Archive of the former Yahoo!Groups mailing list: Homebrew PCBs

previous by date index next by date
previous in topic topic list next in topic

Subject: Re: [Homebrew_PCBs] Soldering a LQFP 48 on a home made PCB

From: Erik Knise <elknise@...>
Date: 2014-11-07

Dwayne,

I too have worked on a lot of raw uncoated copper boards.  One of my customers had a board with tons of surface mount parts on it that we went through 30-40 iterations of until he got his design nailed down.  Once we got solder masked boards and were able to run them on the pick and place it was a breeze except for the few 0201 and 0402 parts I still had to hand solder.

I agree that tinning them would be a good idea but I've always soldered my prototype boards without tinning.  I found that for larger parts with lots of leads that it was almost impossible to get the tinning consistent enough to have the part lay flat on the board.   It's hard enough with an unmasked raw board getting the parts position correctly, but red neck tinning doesn't seem to work well.  Sometimes when tinning those boards you can put too much solder on the traces under the part and it won't sit flat.  The few times I had to tin, I tinned lightly with a soldering iron and then reflowed with more flux and a hot air re-work tool to get it to smooth out.

I honestly haven't had that many issues with it.  I fluxed the leads of the part and the pads on the board.  Put a blob of solder on my soldering iron and after positioning the part I tap two opposing corners to tack it down.  Works 99% of the time.  Then I drag solder everything else and clean up the tacks last.  We were using water soluble solder and liquid flux.  Haven't had as good of results with rosin core or no-clean.

--
Erik L. Knise
Seattle, WA

On Fri, Nov 7, 2014 at 8:37 AM, Dwayne Reid dwayner@... [Homebrew_PCBs] <Homebrew_PCBs@yahoogroups.com> wrote:


The problem with all of the suggestions made so far is that everyone isassuming that the pads already have solder on them.

But this is a brand-new home-made board with bare copper pads.

The first step is to tin all of the pads.  Use lots of flux - gel orpaste flux works best but standard liquid flux will also work. Flood all of the pads with solder, then apply more liquid flux on top ofthe solder blobs.  Hold the board vertically with the pads vertical,then use a soldering iron to draw the solder down to the ends of the padsand on to the tip.  Rotate the board 90 degrees and repeat until allof the pads are tinned and cleaned.  The pads should now have athin, smooth layer of solder on top of the copper.

Do a very good inspection with magnification to ensure that you don'thave any solder bridges.  Then follow the (excellent) advice alreadygiven.

Hope this helps!

dwayne

PS - we used to make hundreds and hundreds of our own PCB's in the earlydays of our business.  We would clean the boards mechanically with afine-grit sander followed by a 3-step chemical cleaning process. Blow the boards dry with warm air from a vacuum-cleaner motor, then feedthe boards through a GBC laminater (modified for lower temperature andhigher speed) loaded up with Dupont Riston dry-laminate film. Exposed the boards with a modified mercury-vapor expose lamp and vacuumframe, then developed the exposed boards with a soda-ash (potassiumcarbonate?) solution.  Stripped the remaining laminate in acaustic-soda (sodium hydroxide) bath, rinsed, then etched in an AmmoniumPersulphate bubble etch tank.

The whole process worked very well but it just took too much time. We eventually went to CNC milling for PCB prototypes (several years),then quit that and just started using APC for prototype boards.  Allof our production boards are now made in China - the quality is extremelyhigh and the cost is astonishingly low.  APC still does ourprototype boards.

dwayne


--
Dwayne Reid   <dwayner@...>
Trinity Electronics Systems Ltd    Edmonton, AB,CANADA
(780) 489-3199voice          (780)487-6397 fax
Custom Electronics Design and Manufacturing