Making a dry joint

Clip something that shakes the lead as you solder it. A vibrating transducer from an old mobile phone. Use a small low power soldering iron, and a cold fan blowing over the assembly.

That will hopefully bugger it up.

If not, use lead free solder.

A smear of silicone grease should do it.

Cheers

In message , whisky-dave writes

Clear nail varnish?

dab of solder mask?

Cheers, T i m

I can see some great ideas from other people, but I had a question. Having found the fault, are the students expected to remake the joint correctly? If so, that could be quite difficult with some of the suggestions.

Solder resist.

leads so they can't be soldered too.

ost things can be faked, I can blow up LEDs, transitors, etc or put the wr ong value component in. But I want to create a dry joint the standard coppe r stripboards is a bit more of a problem, it can't be too messy or obvious as I want them to fault find using a meter on continuity.

No they are not meant to repair anything. I give them simple circuits with faults. A blown LED wrong resistor value, capacitor round the wrong way. Shorted track, broken track, as many things as I think they need to know ab out. Usually 1 or 2 faults per board and about 5 boards, trouble is I need to ma ke them all up and they should all look the same at a glance.

Get me to do them.

:o(

Reminds me of a similar sort of course I ran for engineers who would be maintaining the 1983 Tube Stock (now retired for service). The traction control equipment was predominantly electromechanical, and there was quite a useful test box which enabled you to simulate the inputs the train would normally provide, and monitor LEDs showing circuit behaviour.

After a spell of familiarisation, they got to look for faults I had created. The one that took longest to locate was a bit of masking tape on an auxiliary contact on one of the power switches.

Chris

I've tried this

but it just seemed to move out of the way as I soldered I think it's mainly for flow soldering use.

Solder resist is on virtually every PCB you see. Often green, but can be any colour. Possibly even clear. To prevent solder flowing beyond a solder pad down the track, etc.

Best way to attempt a dry joint would be to use solder without flux and just enough temperature to soften it.

But many dry joints are the result of the joint heating and cooling and flexing in some way. So really a mechanical, rather than chemical, failure. On through hole stuff you can sometimes see them - a ring round the component lead.

A squirt of freezer spray at the instant you remove the iron might help.

Cheers

I do a lot of soldering, so some points:-

Never use lead free. Spawn of the devil. A decent 60/40 is fine for GP use.

Generally, thinner multicore than the standard 1.2mm diameter stuff is more useful, especially on a PCB. 0.8mm is the one I most use. Despite having 4 sizes on my home made dispenser. Less chance of creating solder bridges and also seems to have a better flux to solder ratio. For larger things you can just use more.

A decent iron makes life easy. Best by far is a good low voltage 50 watt temperture controlled type.

And don't be in a hurry to make the joint. Make sure the solder has flowed properly. Especially on many multi layer PCBs where it has to go from front to back, and make a joint to the inner layer. With the iron at the right temperture, you get far longer than you might think without risk of damage to components. About 3 seconds will do most.

I tend to use a 450C setting on my iron.

Agree with almost all you say, especially lead free., personally I use around 390C as this limits insulation damage I find.

I just did 18 solder joints, and another 18 tomorrow (along with

(OOI) These would be wire 'terminals' though wouldn't they, rather that through-hole PCB, or are they all mounted on a filter board / PCB?

When at BT I was taught all wire terminations should be 'mechanical' first, then soldered, the solder just reinforcing the 'electrical' (not mechanical) part of the joint. So, if an axial cap lead was being soldered to a tag or terminal (as seen on the back of a loudspeaker) you would bend the end of the cap lead through the hole in any solder tag and lightly crimp it down first, then solder it.

Some do, our Yamaha active sub does for example, and most 'PC speakers'. Not sure you should try it on yours (well, unless you want to hear what the mains sounds like for a couple of milliseconds that is). ;-)

Yes it would be ... and most other 'real world' things for that matter .

Cheers, T i m

:-) Yes I still have some Multicore from the old days. Plus a reel of fairly fine stuff that my bro brought back from the Navy 55 years ago.

I used my 30W that I bought while I was at CERN. Since its a 220V one it may be running a little hotter.

Luckily the x-over boards are fairly simple single layer. Biggest problem is I'm getting to old for this stuff. I have to lay the speaker on its back, take the big unit out ands that gives access to the x-over. Remove that from its foam backing with cheesewire and off we go. Luckily I've got a solder sucker too.

Thanks for the hints.

No, its a pukka PCB with 9 caps on it, a resistor, and 4 chokes. The box is very heavy and has four drive units. I've been replacing all the caps. Done one so far, do the other tomorrow.

Yes, 240V on the speaker terminals would be amusing.

This *is* a real-world thing, old boy.

I'd not guarantee my iron is actually 450C just because the readout says that. ;-) But it is at the upper limit of what I'd use.