Low temperature solder wire

[michaeliv]

Hello,

I can easily find Sn42Bi58 solder paste that melts around 140C but no solder wire.
I found THIS but reviews indicate that the melting temperature is same as regular solder.
What is the lowest temperature solder wire that is generally available(it would be good if it's available on AliExpress / eBay).

Thanks!

[uncle_bob]

Hi

Do you mind shipping it in refrigerated containers?

The problem with lower temperature wire is that it's mighty soft. It also does not have a flux core. You can get stuff in wire form that melts at just above room temperature. The only way to get anything to you other than a blob is to ship it cold.

Do you mind the fact that it does not work?

Not all "solder" wets to all surfaces. Not all flux works with all solder. When you get into the low temperature stuff, fun and exciting things like manual cleaning of surfaces come into play ..... uggg  Next up there are issues with plating and dissolving into the solder.

Do you have a really big limit on your Visa card?

You quickly move into the "specialty metals" world with this stuff. Neat thing like lot charges, minimum orders, and 16 week lead times. That's all fine if you are in business using this stuff. It's utter nonsense for an individual. 

Do you ever run a part near it's rated temperature?

Solder does not suddenly go from liquid to solid. It spends a lot of time in an intermediate "plastic" stage. It's pretty weak in this form. It also oxidizes faster. If you go to a lower temp solder, be very careful about hot parts.

For normal electronics use, good old tin/lead is the lowest temperature stuff you should be using. It can be trusted to work with the plating on your parts. You can get soldering gear to melt it. It holds up well when parts get hot. Get the parts to hot and yes the tin lead will melt. You can get it with flux cores in a variety of formulas.

What are you trying to do with the lower temperature stuff?

The only time I've used the exotic solders was when attaching to some very unusual materials. Running a fancy cleaning system was not a big deal. After all the hassles ... not going back to that stuff.

Bob

[michaeliv]

Well I would like stuff that is "generally available" so that would exclude anything that
 - Ships refrigerated
 - Has poor enough solder-ability to not be at least somewhat popular
 - Is obscenely expensive
 
Obviously I wouldn't use it for anything that runs 50C+.
Ideally I was looking for the wire equivalent for Sn42Bi58. I'm guessing that this doesn't exist in wire equivalent because it's brittle and the wire would shatter ? Maybe someone knows more ?
The Sn42Bi58 paste is not significantly different than regular leaded solder paste in either performance, usability or price. The only problem I've faced so far when using it is the lower strength.
There are lots of uses for low temperature solder such as soldering temperature sensitive components such as LEDs/VREFs or soldering on low quality PCBs which easily get heat damaged.
My thinking was that since there is an AliExpress listing of a fake version of "low temperature solder wire" there must exist a non-fake version of it, but I'm having difficulty finding it.
Are you saying that there is simply no lower temperature solder wire with decent performance that would be commercially available ?

[Ian.M]

It *may* make sense for soldering really sensitive precision voltage references, but if you hand solder one pin at a time and allow the chip to cool to ambient after each joint, drift due to thermal stresses even using ordinary 60/40 solder can be minimised.   

All except the most exotic LEDs are rated for normal reflow processes, so there is no real need for it there.

The cost of good low temperature solders is high enough that it just isn't worth using them on low quality boards. Its more cost-effective to spec a better board that can stand up to normal soldering.

OTOH for some sensors and exotic substrates it may be the only game in town.

There are other problems like fluxes not being active enough at 140 deg C and long-term reliability issues.  You should read: http://www.hpl.hp.com/hpjournal/96aug/aug96a10.pdf

[uncle_bob]

Hi

There are an enormous number of things you get into trying to use low temperature alloys for normal soldering. For something like a voltage reference, thermocouple effects would be pretty high on the list. For general stuff, fatigue resistance is a big one. You made a soft joint and it will not last as long as one made of the "right stuff".

Unless you are dealing with Germanium based semiconductors, normal soldering practices with proper equipment are not going to damage anything. If you are having over heat issues. This is even more true in our modern era of lead free / high temperature solders. If you use bad old 63/37 lead solder, you already *are* using low temperature stuff.

