Soldering directly to pure silver?

[LLCoolBeans]

Can a copper wire be soldered directly to a pure silver heat sink? Does one need to do anything special to prep the surface or use a specific type of solder?

Thanks

 

[JCup]

Can a copper wire be soldered directly to a pure silver heat sink? Does one need to do anything special to prep the surface or use a specific type of solder?

Probably possible, although the thermal conductivity of silver is such that you will likely need a lot of heat supplied. Try heating the sink with a hotplate or in an oven to get the temp of the metal up.

Silver has a pretty high solubility in tin based solders, and I've soldered silver plated copper, but not pure silver...

Suggest you put put the heatsink in an oven at about 350 F or so, remove it and place on an insulator, work quickly - the heat required to melt solder can be supplied more easily in this manner.

Heat in the oven, remove an apply a type RA liquid flux to the spot for the wire solder joint, then place the wire, heating the wire with your iron, and run flux core wire solder on the wire and the nearby metal of the sink.

 

[LLCoolBeans]

Probably possible, although the thermal conductivity of silver is such that you will likely need a lot of heat supplied. Try heating the sink with a hotplate or in an oven to get the temp of the metal up.

Silver has a pretty high solubility in tin based solders, and I've soldered silver plated copper, but not pure silver...

Suggest you put put the heatsink in an oven at about 350 F or so, remove it and place on an insulator, work quickly - the heat required to melt solder can be supplied more easily in this manner.

Heat in the oven, remove an apply a type RA liquid flux to the spot for the wire solder joint, then place the wire, heating the wire with your iron, and run flux core wire solder on the wire and the nearby metal of the sink.

I guess I should elaborate, the item in question is not a complete heat sink, merely a riser/shim in between a Cree MC-E and an aluminum heat sink. Total mass of the riser will be pretty small as it will only be about .045" thick and about the length and with of the emitter itself. I don't anticipate having trouble getting it hot enough with my soldering iron. Of course, I'm new to all of this so let me know if you think I'll have difficulty with a part that small.

So, thermal issues aside, I should be able to use regular old lead/tin rosin core solder to make a reliable connection between a lead and a piece of pure silver. Is that correct?

Thanks again.

 

[JCup]

I guess I should elaborate, the item in question

So, thermal issues aside, I should be able to use regular old lead/tin rosin core solder to make a reliable connection between a lead and a piece of pure silver. Is that correct?

I believe so. Silver is solderable (although in my experience it does not wet nicely), and many connectors have silver plated surfaces, in the 80's silver plated leadframes were common for integrated circuits. It is one of the best metals for both heat and electrical conduction.

I would recommend eutectic tin/lead Sn63 metal. You will need a flux to reduce the oxide present on silver, rosin should work, but the organic acid type may give you better results.

 

[LLCoolBeans]

I believe so. Silver is solderable (although in my experience it does not wet nicely), and many connectors have silver plated surfaces, in the 80's silver plated leadframes were common for integrated circuits. It is one of the best metals for both heat and electrical conduction.

I would recommend eutectic tin/lead Sn63 metal. You will need a flux to reduce the oxide present on silver, rosin should work, but the organic acid type may give you better results.

Thanks, I'll give it a go with the rosin core, as that is what I have, If I have trouble, I'll go to the store tomorrow and grab some flux.

 

[qwertyydude]

You'll need a lot of heat best way to solder to silver is use a jewelers torch, I think you can get one that goes to regular oxyacetylene hookups at harbor freight. Pinpoint HOT flame right to the joint and you can solder or braze no problem.

 

[LLCoolBeans]

I used my soldering iron on high and Sn63 to solder a 24ga stranded wire to a test piece. No problem, joint is clean and strong.

Silver riser is approx: .5x.25x.048".

 

[JCup]

Cool, LL! Good work, thanks for the report. I'll bet that will be one effective little heat sink. Just a question...

I'm working on an opportunity to market solder.
Where did you get your solder?
What details - wire diameter, core, flux type, spool size...

 

[LLCoolBeans]

Here are the specifics on the solder I used.

Brand: Multicore
Metal: Sn63
Type: X38B
Dia: .015" (.37mm)

1/2lb. spool

Got it at the local electronics shop last week.

