Solder Recommendations for Recapping

[bgavin]

I checked out the Kester site, only to find a bewildering array of products, and subgroups within products.

I'm looking at Kester 44 in 63/37, 8mm/0.031". My rig is the Hakko 936.

Can somebody recommend the optimal solder for motherboard recapping?

[linuxguru]

Re: Solder Recommendations for Recapping

63:37 is useful for unsoldering the old caps and getting the solder out of the through-holes, since it flows at a lower temperature. However, I use regular 60:40 for resoldering new caps - it wets the joint better; the beads tend to be shinier, smoother and more ductile (softer and less brittle); and the bead resistance is likely to be lower because lead conducts better than tin. That's important when the ESR is measured in milliohms.

Also, 63:37 tends to corrode metal tips faster because liquid tin dissolves metals faster than lead. This is noticeable on the Radio Shack bulb-type desoldering tool. But one soldering application where 63:37 is useful is on large power planes, on both PSUs and mobos.

[hardwareguy]

Re: Solder Recommendations for Recapping

I use 60/38/2, 60% tin, 38% lead and 2% silver. I LOVE this alloy, it makes excellent joints, conducts very well and has high mechanical strength.

Radio Shack sells the stuff, look for "High Tech Silver Bearing Solder".

[999999999]

Re: Solder Recommendations for Recapping

Someone is using bad solder if they have a problem with 63/37 eating up tips, not flowing well or not making good joints, it is practically as easy to use as 60/40 but really even easier because there is such a reduced chance of bad joints due to 63/37 being eutectic. The prior description of the difference between 60/40 and 63/37 almost sounded like it was meant to contrast 60/40 with lead-free mixes instead of 63/37. Lead-free mixes do tend to have inferior wetting and so use more aggressive fluxes.

Also, Tin is more conductive than lead, but dozens of years and billions of solder joints have shown the differences here are negligable. If the base properties were to be considered, if anything the critical factor for surface mount parts might be that a general tin/lead alloy tends to be more malleable, somewhat lessening the chances of stress cracks if a board is flexed (not much though, flex any board enough and it's going to be a problem).

The flux is not inherantly different in 63/37, but either metal % solder could have any one of a number of different fluxes in it.

You're not supposed to pick solder in general, rather using what was already there, the existing solder on the part being repaired or the closest match reasonable (I suggest having at least two types, lead-free and 63/37).

Kester 63/37 RMA 0.025" part # 24-6337-9718
Kester Lead-Free 96.5/3/0.5 part # 24-7068-7617

Note that above part # are for 1% flux on the tin/lead (core 50) and 2% flux on the lead-free (core 58). It might be easier to use the 2% core 58 tin/lead and 3% core 66 lead-free due to having more flux in them, though it's also more flux left on the board or that you clean off (with the 285 RMA type). Your choice as to which to use, personally I prefer to use the 1 & 2% part numbers listed but I also enjoy using a bottle of liquid flux for some jobs (typically drag soldering fine pitch SMT, not needed for a typical cap repair though adding extra flux to desoldering braid also helps to clear solder out of holes or around parts you don't want to subject to a spring-action solder sucker since it can kick back and damage some SMT parts (though a bulb or motorized pump, vacuume sucker is gentler too).

You could use good old (more popular) Kester 44 for the leaded parts repairs but the main change is a more active flux and you don't "need" a more active flux for these kinds of electronics work.

There is another issue about getting solder to flow though, the typical 25-35W cheap soldering irons like you'd find at Radio Shack have terrible recovery times and just don't heat up larger power and ground plane areas very well so they can tend to make any soldering job more difficult, especially lead-free work.

[999999999]

Re: Solder Recommendations for Recapping

There's also the issue of lead-free (and/or leaded that's silver bearing, unless enough copper is added) solders typically having a higher melting point so there is a greater chance of overheating a part or trace, but if that's the pre-existing solder on the part, you want it all to reflow anyway so you just need to exercise a little more care.

[acstech]

Re: Solder Recommendations for Recapping

I've been using some stuff I got at Radio Shack years ago. It's their regular "Silver Bearing Solder." 62% tin, 36% lead, 2% silver. .022 thickness. Absolutely love the stuff. Melts easily, and easy to work with. I have noticed it's either fully melted and fluid, or fully solid. There is no in between with it.

[severach]

Re: Solder Recommendations for Recapping

All the new solders are quite confusing and I found the information here to be too incomplete to be helpful. Here are some links that will straighten things out.

http://www.logwell.com/tech/servtips/solder.html
http://www.kester.com/en-us/leadfree/lfsfaq_index.aspx (Lead Free Alloy Selection)
http://www.rfcafe.com/references/electrical/solder.htm

Sn95Cu5 (best guess) is available at Lowes.
Sn96Ag4 is available at Sears and Radio Shack.
Sn95Ag5 is available at Sears
SnAgCuBi is available as Oatey Safe-Flo solid plumbing solder.