Take a look at your soldering equipment. If you are working on big ground plane boards without pre-heat ... fix the problem. If your soldering iron has a blown tip ... replace it. If you are trying to do this with an iron that is not correctly temperature controlled, spend $100 or so and get one that is.

Take a look at your technique. Often things get blamed on equipment that are really a matter of technique. Keeping tips clean. Choice of tops. Pre-fluxing or not. Surface prep before soldering. Even things like lighting and magnification matter a lot. Don't take this as a knock. It is not at all uncommon to take somebody who has been soldering for a few decades, send them to soldering school, and they have problems. Don't ask me how I know this ...

=====

Solder wire used with a flux bottle is a pain. You wind up with an enormous gooey mess all over the board. Back in the days we had a tank full of a now banned substance to spray them down, that was not a big deal. Today you don't have neat gear like that. Leaving flux all over the board is a *bad* idea. Cleanup with modern chemicals is no fun at all. You don't want to make it worse.

I do not know of any flux core low temperature solders.

Bob

[tooki]

Do you ever run a part near it's rated temperature?

Solder does not suddenly go from liquid to solid. It spends a lot of time in an intermediate "plastic" stage. It's pretty weak in this form. It also oxidizes faster. If you go to a lower temp solder, be very careful about hot parts.

Solder does go directly from liquid to solid and back if it's eutectic solder. 63/37 solder is eutectic and has no plastic phase. (That's why we prefer it for hand soldering, because the lack of plastic phase means fewer cold joints caused by motion during a plastic phase.) SAC359 lead-free (Sn95.6Ag3.5Cu0.9) is also truly eutectic, but it seems impossible to purchase.

There are lots of uses for low temperature solder such as soldering temperature sensitive components such as LEDs/VREFs or soldering on low quality PCBs which easily get heat damaged.

Modern surface-mount components are literally designed to be baked in a reflow oven at temperatures slightly above the melting point of lead-free solder. I suspect you're seeing a problem where none exists. No, you cannot bake some through-hole components, but their leads are certainly designed to take ordinary soldering temperatures. And if you're damaging cheap PCBs, then either switch to better boards, or use some combination of better flux, lower temperature, reduced pressure on the pads, and shorter dwell times when soldering them.

[uncle_bob]

Hi

Not to overly quibble, but in the context of "every solder alloy on the planet" you do see a lot of them go through a plastic stage In this case I guess I was not very clear. My reference was to plastic in a more general sense.

Back to the point, even with 63/37, the joint is *much* weaker at say 175 than an 125. It's even weaker still at 200C. The solder has not yet turned to a liquid, but it has gotten pretty soft. This is one of the reasons for specifying a cool down profile even when using 63/37. You need some margin between melting point and operating temperature to ensure a solid mechanical situation.

Lots of variables. Stick stick with the stuff that has "well known properties". Heading off to Bismuth / Gallium / Indium combos is not a good idea.

Bob

[Gyro]

There was a fun demo last night on last night's QI program (BBC) of melting a teaspoon made of Gallium in a glass of hot water. 

[tooki]

Hi

Not to overly quibble, but in the context of "every solder alloy on the planet" you do see a lot of them go through a plastic stage In this case I guess I was not very clear. My reference was to plastic in a more general sense.

Back to the point, even with 63/37, the joint is *much* weaker at say 175 than an 125. It's even weaker still at 200C. The solder has not yet turned to a liquid, but it has gotten pretty soft. This is one of the reasons for specifying a cool down profile even when using 63/37. You need some margin between melting point and operating temperature to ensure a solid mechanical situation.

Yes, most alloys are non-eutectic, but in electronics, eutectic solder is (or at least was before RoHS) a really significant portion of the solders sold. It's a very significant error to omit them.

Yes, eutectic solder's strength and elasticity changes with temperature, but its melting point is just that — a singular point. If you're going to use the technical term, it's best to stick with its established meaning. The plastic phase of a non-eutectic solder has absolutely nothing to do with the weakening you're talking about. At 182C, 63/37 solder is a solid. Not a paste, or a gel, or whatever you want to describe a plastic phase as. A solid. One degree higher, at 183C, it's a liquid. Not a paste or gel, a liquid. (At 200C, 63/37 has been liquid for 17 degrees already and thus is indeed "weaker still".)

