Searching \ for '[EE] TQFP alignment and soldering tips' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=tqfp+alignment+soldering
Search entire site for: 'TQFP alignment and soldering tips'.

Exact match. Not showing close matches.
PICList Thread
'[EE] TQFP alignment and soldering tips'
2006\10\09@104750 by slippyr4

picon face
Hi All,

I've got me some 18F6620's, which are TQFP 64 packaged, and a
prototype board (the PICPROTO80 from
http://melabs.com/products/picproto/40+pproto.htm). I need the 6620
cos i need two hardware usarts.

I've bought a hot air rework station, and by tomorrow I should have
the 10mm qfp nozzle for it, so i shall be tackling soldering it.

Any tips? i've never tackled something like this before. I've not had
great success even aligning the chip on the board. The pads are
pre-tinned and the chip tends to drop with its legs between the pads,
on the solder mask. Will surface tension sort it out when i heat it?
or do i need to get the aim absolutely perfect?

Any pointers appreciated. I've googled loads, but theres not much on
how to use a hot air rework station- most the sites ive found give
techniques for iron-soldering.

thanks

jon

2006\10\09@110358 by Robert Young

picon face
{Quote hidden}

I have never had good luck soldering this pitch with a hot-air-pencil.

IR reflow or a fine-tip soldering iron have worked better.

Pre-tinned pads may not have sufficient solder present to make a good
joint, depends on how much was removed when the pads were leveled during
the tinning process.  Insufficient solder and uneven heating with the
hot-air pen may not allow the surface tension of the liquified solder to
pull the part onto the pads.

A good mangnifying glass, plenty of extra flux, patience, 0.015" solder,
fine solder wick and a fine tip iron may be your best bet for soldering
one or two pieces rather than trying to use a hot-air pencil.  

If you must use the pencil, consider getting a small syrenge of solder
paste, a fine needle to apply it and laying a small bead on all four
sides of the pads, directly under where the leads will contact the
parts.  CAREFULLY place the part, align it to the best of your ability
and then heat all four sides at the same time.  But no guarantees if you
can't get good pre-alignment of the part or even heat.  Since I don't
have a rig for holding the pen and TQFP tip at right angles to the board
surface, I haven't tried this recently.  Did work for big PLCC packages
but they were to heavy for the air flow to move and the pads are "huge"
as surface mount parts go.

Rob

2006\10\09@111507 by Mike Harrison

flavicon
face
On Mon, 9 Oct 2006 15:47:48 +0100, you wrote:

{Quote hidden}

I wouldn't use hot air - good for desoldering but not great for soldering IME.
The way I usually do TQFPs is to put a dot of solder paste at each end of one row, place the part
using my cheap substitute for a vacuum pick-up (a probe with a blob of sticky blu-tak on the end).
Align using magnifier, and flow the pins once aligned (while holding chip down with a pointy probe)
to tack it in place. Re-check alignment and adjust as necessary (slight misalignements can be done
without reflowing the tacked pins - just bending slightly as you push the package.  
Then put a line of solder paste along all 4 rows, and flow using a wide blade tip, doing the tacked
side last. When flowing, slide blade along the length of the pins, away from the chip, to draw off
excess solder.

Another method I used to use with pre-tinned pads is to use a metal roller to flatten the pads so
the part sits in the right place.  

Whatever method, make sure there is flux on the PCB first.


2006\10\09@124009 by Orin Eman

picon face
On 10/9/06, slippyr4 <.....slippyr4KILLspamspam.....gmail.com> wrote:
> Hi All,
>
> I've got me some 18F6620's, which are TQFP 64 packaged, and a
> prototype board (the PICPROTO80 from
> http://melabs.com/products/picproto/40+pproto.htm). I need the 6620
> cos i need two hardware usarts.
>
> I've bought a hot air rework station, and by tomorrow I should have
> the 10mm qfp nozzle for it, so i shall be tackling soldering it.
>
> Any tips? i've never tackled something like this before. I've not had
> great success even aligning the chip on the board. The pads are
> pre-tinned and the chip tends to drop with its legs between the pads,
> on the solder mask. Will surface tension sort it out when i heat it?
> or do i need to get the aim absolutely perfect?
>
> Any pointers appreciated. I've googled loads, but theres not much on
> how to use a hot air rework station- most the sites ive found give
> techniques for iron-soldering.

