On Mon, 29 Nov 2010 12:37:39 -0800 (PST), DerbyDad03
The devices in the circuit board are mechanically held in place by
virtue of passing through the board - and the solder just stiffens the
wire to make it harder to pull apart, then glues the twisted wires so
they cannot easily move in relation to each other. Two wires laid
together and soldered can be separated relatively easily in comparison
- and a wire just laid on a circuit board and soldered WILL fail.
On Nov 29, 5:16 pm, email@example.com wrote:
re: "The devices in the circuit board are mechanically held in place
by virtue of passing through the board"
Not on any boards I've built or repaired. The wire or component leads
are loose in the holes and will pull right out unless solder is
applied to the tinned pad on the lead side.
Ever encounter a cold solder joint on a circuit board? A nice neat
mound of solder that looks good on visual inspection but a component
lead that will pull right out through the board because there is
nothing holding it in place.
On Mon, 29 Nov 2010 15:15:56 -0800 (PST), DerbyDad03
They are held "in place" except for one direction. Might be more
accurate to say they are "located" by the holes.
Surface mount is a different story - and they DO tend to fail on
Yes I have - but if it was not in the hole the cold solder joint
would just be a spot of copper - and a component rattling around
Also, PROPER installation involves spreading the leads slighly when
they come through the hole so the component stays in place when you
either turn the board over to solder it, or move the board to the wave
Mechanically placed components virtually all have (or at least had)
that little kink that held the part in place.
All the custom boards I had made also did. (we are talking in the
computer industry) And all the automotive computers I've worked
on/repaired (up until the '90s)
And a very large number of SMTs are glued to the board (solder paste
in some cases) which does not do much for the final strength, but
keeps the part in place until it is soldered
A lot of the stuff I worked on was custom - and not high volume - and
the little "hook" was standard, virtually across the board. Dealt with
some pretty high end contract assemblers. Waterloo Region is NOT the
low-tech part of the country. A lot of the work these guys did was
aerospace or high end industrial electronics.
Ever hear of Dalsa, ComDev,Christie Digital, RIM, Agile Systems,
Accellerated Systems, SRE Controls, Navitas, Virtek, Raytheon, NCR,
ATS, just for starters.
Your technique may need revision.
I was taught to push the lead through the hole, bend it over to provide
mechanical connection, solder, then clip the dog leg.
You are evidently skipping the "bend over" step. Your technique makes
removing the component MUCH easier - until the component simply falls out on
On Mon, 29 Nov 2010 22:42:41 -0600, " firstname.lastname@example.org"
When the joint area approaches or excedes the area of the device
itself, and the mass of the solder involved in the joint is a
significant fraction of the mass of the device, and there is no "wire"
involved, you are pretty safe. That pretty well describes SMT.
Lay a wire on a PC trace and solder it on and see how much abuse it
takes (generally before pulling the trace off the board - but
occaisionally pulling the wire off the trace)
On Tue, 30 Nov 2010 01:37:37 -0500, email@example.com wrote:
You're talking through your ass. Think BGAs.
It is perfectly normal to surface solder wires onto SMT boards. Mechanical
connections are *not* required, nor desirable, before soldering. As has been
stated by others here, the only reason to have a mechanical connection before
soldering is if the connection won't make it until it's soldered.
I've spent as many years in the computer repair business as I have
Instead of just dissagreeing with everyone why don't you give us some
real information - like what you work on that things are done
differently. I'm sure there are many reasons for doing things
differently in diferent industries and situations.
My experience has been MOSTLY automotive and computer, with some
communications and audio and avionics thrown in for good measure.
I've done component level repair on automotive electronics and
computers, as well as pre and early solid state radio and audio
I've actually done more "discrete component" assembly than LSI stuff
- and a fair amount of the old "point to point" as well. Printed
circuits were still pretty crude when I got started.
Did a fair bit of vacuum tube stuff too.
Oh. Didn't know that. I suppose the cheap, disposable tips are merely plated
(maybe even just painted in bright colors) - the ones I use are solid
Agreed solder IS a mechanical connection; a piss-poor mechanical connection
at that. Lay two wires side by side and solder them together. Then pull them
apart. Next tie the two wires together in a knot, solder them, and pull them
apart. If you can.
There's a HUGE difference between a soldered mechanical connection and the
mechanical connection of a solder joint.
The cheapy screw-in tips I get at radio hack are copper base. Dunno what
the plating is on top, though.
As for tinning them, after I degrease them (when new), I put a very thin dab
of flux on the tip and wrap it with flux-core solder, then I plug in the
This gets it tinned before it has a chance to oxidize, and seems to work
well for me.
A properly soldered joint doesn't not need a mechanical connection, like
twisting the wires together, in order to make a permanent electrical
connection. Matter of fact, NASA prohibits any such joint. They
prohibit it because if repairs are needed it is too difficult to take apart.
You are using a double negative so appear to be disagreeing with
tourself - but regardless, nasa notwitstanding (if it is true)
acceptable standards for aircraft repair ( I believe it is AC 43.13)
specifies joints must be mechanically secure before soldering, but
also (quite severely) limits where soldering is acceptable in aircraft
wiring. Solder wicking back a stranded wire hardens the wire and makes
a stress point where that hardening stops - a failure waiting to
happen when exposed to vibration.
On 11/30/2010 1:00 AM, firstname.lastname@example.org wrote:
Ah, a typo. Read it as a single negative and you will understand what I
mean. Another reason solder joints shouldn't be mechanically secure
before soldering is because that practice can hide poor solder joints.
It may be a cold solder joint but it doesn't show up easy as if it would
without a separate mechanical connection.
As far as solder wicking up wire, yes I agree the wicking and vibration
is a very bad combo. As far as NASA, I believe that was mostly applied
to components on circuit boards.
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.