I have Sn95Cu5 and SnAgCuBi and neither melt or mix properly with SAC motherboard solder. I decided it wasn't worth buying SnAg solders only to discover that they too won't melt or mix properly. My spool of Kester Sn96.5Ag03Cu.5 arrived today and it melts and mixes properly.

[linuxguru]

Re: Solder Recommendations for Recapping

Sn62Pb36Ag2 is also eutetic or very close to it, and it's as good or better than stock 60:40 or 63:37 for soldering, for several reasons:

1. The Ag retards leaching/dissolution of barrier/coating metals like Ni, Ag, Pd, Cu, etc. into the liquid solder - pads and tips will last longer.

2. The melting point of SnPbAg eutetic alloy is 179c, slightly lower than SnPb 63:37 eutetic at 183 and SnPb 60:40 at 186.

3. Intermetalic compounds at the boundary between the bead/lead or bead/pad are stronger at room temperature than the bulk solder bead, for a wide range of barrier/coating materials, leading to a more reliable joint.

4. AgCu intermetalic is highly conductive, reducing the probability of a cold or non-ohmic joint.

In short, SnPbAg is an excellent solder - no worse than stock SnPb in most parameters, and better in a few. In practice, bulk Pb is usually contaminated with a small amount of Ag in natural ores, so most SnPb solders will have trace quantities of Ag in any event. This will generally improve the properties of the solder up to about 3.5% Ag - higher percentages of Ag will lead to higher melting points of the alloy.

[linuxguru]

Re: Solder Recommendations for Recapping

(Addendum: 10-min edit limit expired)

5. Future rework of SnPbAg soldered boards with SnCu, SnAgCu or SnAgBiCu solders (or vice versa) is still possible, with excellent reliability - check out Table 3 on Page 7 of publications.npl.co.uk/npl_web/pdf/matc85.pdf

[999999999]

Re: Solder Recommendations for Recapping

Ag does not retard leaching or dissolution of board or capacitor leads to any significant extent unless they were silver plated, which they are not. People who solder with silver bearing solder do report their iron tips worn away faster. While these reports can't be taken as strict evidence due to too many potential variables, it does still suggest what had been suspected all along, that the somewhat more active flux is to blame.

The mentioned paper is contrasting the mixing of alloys against two specific tests applicable to surface mount parts, which should never be factors for our (recapping) purposes. Before a board could flex enough to effect these capacitors, much of the rest (other original SMT parts) would already be damaged or their joints cracked. The extreme thermal cycling needed to make a difference just won't be happening with capacitors, else the cap itself would fail too.

In the end, I feel the hypothetical advantages of some silver are outweighed by it not flowing as well. In practice either type does the job fine as evidenced by billions of solder joints, but one is less expensive and slightly easier to use, making no benefit at all to silver in solder for cap rework as we'd realize in our uses.

[linuxguru]

Re: Solder Recommendations for Recapping

Good points - perhaps the best advice is to try both and use whichever is easier to work with. I'd probably go with SnPbAg for audio.

[Fizzycapola]

Re: Solder Recommendations for Recapping

what linuxguru said

or just find a hi-fi spares shop most expensive solder or simply best; [surly mode on] if what they give you says materials you've never heard of like lithium-23 / uranium-239 / chlorine-13 / mercuric-7000 then you probably in wrong shop try another.

[linuxguru]

Re: Solder Recommendations for Recapping

There's also SnAu 98:2 solder, but I'm not sure I'm willing to pay for it.

[Fizzycapola]

Re: Solder Recommendations for Recapping

Sorry for my last post (was just jogging memory it gets *really* slow sometimes) finally found the one I was thinking of:

Mundorf 9.5% silver solder supreme ~$40 per 100g
88.6%S-Sn, 1.8%Cu, 9.5%Ag, 0.1%Au

Never tried it, it's too expensive for me.

General hype for gold in solder:
"Gold-tin solder alloys, ranging from the eutectic 80/20 Au/Sn composition to 70/30, are of proven utility for integrated circuit chip or die bonding, because of their high bond strength, adhesion, resistance to creep, low stress, corrosion resistance, good thermal conductivity and absence of whisker formation." - Jet Process Corp (Website)

Retailers I know of:
hificollective.co.uk

[999999999]

Re: Solder Recommendations for Recapping

I suggest that audio dealers frequently peddle snake-oil. In a rare case, a large brige of certain alloys might have some minor difference, but in audio there is a focus on preserving AC signal purity, while in DC supply scenarios, the focus is not in preservation but rather ripple. We don't need to mind whether a noise is linearly preserved, only that it is minimized as much as reasonably possible.

Who has reported problems using plain old boring 63/37, or 60/40 if they keep the work stable while it hardens? IMO, there is no gain at all using exotic solders, the typical Kester/etc blends are great for our uses, and better than lead-free at retarding or eliminating tin wiskers.

This does not mean other solders can't do the job. Certainly many can, and do, but to pay more without any demonstrated benefit seems counter-productive?

[bgavin]

Re: Solder Recommendations for Recapping

Who has reported problems using plain old boring 63/37, or 60/40 if they keep the work stable while it hardens?