Take a look at the tutorial under:

http://www.sparkfun.com/commerce/hdr.php?p=tutorials

Orin.

2006\10\10@061738 by slippyr4

picon face
well, i tried, and i failed badly.

i used solder wick to remove most of the tinning on the board- this
allowed me to align the chip better and stick down one corner with an
iron.

then i added flux and paste, and soldered reasonably well with the hot
air. but there were a couple of bridges that i just could not remove
no matter how much i tried.

so i decided to remove the cheap, clean up and try again,

except i ripped a load of pads off the board whilst removing it.

:-(

On 09/10/06, Orin Eman <EraseMEorin.emanspam_OUTspamTakeThisOuTgmail.com> wrote:
{Quote hidden}

> -

2006\10\10@064646 by Wouter van Ooijen

face picon face
> but there were a couple of bridges that i just could not remove
> no matter how much i tried.

that's where you could have used the solder wick!

> so i decided to remove the cheap, clean up and try again,
> except i ripped a load of pads off the board whilst removing it.

How did you do that? Hot air might not be ideal for soldering, but in my
experience it is perfect for removing. Just apply hot air, and remove
with a pencil. Or when it is the only component: by a gentle wack on the
back of the PCB.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\10\10@072401 by Mike Harrison

flavicon
face
On Tue, 10 Oct 2006 11:17:36 +0100, you wrote:

>well, i tried, and i failed badly.
>
>i used solder wick to remove most of the tinning on the board- this
>allowed me to align the chip better and stick down one corner with an
>iron.
>
>then i added flux and paste, and soldered reasonably well with the hot
>air. but there were a couple of bridges that i just could not remove
>no matter how much i tried.

1) Clean the old flux off, and add more fresh flux.
2) cut the end of the solder wick at an angle to give a fine pint
3) apply the point to the short and heat from the other side of the wick

Remember solder likes to flow towards heat, to where possible, apply the iron to the braid, and the
wick to the joint.

If it doesn't draw the solder off, get better wick - there is a big difference in effectiveness
between brands. 'Soder-Wick', size 4, from Chemtronics is by far the best I've found.

2006\10\10@083217 by Robert Young

picon face
>
> then i added flux and paste, and soldered reasonably well
> with the hot air. but there were a couple of bridges that i
> just could not remove no matter how much i tried.

Sometimes if you reflux the joint and ADD a little more solder (ie flood
the joint) then go back and use a good quality pre-fluxed wick, the
solder will flow out.

The key is having sufficient flux, even heat and a steady hand.

You picked a pretty small pitch part for your first try though.

Rob

2006\10\10@104524 by slippyr4

picon face
maybe i should have added some more solder. the bridge i couldn't
shift was kind of where the pins are vertical - the horizontal parts
were all nicely on the pads.

I'm seriously considering crossing to the dark side, and using an AVR
for this project. They make a 40pin DIP part that has two hardware
UARTs.

2006\10\10@115608 by Mike Harrison

flavicon
face
On Tue, 10 Oct 2006 15:45:21 +0100, you wrote:

>maybe i should have added some more solder. the bridge i couldn't
>shift was kind of where the pins are vertical - the horizontal parts
>were all nicely on the pads.

Decent solder wick would remove this easily.

2006\10\10@120238 by olin piclist

face picon face
slippyr4 wrote:
> I'm seriously considering crossing to the dark side, and using an AVR
> for this project. They make a 40pin DIP part that has two hardware
> UARTs.

Microchip makes a 28 pin DIP with two UARTs.  It's called the 30F3013.

********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2006\10\10@121615 by M. Adam Davis

face picon face
It just takes time an experience.

