To blow or not to blow on the joint, that is the question

Question

In a certain Electronic Physical Design book, we're advised that, after soldering:

Do not blow on the joint, as this will cause the solder to cool too rapidly, leading to crystallization and embrittlement.

On the other hand, a (if not the most) famous user of this board, advises us to

blow on the joint gently until it hardens.

Now this sounds like one of those issues that the EEVblog or even Mythbusters would tackle. So, does anyone know of experiments where the effect of blowing on the joint has been studied?

Updates:

• As was pointed out in a comment below, the latter advice may be impractical as it was written because a small joint may harden too fast anyhow for blowing to be helpful in that regard. Still there may be other practical incentives to do it, like cooling the board/parts faster so you can move on to making the next joint without burning yourself (by accidentally touching the traces, parts etc.) So I think it's fair to ask if the advice given in some textbooks (against blowing) is purely ex cathedra or backed up by some empirical evidence. Alas the book I mentioned doesn't cite anything in support of their stance.

• After a bit more searching, I found some anecdotal evidence in an EDN blog supporting the claim from the book. Still it seems rather unsatisfactory and possibly not scientific enough since this blog said that all the joints examined at that site were cold joints ("solder was visibly cracked and crystallized in all different directions"), but that could have happened for other reasons, i.e. this anecdotal piece of evidence lacked a control.

• As discussed in comments below, blowing on the joint is sometimes the poor man's fume extractor (or deflector anyway). Now, since real fume extractors are standard in most shops/labs and these have non-trivial airflow, I suspect some boffin has studied what level of airflow becomes dangerous for joint reliability.

Answer 1

It depends on what and how you're soldering. If you are manually soldering things that come off a board, like a wire, then it's usually good to gently blow on the joint to cool it quickly. The advantages are:

1. It hardens the joint quickly. This reduces the time you have to hold things still, which in turn reduces the chance that things are wiggling around as the solder hardens. When you're holding one or more things with your hands, this is useful.

2. You can get a visual indication of the quality of the joint. Seeing the solder harden shows you a bit about how well the solder made thermal contact, which depends on how well it flowed around all the metal parts of the joint. It's hard to explain without showing it, but sometimes you can catch a problem by noticing that the solder just didn't look right after it cooled, or in the process of cooling.

However, when things are held in place on their own, then you get advantage from letting the solder cool slowly. This is the case, for example, with soldering a single pin of a larger component on a board. If the other pins have already been soldered, then they will hold the part in place. The part joint won't be weakend due to the part wobbling as the solder was cooling. Now letting the mechanical stresses due to uneven heating dissipate a bit by slowing the hardening process helps.

Of course when you're doing reflow soldering, follow the recommended temperature profile. In that case the equipment handles that outright, and you shouldn't be in there altering the process.

With manual hot air soldering, you generally don't want to blow on joints either. You wouldn't use hot air to solder two things you have hold still yourself. Generally you're heating a entire part. Hot air heats the board and other parts around where you are soldering. Give them a chance to cool down slowly and with the least thermal stress.

Answer 2

Better not to blow if you are 100% sure that it won't move until it cools down. Better to blow if there are chances of movement before cooling. Your decision depends on your patience. After all, brittling/crystallizing caused by blowing is better than brittling/crystallizing caused by moving.

Answer 3

The effects of cooling as the last stage of a reflow oven are well understood. In brief, you want it to cool as fast as possible below a limit which I think is thermal shock. As a rule of thumb, 4 degrees per second is about right. I would hazard a guess that for a small to medium size joint in free air it would be about that, but also that blowing on it wouldn't change it much. The flux flies away pretty quickly and the heat is localised. Tldr: blow on heavy things like d2pak pads and connections to high current copper. Do what you like for the rest.

Answer 4

I'd doubt that blowing would make a vast difference in joint characteristics.
As loganf notes, reflow cooling rates are usually in the 2 to 4 C/second rate - more likely closer to the latter. In hand soldering you will have moved on to the next joint long before a 4C/s rate reduces joint temperature to anything like ambient.

Arguably the critical area is from somewhat above to somewhat below the liquid to solid transition point. Traditional 60/40 lead solder is a "eutectic mix" with a very well defined ytransition point. Modern lead free solders tend to be close to eutectic mixes but not xactly so and the transition occurs across a range of temperatures with the mterial being "sludgy" as it transitions. (This is used to good effect for eg "lead-wiping" by cable jointers and others who use molten metl as a seal and hold the metal in the transition zone while "working" it.)

In the case of hand made joints (not roll your owns) I suspect you'd be hard put to limit cooling rates to only 4C/second even in still air. This tends to lead to haarder joints and blowing will increase this effect somewhat. Arguably a more malleable joint is better if mechanical support issues matter.

FWIW (debatable):

The following references largely relate to cooling after wave or reflow soldering but gives some guide to the processes involved.

Wikipedia - reflow

The last zone is a cooling zone to gradually cool the processed board and solidify the solder joints. Proper cooling inhibits excess intermetallic formation or thermal shock to the components. Typical temperatures in the cooling zone range from 30–100 °C (86–212 °F). A fast cooling rate is chosen to create a fine grain structure that is most mechanically sound.[1] Unlike the maximum ramp-up rate, the ramp–down rate is often ignored. It may be that the ramp rate is less critical above certain temperatures, however, the maximum allowable slope for any component should apply whether the component is heating up or cooling down. A cooling rate of 4°C/s is commonly suggested. It is a parameter to consider when analyzing process results.

Wikipedia - solder - search for "cool" - some useful observations

Cree - wave soldering

The last phase of the reflow process is the cooling stage. Proper cooling is vital in the soldering process and enhances the strength of the final solder joint. Fast cooling results in a stronger solder connection but too fast could result in thermal expansion stresses on the components. Cree recommends a cooling rate between 2°C and 4°C per second.

PCB cooling post soldering

It is not desirable for the assembly to cool too slowly (excess dwell at liquidus temperatures) or too quickly (thermal shock resulting). Controlled cooling prevents excess intermetallic formation, de-wetting, oxidation, thermal shock, and other problems

Useful - lead free solder related

Related"

Wikipedia - Wave soldering

http://www.electronics-cooling.com/2006/08/thermal-conductivity-of-solders/

Abstract only

Answer 5

Do blow. Anyway any handwork you do is not professional enough to be airborne, so don't worry. Blow, save time, everything will be fine.