I am not a component level tech by trade. I've been using 63/37 on my board recaps (6 so far) and am delighted with the results. None of my recap jobs have failed so far.

The best value in tips for me is Willawakes' sticky on using a proper solder station and recapping technique. The stainless needle method works perfectly for me. I use a modest Hakko 936 solder station and love it. It was under $75 USD from an internet vendor.

[filterpunk]

Re: Solder Recommendations for Recapping

I hate to dredge up an old thread, but I've referred back to this one often enough that it seems like the most appropriate place to ask.

I'll be attempting my first recap this weekend. Due to budget constraints, I'll be using a cheap 40w soldering iron from Radio Shack. However, I'm not really sure what type of solder would work best, given the potential limitations of this iron.

From what I've read here and elsewhere, it seems like 63/37 might be easier to work with, as it melts at a lower temp. than 60/40 and may help minimize poor solder joints. Related to that, should newbies be even bothering with SnPbAg or stick with the tried-and-true SnPb?

I've read the discussion here, but it's difficult to get a sense of what I should really be using as a first-timer.

[999999999]

Re: Solder Recommendations for Recapping

I hate to dredge up an old thread, but I've referred back to this one often enough that it seems like the most appropriate place to ask.

I'll be attempting my first recap this weekend. Due to budget constraints, I'll be using a cheap 40w soldering iron from Radio Shack. However, I'm not really sure what type of solder would work best, given the potential limitations of this iron.

From what I've read here and elsewhere, it seems like 63/37 might be easier to work with, as it melts at a lower temp. than 60/40 and may help minimize poor solder joints. Related to that, should newbies be even bothering with SnPbAg or stick with the tried-and-true SnPb?

I've read the discussion here, but it's difficult to get a sense of what I should really be using as a first-timer.

How much soldering experience do you have? Ideally one would practice on some junk piece of electronics first and this would require having solder already, so do you have some?

Ideals and reality are a bit apart, practically speaking you could use 63/37, 60/40, or one containing silver. 40W is enough to do the job even if the silver bearing solder melts at higher temp, though with a Radio Shack iron that heat may not be transferred as effectively to the board.

The generic answer is use what you have if you have one of these, or if buying now 63/37 would be slightly easier to use but 60/40 similar enough. The ideal is using the same solder that's already on the board but unless you know what that is, we can skip to practial applications in which many of us have used all these with good results on both older leaded solder boards and newer lead-free boards. Just because a board is lead-free doesn't mean you could just grab some lead-free solder and assume it's a closer match, so the common and inexpensive 63/37 seems the best option for you.

To minimize poor solder joints there are a couple other things that can be as important. Regularly clean the iron tip (especially on a Radio Shack iron, some of those are so terrible that in past years cleaning wasn't enough, they had to have a new surface filed onto them because they just wear so fast compared to a good iron plated tip found in better irons).

Get some liquid flux if reasonably possible, often people who haven't soldered much will end up adding too much solder just trying to get enough flux onto a joint to make the (lesser amount of) solder they needed, flow properly. I can't say how well matched your iron is for the job but eventually if you do enough soldering you may want a lower wattage iron that stays at a lower temp but with faster recovery time, because the lower temp tip (among other benefits) burns up the flux in the solder slower allowing it to be more effective.

Nothing really beats practice, beyond the basic fundamentals of soldering you just need to do a few boards to get proficient at it. I had an old Radio Shack iron once which had a threaded tip that screwed into the handle, the heat transfer between handle and tip was terrible because the heat degraded and accelerated oxidation of the threaded areas. It would be bad to hold such an iron on an area for a long time trying to transfer more heat than it's capable of if in bad condition so I suppose my point is, if it's not going fairly easily stop and reassess what part is going wrong.

Since the Radio Shack iron is a fixed temp or at least not likely to be temp feedback controlled, remember to leave several seconds between doing each joint so the tip has time to reheat.
Maybe I'm emphasizing the potential problems with Radio Shack irons too much, but there are these little things that make most of us want to move on to a better iron eventually (though I could hardly recommend buying a replacement just to solder one motherboard).

[filterpunk]

Re: Solder Recommendations for Recapping

This will be my first experience with soldering, ever. Luckily, I do have plenty of old parts at my disposal, some of which were previous victims of bad caps. I'll be buying the tools needed this weekend, but am doing some last minute research before diving into it.

I can't be 100% sure, but I'm fairly certain that the board I'll be working on is not RoHS compliant. Could be that E-Mu simply left that info out, or that they released an RoHS version strictly for Europe. I'd try asking them, but frankly, they've been unresponsive to any of my attempts to contact them. Thanks for buying them out, Creative Labs!

Anyway, I think I'll probably start with the 63/37 and Radio Shack iron, see how it handles and how I do during practice, then assess from there whether or not I should try some liquid flux. I may also look around locally for a better electronics store, see if there are maybe some decent options outside of Radio Shack.

Thanks very much for the help.