Soldering question

Factual Questions
21

In some of the video segments you can see smoke coming off of the tip of the soldering iron before it touches the components’ leads. In others you can see (much more) smoke coming off only after it touches the leads.

I believe that the person making these devices touches the underside of the soldering iron to a wire solder which has flux and picks up a droplet AND the person applies a liquid flux to the joint. Then the person brings the iron up to the components and “makes” the joint. Sometimes the flux has vaporized off of the iron before it enters the picture, and there’s no smoke until it touches the joint. And, the droplet of solder is not visible - its on the underside of the iron’s tip.

Even for a “quick & dirty” assembly like this, I’d prefer to have the leads run alongside of each other (parallel) wherever there is a joint to be made, even if it means bending one lead into an L. But then again, these are bench assemblies. If a joint fails it can be quickly reinforced and remade.

22

A lot of prototype circuits used to be lashed up in this way, back in the days when things generally ran a lot slower and one could get away with air-wiring a circuit with no ground plane. The late, great Bob Pease was a fan of this technique, and for high-speed and high-precision circuits he’d wire together the components above the copper side of a blank PCB and use the copper foil as a ground plane:

What’s All This Pease Prototype Stuff, Anyhow?

I’ve a similar technique for surface mount components, where I stick the SMD parts to the insulating side of a bit of single-sided PCB, wire them together, and drill little holes whenever a ground connection is required. It’s great for things like simple switched-mode power supplies that are sensitive to component layout and grounding technique. Most people mount their chips in the dead bug fashion (i.e. legs in the air) as it’s easy to tack them down with a blob of glue, but I prefer to put ICs the right way up as I want it to be more representative of a PCB layout.

The master of the art of air-PCBs was Bob Pease’s close friend, the also great, and sadly also late Jim Williams, who did go out of his way to introduce an artistic aspect. Here’s a thermometer, which works by measuring the speed of sound of air in a jam jar (it’s the one with the parrot):

Remembering Jim Williams, 5 years later, with the engineering behind it in the white paper “An Introduction to Acoustic Thermometry”

I’m a big fan of flux, me, and I wish I’d had cottoned on to its awesomeness years before I did. I am grateful to the excellent technician colleague who enlightened me. Lead-free solder is not as easy to solder with as the old, toxic PbSn types, but a bit of flux works magic - just remember clean it off afterwards as it can be corrosive in the long term. Decent lead-free solder helps a lot; I’m a fan of Stannol HS-10 (another top tip from a technician colleague - they’re worth their weight in gold, good technicians). I prefer the old-style rosin solder, which has fallen out of favour with most PCB assemblers due to the need to clean it off afterwards. I hate no-clean flux (which is the preferred type these days) as the faint residue it leaves behind is very difficult to clean off, and it’s also very slightly conductive; not a problem for most circuits, but can cause big issues on sensitive high-impedance/low current designs. The right solvent will shift it though, and I’ve had excellent results with Flux-Off No Clean Plus from Chemtronics, and it works extremely well with rosin fluxes too. Flux-Off do a rosin-specific solvent, but I find the No Clean Plus works for everything (fluxes, sticky labels, gum, ink etc.).

23

It’s also a little conductive and can lead to some leakages between contacts.

24

The trick to avoiding cold solder joints is to heat the part and let the solder flow onto it. This is true regardless of whether you are soldering to a PCB or if you are stringing parts together in a 3d sculpture sort of way or anything else you are doing.

You’re not the only one. I suspect that we have similar techniques.

I have adjustable “hands” (well, more like alligator clips) to hold parts in place as well as a clamp sort of thing with a grippy bit that you turn a little hand wheel to tighten into place (I guess you’d call it a PCB vice?). They do come in handy from time to time.

25

Thanks for all the answers. Adding flux might be the key to mystery. Yes, I use solder with a flux core, but the lack of smoke from the solder iron in most shots indicates that if there is solder there, its flux has been burned off already. The components could be tinned, but that’s not apparent to my eye. If there’s a coating of flux on the component and a hidden dollop of fluxl-less solder on the iron, that may explain what’s going on.

I agree about burning up the ICs. The few times I’ve worked with anything advanced enough to need one, I’ve soldered its socket to the board and just pressed in the IC later. Of course, I’ve never tried to see if I could solder it fast enough to avoid destroying it.