Hand-soldering QFN-32's

Go To Last Post
19 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is this even POSSIBLE? Anyone here do it with success on a regular basis?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For me, TQFP's are the limit. But I know that there are freaks that go even further.

Nard

A GIF is worth a thousend words   They are called Rosa, Sylvia, Tricia, and Ulyana. You can find them https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've done a couple, it isn't hard. Just extend the pads a little and use drag-soldering.

Leon

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've done it with a toaster oven and a temperature probe. I used a syringe of solder paste and it was very easy since the solder surface tension pulls off the solder mask and onto the pads, just don't use too much and it causes the part to float into position.

See the tutorial on www.sparkfun.com.

Smiley

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I agree with the solder paste method. It gives very good results. It's kinda fun to watch too. Really neat how it pulls the part into alignment.

Go electric!
Happy electric car owner / builder

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

extend the pad a little, with the default pcb packages once the qfn is in place it almost totally covers all the pads, this make positioning very hard as you can not really see the pads.

Flux does 90% of the work. Tack on 2 pins with solder, then flood all the pins with solder, bridges are good as you wants lots of solder. Then apply flux, move the iron across the pins and presto it is like waving a magic wand, everything looks good with no bridges.

I never used the toaster oven method, but I am sure it has got to be much easier and yeilds a higher rate of working boards. Hand soldering qfn is brutal, it gets better as you get experience but it still seriously blows.

Regards,

Alan To

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I had tryed once to follow the toaster oven method, but it seems that my oven is too slow to raise the temperature. Is there something special about them? Some one preferred? Models? Small ones?

I had luck with toalan/Alan To method, but I don't want to repeat.

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Guillem,

I was able to follow the heating profile chart timing using a cheap $20 toaster oven, wall clock, and a voltmeter with a temperature probe. I thought about building a controller for doing, but since I could do it good enough by eyeball I couldn't see investing the time. Sparkfun now likes using an electric skillet. Have you read their site tutorial? Could it be something about specifically Spanish appliances or voltage that causes the slow heating time?

Smiley

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I had to quickly hand solder a 16 pad LFCSP package (which is quite similar to a QFN) yesterday. I had no access to an oven or paste or even a PCB. I glued the IC upside down to the non metallic side of a piece of stripboard, then hand soldered very thin strands of wire (single strands from a multi-strand ribbon cable) to the pads. I used one of those desktop magnifying glass (with the circular light around the lens), an extremely fine tipped solder iron and flux. I dropped small amounts of flux over the IC, then placed a tiny amount of solder on the tip and dabbed it about the pads. this gave me solderable pads. Ran a leeeetle bit of solder over the end of the really thin wire and then used the magnifying glass to patiently solder each pad that I required. The trick is to keep changing the orientation of the IC such that you can position the solder iron so that it DOESNT melt already soldered pads (wow is that irritating). After each wire was done I pulled it through a hole in the stripboard and soldered it to a strip on the other side. It looks like a spider web.

NOTE this was a quick HACK not recommended for more permament applications, but it can be done.

"You live and learn. At any rate, you live."

Douglas Adams

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've tried LQFP48 and LQFP100 (0.5mm pinout) pin by pin and it's possible.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There are also surface-mount-to-through-hole testing sockets available, as found at:

http://www.enplas-ets.com/enplas...

"You live and learn. At any rate, you live."

Douglas Adams

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, Smiley, seems that I had tryied with a too big oven (too much air to heat, and not so big resistors, about 1500W), not an small toaster oven. I should try these ones.

I had succesfully soldered some QFN 16, and similar with a fine soldering tip. Also done dinofizz solution (we call these 'spiders' or 'spiderwebs') with success. But not 100% success always, and also some minor problems to solder the bottom pad. I had also managed to solder LQFP's (Atmels ARM and NXP's LPC uC) and FPC connectors, all with 0.5mm pitch, and this is about my absolute limit right now (and also of my prototyping process resolution). All by hand, fine tip iron, fine soldering wire, lots of flux, desoldering station, and much, much patience.

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

http://www.youtube.com/watch?v=4...

Wow. This guy is, for lack of a better word, "sickhouse."

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Guys, remember that a 'gas soldering iron' can be used to solder surface mounts using paste, without using an oven. and without shelling out for a 'rework station'.
these lighter-gas powered thing usualy come with a hot air tip that can melt solder paste in seconds. You can even solder big parts (Mega128) using that system.

I use a frying pan + paste for large pads like connectors (hot hair can melt the connector's plastic too), but I use the 'hot air' pen to solder the rest, because you can do that at your pace and test the board as you go along...
A thin filet of paste across the pads and hot air can solder pretty much anything, .5mm included!

This is the thing I use.

Author of simavr - Follow me on twitter : @buserror

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've seen a hot air gun used successfully with a QFN32, the caveat is to not blow the chip off the board.

Smiley

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Guillem Planisi wrote:
I had tryed once to follow the toaster oven method, but it seems that my oven is too slow to raise the temperature. Is there something special about them? Some one preferred? Models? Small ones?

I had luck with toalan/Alan To method, but I don't want to repeat.

Guillem.

I've attached a profile that many use to solder SMD technology with a Toaster oven.

It might help.

A key part of successful Toaster oven reflow is that, you have to have enough BTUs (Watts) to exceed the anticipated temperature rise, and then control that temperature rise to where you it needs to be.

A movement of a few cubic feet per minute thru the oven work chamber also helps distribute the temperature uniformly, as well - but not so much air flow that you blow the components off of their pads.

Attachment(s): 

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks, Carl. That's exactly what I found. Not enough watts (we don't have British Thermal Units in Spain, jeje) for a big kitchen oven for Pizzas and so on. I would try some toaster oven, but I never saw anyithing similar to the one posted at sparkfun in Spain. We have small ones that could fit the job perfectly, so I would give them a try when I have time (quite sparse, that's the main difficult I have in fact).

Also the temperature profile you give us is nice, since it its for leaded alloy, that is what finally I use for prototyping. I was using the nasty lead free alloy for production (if some people doesn't know, Spain is in Europe), that has a higher peak temp (thus even more problems with temperature rising, since the maximum our oven achieves is about 230 - 240 ºC). But for prototyping (that's where I need this, production has it own soldering process) there is no problem to use leaded alloy and your temperature curve (that I didn't have).

Thanks again,

Guillem.

Guillem.
"Common sense is the least common of the senses" Anonymous.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for all the replies. I'll have to attempt the toaster oven reflow, as I happen to have a spare toaster available. If not, I now know that soldering it is possible....

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I recently hand soldered a 64pin QFN (LFCSP) 0.5mm pitched for the first time. Using the same technique I learned with TQFP's I tacked to opposing pads, then flood the area with liquid flux (water soluble flux). I use a soldering iron tip with a tiny cup that run along the perimeter of the package, making sure the molten solder blob contained in the tip touches the edge of the package and each pad. After doing that, there shouldn't be any bridges, but if there are, you can solder wick it. You must have lots of flux and it must bubble and fume while your soldering. But from what I here, the solder paste and toaster oven method won't raise your blood pressure as much trying to soldering these buggers.