 

[hank]

I've done this a lot doing sculpture years ago.
Three notes, maybe/maybe not useful:
1) "adequate ventilation" means having a big honking exhaust fan running, on the far side of the work so the airflow is over your shoulders, across the work, and away from you. Lead boils off before silver gets hot enough to solder; so does zinc; neither is good to have in your air or deposited as dust around your worksite. Once you've had metal fume fever, you'll understand. It's kind of like learning to do backups of data ....
2) silver-tin solder needs a higher temperature ("hard soldering" or "silver brazing" needs the metal to get up to dim red) but also makes a much stronger bond that doesn't corrode, and removes the problem of putting lead into your area. This has a significant amount of silver in it, comes in three different temperature ranges so you can build up something out of several different pieces by using them in sequence. (This isn't "silver-bearing" solder, though that might also be useful; that's a much lower temperature range material, only slightly above the old lead solder.)
3) Have a little electric teakettle going and plunge the work into the boiling water as soon as the solder sets, and the residue will explode off the metal instead of gluing itself down as it cools.

 

[LLCoolBeans]

Photos of the silver riser and the emitter I plan to solder to it.

Riser will act as the jumper connection for 2S2P configuration and transfer heat to the aluminum heat sink.

 

[JCup]

I've done this a lot doing sculpture years ago.
Three notes, maybe/maybe not useful:
1) "adequate ventilation" means having a big honking exhaust fan running, on the far side of the work so the airflow is over your shoulders, across the work, and away from you. Lead boils off before silver gets hot enough to solder; so does zinc; neither is good to have in your air or deposited as dust around your worksite. Once you've had metal fume fever, you'll understand. It's kind of like learning to do backups of data ....
2) silver-tin solder needs a higher temperature ("hard soldering" or "silver brazing" needs the metal to get up to dim red) but also makes a much stronger bond that doesn't corrode, and removes the problem of putting lead into your area. This has a significant amount of silver in it, comes in three different temperature ranges so you can build up something out of several different pieces by using them in sequence. (This isn't "silver-bearing" solder, though that might also be useful; that's a much lower temperature range material, only slightly above the old lead solder.)
3) Have a little electric teakettle going and plunge the work into the boiling water as soon as the solder sets, and the residue will explode off the metal instead of gluing itself down as it cools.

Tin based solders will solder to silver without reaching brazing temperatures, refer to a phase diagram to see the temperature, but certainly you don't need glow temps.

http://www.ami.ac.uk/courses/topics/0244_tsm/index.html

Out in Berkley, you may want to avoid all lead. My recommendation is that the tin-lead solder alloys are the best and most reliable, notwithstanding RoHS intiatives and other green and fuzzy concepts. If I were soldering something to use for food or drink, then yes, avoid the lead.

Especially for soldering copper or copper alloy base metals, tin-lead is superior, and quite safe for the environment.

 

[JCup]

I see what you have in mind, LL. If you have not done the soldering, apply some thermal adhesive too. Looks like you will have a fine sink.

 

[mudman cj]

Yeah, you can't beat pure silver (OK, pure diamond :ironic for thermal conductivity. The only thing I will add is to prepare the surfaces of your silver so that they are mirror finish and flat. This will decrease the average distance the heat has to travel through the thermal grease, which is of course very poor compared to silver.

 

[hank]

Ah, you don't want to solder an _emitter_ to silver using high temps for sure. It'd cook.

As to avoiding lead and zinc fumes and dust -- this isn't news, nor a political statement. I'd've given you the same advice when I lived in the deep south. It's textbook.

Lead's slow [edit-- sometimes it's fast*] and accumulates; zinc's very fast and you get over it quickly. Avoid both.
"Adequate ventilation" is defined so it can be said in two words on product labels.
Many amateurs never looked the words up. That's it: forced crossflow exhaust airflow ventilation.

Like I said, it's like advice to back up your data. After you didn't, you will thereafter.
http://www.google.com/search?q=zinc+fume+fever

*http://www.anvilfire.com/iForge/tutor/safety3/index.htm

None of this is going to hurt you doing one little job once.
Starting good practice doesn't hurt either. That, too, is cumulative over time.

Anyone new to soldering should read and understand so they can judge for themselves.
Yes, all us old guys survived not taking those precautions. We could be smarter now.

 

[Oznog]

Well it's interesting, but I don't think it's going to help with thermal conductivity in any significant way over copper. In fact, I suspect a wider piece of sufficiently thick copper will outperform this fairly narrow bit of silver. Or a copper sink.