When you have some bridges that don't seem to go away, add _a lot_ of
flux to the pins that are bridging.  Cover the pins completely in a
little pool of flux.  Then make sure your solder tip is shiny and free
of any extra solder.

Now place the solder wick against the pins , and press on the wick
with the soldering iron, so the wick presses against the pins.  Hold
it there for at least 12 seconds, while the flux boils away.

The bridge should be gone.  The main point is that you have to use a
lot of flux.  If you can't see the flux color without magnification,
add some more.

-Adam

On 10/10/06, slippyr4 <@spam@slippyr4KILLspamspamgmail.com> wrote:
{Quote hidden}

2006\10\10@134043 by slippyr4

picon face
On 10/10/06, Olin Lathrop <spamBeGoneolin_piclistspamBeGonespamembedinc.com> wrote:
> Microchip makes a 28 pin DIP with two UARTs.  It's called the 30F3013.
>

wow, i hadn't thought to look at the dsPIC range- i've always regarded
them as too "hardcore" for me, since i've no need of any dsp stuff at
all.

OOI, how do the pro-embedded engineers prototype? do you work entirely
theoretically until you're sure you have a working design, and then
make a pcb? seems most the newest chips are in qfp's which are
awkward.

2006\10\10@134636 by Brooke Clarke

flavicon
face
Hi Orin:

I'm just starting to learn how to solder surface mount parts and have a
web page about it at:
http://www.pacificsites.com/~brooke/SMT.shtml

One of the tricks of getting solder wick to work is to wet it with
LIQUID flux.  The dry flux that may be on the wick is not anywhere near
as good as using liquid flux.  The propose of the flux is not so much to
act as a wetting agent but to act as a thermal conductor.  It's much
better than air at conducting heat and that makes the wick work so much
better.  This is especially true when you're working on something that's
about the same size as the air gaps in the wick.

Have Fun,

Brooke Clarke

--
w/Java http://www.PRC68.com
w/o Java www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com

2006\10\10@135745 by Shawn Wilton

picon face
I burn a proto board either using PCBExpress, ExpressPCB, Advanced Circuits
or BatchPCB.  Any of them will work.  But I'm also reasonably certain my h/w
will work the first time around.  Otherwise a dev board/environment is best
in my opinion.

I also use either a mylar or stainless steel stencil to run my board.  Apply
paste with the stencil, apply parts with tweezers (some day I will get a
vacuum pencil) and then reflow in an old toaster oven.  Works perfectly.
Trick is how long you let the parts heat up.  I have found for the
particular solder paste I use, that a 2-4 minute warm up period is just
right.  Too much and I end up getting some nasty looking boards that will
not reflow properly presumably because my flux begins to evaporate.

Also, when you go to do that TQFP part (assuming you have another board to
use), do opposing corners first, then go pin by pin with a fine tip and fine
solder.  The finer the better.  Turn the iron up to around 650-750F so that
the heat is transferred quickly.  This will prevent you from having to apply
heat for in duration potentially damaging the part.  Remember it takes only
a second to heat that pin enough to reflow the solder.  The other option is
to pretin all the pads, put the part over the pins, do opposing corners
first.  Then just go pin to pin with the iron heating each pin and pushing
it down in to the solder.  Both have worked great for me.  Experience will
determine time required.



On 10/10/06, slippyr4 <TakeThisOuTslippyr4EraseMEspamspam_OUTgmail.com> wrote:
{Quote hidden}

> -

2006\10\10@164535 by Gerhard Fiedler

picon face
Shawn Wilton wrote:

> I burn a proto board either using PCBExpress, ExpressPCB, Advanced
> Circuits or BatchPCB.  Any of them will work.  But I'm also reasonably
> certain my h/w will work the first time around.  Otherwise a dev
> board/environment is best in my opinion.

Right. And you don't have to breadboard your whole circuit, with the final
target processor. Usually, it's only a small part of the hardware that's
doubtful, and usually many different processors serve to test it.

Gerhard

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...