X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Virus-Scanned: Debian amavisd-new at mail.linetec.nl
Message-ID: <66c88bcb-820c-9a1d-1698-d0b36f32e3f3@linetec.nl>
Date: Wed, 17 Aug 2022 13:15:47 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.11.0
To: geda-user AT delorie DOT com
Content-Language: en-US
From: "Richard Rasker (rasker AT linetec DOT nl) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Subject: [geda-user] [OT] Solder paste woes
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Reply-To: geda-user AT delorie DOT com

Hello,

My apologies for this off-topic question, as this has not much to do 
with gEDA - but I could do with some expert advice on PCB assembly for 
reflow, and perhaps some people here have good ideas.

I've been working with SMD reflow technology for many years now, and 
very step of the process goes smooth except one: applying solder paste 
for small-pitch components.

My main problem is that even after hundreds of PCB's, the result is 
unpredictable. Only 2 out of every 10 PCB's have a good crisp result, 
like this: http://www.linetec.nl/electronics/paste_crisp.jpg

The other PCB's often look like this: 
http://www.linetec.nl/electronics/paste_mess.jpg

So applying paste is rather a hit-and-miss affair, and it sometimes 
takes half a dozen attempts to get one PCB right.

This is my set-up and work procedure:

* I only use flash gold finished PCB's (the tinned ones have bumpy pads, 
hugely increasing the error rate).
* For positioning of the PCB's, I use an L-shaped piece of PCB taped 
down with masking tape.
* The SMD stencil is stainless steel, 100 microns thick, taped into 
position along the bottom edge and at a top corner.
* The wiper is also stainless steel, but of course rather thicker.
* I apply the paste slowly with the wiper at an angle of ~30° relative 
to the horizontal, with small sideways movements in addition to the main 
movement downwards.
* I use moderate force -- difficult to estimate, but I'd say between 500 
and 800 grams, so roughly 1 - 1.5 pounds.
* After each single paste application, I wipe the underside of the 
stencil with an acetone cloth, to prevent any paste on the underside 
getting squished, causing shorts etcetera.

Now admittedly, those 'shorted' pads often turn out OK due to the 
solder's high surface tension and tendency to collect at the metal parts 
of the components, but I still get rather a lot (ca. 1 in 5) of PCB's 
with nasty shorts, especially underneath those QFN housings.

There is of course also the problem of the solder paste degrading and 
getting tougher and 'dryer' over time; however, that is not the problem 
here. If anything, fresh solder paste is more runny and thus tends to 
short out between pads faster.

So the question is if someone has any tips to improve the process, or 
maybe point out what I'm doing wrong. (Yes, of course I can have a PCB 
house do the assembly, but that typically costs > $1000 per run, and 
that is not always an option.)

Thanks in advance,

Richard