 

[LLCoolBeans]

Well it's interesting, but I don't think it's going to help with thermal conductivity in any significant way over copper. In fact, I suspect a wider piece of sufficiently thick copper will outperform this fairly narrow bit of silver. Or a copper sink.

Not looking for any sort of ultimate performance or anything. Just that the silver was not as soft as the copper that I have, so it was easier to machine a thin piece without bending it while cutting on the mill.

The direct solder jumper connection idea has been scraped. Getting the silver hot enough to solder would almost certainly cook the emitter.

 

[JCup]

Ah, you don't want to solder an _emitter_ to silver using high temps for sure. It'd cook.

As to avoiding lead and zinc fumes and dust -- this isn't news, nor a political statement. I'd've given you the same advice when I lived in the deep south. It's textbook.

Lead's slow [edit-- sometimes it's fast*] and accumulates; zinc's very fast and you get over it quickly. Avoid both.

Who said anything about zinc? I have been a process engineer in the electronics assembly business since 1978, and some consider me expert.

At soldering temperatures you can generate with an iron, no metallic gases are present. You suggested brazing, which is inappropriate for joining electronics components.

See:

"process known as soldering is generally similar to brazing except that the filler metals used melt attemperatures below 427C (800F). In actual practice, most brazing alloys melt at temperatures well above 427C, most solders at temperatures well below 427C. Many of the brazing alloys based on silver (all ofwhich melt above 600C) were formerly termed "silver solders". Avoid that term, and its relative, "silversoldering". Even the term "silver brazing" is sometimes misleading, since some brazing applications for which silver alloys are generally used can also be handled with alloys which contain no silver."

As to the safety of soldering with tin-lead, I've had extensive experience with actual personnel monitoring, blood tests, and read many articles. There is no exposure to lead in the "fumes" from typical soldering. Lead does not vaporize below 3,000 F.

http://www.esabna.com/euweb/oxy_handbook/589oxy19_1.htm

I don't particularly enjoy online debating, but in this case wanted to point out you were inaccurately stating how tin based solder could be used to join a copper wire to a silver substrate in an electronic assembly.

Regarding the safety of tin-lead solder alloys, I will not continue this beyond this post, but do assert that these alloys are less expensive and superior in practical usage to the "lead free" alloys and are safe and legal to use in the US for both private and commercial purposes so long as they are not exported.

Ingestion is the main risk for handling lead bearing alloys. Never solder in an area that will be used for food handling and preparation, and wash hands prior to eating or smoking. Lead and its oxides cannot be absorbed through the skin.

Welding and brazing temperatures can readily, on the other hand, produce toxic exposure to metals via vaporization. Use of an oxy-acetylene torch as you suggested would bring about a whole new realm of safety issues which you did not address other than to suggest the avoidance of tin-lead solder alloys.

 

[hank]

No argument there, you're right, no professional would imagine doing this that wrong. I know heat sinks can take a lot of heat before you get something properly soldered to them (I was guessing soldering a wire connection -- the photo came in after my first post).

I've seen people try to solder to a hunk of metal, get bored trying to heat it with an iron, pull out a propane torch (flame about 1995 °C/3623 °F), get bored trying to heat the metal slowly, hit the solder with the flame, and oops, "Lead does not vaporize below 3,000 F."

Nobody I've ever seen make that kind of mistake had a clue what they were doing or that the torch flame was that hot. Word to the wise, Murphy's law, ventilation is good.

 

[Oznog]

The direct solder jumper connection idea has been scraped. Getting the silver hot enough to solder would almost certainly cook the emitter.

But you can still use standard tin/lead electrical solder, right? These emitters were MADE to be soldered and take the temps of electrical solder nicely. It works brilliantly. The high silver content silver solders will be trouble, regardless of whether you spread the thing out to bridge the electrical contacts or not.

And while Arctic Silver is "good" in most cases, solder should be better. I've come to the conclusion that a big issue with AS is the thermal resistance is so dependent on application method and varies quite a bit. In most applications it probably wouldn't make much difference, but since this whole thing seems to be an exercise in "how low can it possibly be", then that's quite significant.

You could always make the whole sink out of copper... that rocks. Well, would have incredibly low thermal resistance anyways. But at some point you're gonna have the situation where the limiting factor is the thermal resistance from sink-to-air and/or the junction-to-pad resistance inside the package and there's nothing significant to be gained by trying to improve the pad-to-sink conductivity.