28 on: January 23, 2006, 10:37:12 AM »

So in general you should make sure that the voltage rating on your caps exceeds the voltage that will be applied to them. I think we can all agree on that.

Looking at your particular situation, BF, I'd be pretty tempted to do what you did. At the end of the day, it's your gamble to take. So I don't exactly look at it like you ignored the advice you got, you evaluated it and assessed that the risk was worth taking (worst case is you pop that cap out and try again with a new one).

Replacing that cap with the out-of-spec part may have any effect from causing the subsystem that uses it to be less reliable (in which case you may not see failure for a long time) to instant dramatic failure. None of us could predict with certainty.

So I look at your choice to proceed with the replacement as a calculated risk. I don't think you can generalize from your experience at all--it's still a bad idea to replace a cap with one that has a lower voltage rating. But you may have gotten away with it here.

Not too many people have soldered on a mobo and come out the other end with a working system, so hats off to you for that in any event.
Logged

tb421ny
Strong like bull
*
Offline Offline

Posts: 3


« Reply #29 on: August 28, 2008, 09:14:22 PM »

You can get motherboard capacitors from either www.thecapking.com or www.badcaps.net.  Soldering instructions can be found at capacitorlab.com.  Those are the sites I use and have been happy with them.  I fix Dell motherboards on the side and have a lot of experience recapping motherboards.
Logged
Pages: 1 [2] Print 
linear1 forums  |  LED discussion  |  Electronics discussion  |  Topic: Where to buy capacitors. « previous next »
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.7 | SMF © 2006-2008, Simple Machines LLC Valid XHTML 1.0! Valid CSS!
Page created in 0.614 seconds with 19 queries.