[sdiy] 0.100 pin header reliability.

[Harry Bissell]

I have rarely (in fact never) seen aluminum as a metal in connector pins.
Usually they are steel (cheap) or brass, or beryllium copper...

Aluminum does not solder well at all using any standard electronic
solders... (i hear there are fluxes for this, but have never needed them...)

H^) harry


----- Original Message -----
From: cheater cheater <cheater00 at gmail.com>
To: Dave Kendall <davekendall at ntlworld.com>
Cc: synth-diy at dropmix.xs4all.nl
Sent: Mon, 15 Nov 2010 04:05:11 -0500 (EST)
Subject: Re: [sdiy] 0.100 pin header reliability.

Hi Dave,
to prevent breaking of soldered cable, have it zip tied to the PCB
about 2 cm away from the spot it was soldered to. Zip tie first,
solder second, so that the solder solidifies in the right shape.

With regards to soldering of crimped connectors: there might be
several things at play here, but I'm not good enough with my chemistry
to know if they actually make an impact. Maybe someone can tell me if
my intuitions here are right or wrong.
1. three/four metals (aluminum for the connector, copper for the cable
and solder, tin for the solder) make for different electrical
properties, oxides migrate differently
2. the solder crystals change shape over time. I understand there's
some sort of evolution to solid solder that makes it eject oxides?
This could not only separate the cold weld but also make it dirty with
oxides. Maybe someone can confirm this.
3. Is solder perhaps more reluctant to make a cold weld with aluminum,
than bare copper touching bare aluminum? If you solder first and crimp
second, you're crimping against tin. This also has the effect that the
tin soldered to the cable and/or connector might separate off what it
was soldered to, making an even worse connection
4. If you crimp first and solder second, the heat is going to expand
the aluminum more than it expands the copper wire - aluminum has a
higher linear coefficient of thermal expansion - effectively
un-crimping the connector. I don't think it's going to re-seat itself
nicely back to being crimped afterwards, but I don't know. My guess
here is that you actually need a lot of force to create a cold weld,
and this force is not going to be supplied by the shrinking connector
alone. Besides, there might be solder in that space now, which messes
up the connection
5. My intuition is that at least the better connectors have the leafs
produced in such a way that the crystalline structure is good for
making cold welds. Heating up changes this structure, quite possibly
for worse.

It would be interesting to know if any of this actually holds up
according to theory.

Regarding saving time: I read this thread and I think to myself...
isn't it much better to just buy ready cables?... They'll be crimped
much better than you can do it anyways, they'll be much cheaper in the
end, and they don't cost you time..

Cheers,
D.

On Sat, Nov 13, 2010 at 02:58, Dave Kendall <davekendall at ntlworld.com> wrote:
>
> On Nov 13, 2010, at 00:50, Paul Perry wrote:
>>
>> I think that when soldering decreases reliability it is because
>> the solder wicks up the wire away from the joint & stiffens the wire,
>> which then tends to flex sharply at the end of the solder & break there.
>
> That makes sense. But wouldn't that same potential weakness also apply to a
> wire either directly soldered to the PCB pad, or to a PCB pin?. I've had
> that happen on some perfboard prototype builds using stranded wire that got
> some rough handling. (the soldered joints to PCB pins weren't reinforced by
> heatshrink or hellerman sleeves or anything, so I wasn't too surprised...)
>
> This is an interesting thread, thanks for everyone's input.
>
> cheers,
> Dave
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
_______________________________________________
Synth-diy mailing list
Synth-diy at dropmix.xs4all.nl
http://dropmix.xs4all.nl/mailman/listinfo/synth-diy

-- 
Harry Bissell & Nora Abdullah 4eva