Indeed, mechanical strength is part of the reason for a cooldown profile, but I thought the main reason was to reduce mechanical stress because of different temperature coefficients of shrinking.

[uncle_bob]

There was a fun demo last night on last night's QI program (BBC) of melting a teaspoon made of Gallium in a glass of hot water. 

Hi

Yup, melts in a warm room.

There is no lack of nutty alloys to play with out there. Look at all the trouble we still have finding a good lead free solder. There are LOTS of people trying this and that. It's an issue on plumbing. It's a big deal in electronics. Finding something that works as well as / as quick as/ as easy as lead solder .... not so easy. I'd bet that they have put billions of dollars into researching the stuff by now.

Bob

[michaeliv]

Here is what I've managed to find so far. If anyone knows of better options please share.

http://www.ebay.com/itm/141643380681 130C $16 / 18g With flux core -- Contains cadmium (toxic)
http://www.ebay.com/itm/301863621504 118C $20 / 10g  No flux core
http://www.ebay.com/itm/201401318719 138C $5 / 3g No flux core

[uncle_bob]

Here is what I've managed to find so far. If anyone knows of better options please share.

http://www.ebay.com/itm/141643380681 130C $16 / 18g With flux core -- Contains cadmium (toxic)
http://www.ebay.com/itm/301863621504 118C $20 / 10g  No flux core
http://www.ebay.com/itm/201401318719 138C $5 / 3g No flux core

Hi

They all are mighty proud of their solder aren't they?? Looking at how little you get, that's very expensive solder.

I've used the one in the middle of your list or something like it. I would avoid it for electronic use.

Bob

[KL27x]

Years ago, you used to be able to buy 43/43 solder with 14% bismuth. I read that it was used in the telecom industry for awhile, before they dropped it due to long-term structural failure. Even 14% bismuth makes the wire fairly brittle. While it doesn't lower the melting temp all that much, it keeps the solder joint liquid for way longer. And even after it solidifies, you can pull the component/bead clean off the pad for a good while. Kinda what uncle bob described.

There's apparently a problem with long term strength due to creep, which creates low tensile strength shear lines if I am reading correctly.
http://www.emeraldinsight.com/doi/abs/10.1108/eb037766

The roll I have is Kester 43/43/14 with rosin flux core. It was an MPJA closeout, maybe 5-6 years ago. I use it like ChipQuik.

that's very expensive solder.

The closeout price on this roll was less than a roll of regular solder! (Given its overall utility, that is about right).

[uncle_bob]

Why you need one? Hand soldering with iron, if done properly, even delicate GaAs RF devices with SAC (217C!) solder will be fine.
The reason people develop low temp solder is because thermal damage in reflow or wave soldering process is more severe, and some parts won't survive 260C reflow.

Hi

Actually, after running a few 10's of millions of circuits through hand soldering and the same stuff through reflow, I would say the reflow is by far the less damaging process. Yes, they all used pretty conventional parts on them. Indeed there are some parts that do not like high temperature. With the world having gone to ROHS more than a decade ago, those parts are becoming pretty scarce. It is very unlikely you will run into a part like that at home. It's even less likely you will be set up with the ESD system to keep from blowing it out.

Bob

[woodchips]

Now this thread has turned up at a very opportune moment, I need some 140C low temperature solder.

Why? I have a boiler that uses a thermal link using this BiSn solder to stop any thermal runaway. The suppliers of the links don't seem to understand that the ones they sell don't work, they are smaller so the I^2R losses, at 40A, cause them to fail after a short period. I need to be able to resolder them together and save the silly replacement cost.

Anyone use anyone in the UK? I have found Somerset Solders who do a 250g jar for £29, also 0.8mm and 1.2mm solid wire for very much more.

The article in the HP magazine referred to is interesting, particularly the problems with lead contamination and the effects it has. Could use a new soldering iron tip, but has anyone uses a small butane flame to solder it?

Thanks to the various posters.