_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

I wouldn't hold my breath. The last update on the site is from June 2011. Too bad, because I was curious about his results.

Indeed.

Sparkfun has a reflow oven tutorial:

http://www.sparkfun.com/tutorials/60

I don't know about ovens, especially for reflow. What if you've got plastic connectors onboard after a suspected failure of BGA mount? They're going to melt.

Take, for example, a person trying to make a video board trying to tackle BGA mounting. In order for a first test, one needs the JTAG and DB15 VGA connectors along with the BGA soldered. If the project doesn't work, what will be the first and final suspect if the circuit doesn't work, especially if it has progressed from a previous successful design based on QFP IC packages and the only detail that's changed is a BGA IC for the FPGA, or some other IC?

For this reason, I would lean towards a SMT reflow station with a hot air gun. These stations are cheap, usually come equipped with a soldering iron with many size tips, both for the iron and the air gun, and have temp control for both.

However, since my experience is/has been mainly QFP where the conductive metal leads stick out and absorb the heat before the QFP plastic package, I would be worried about heating up the plastic BGA package, in the same manner, to a point before the balls underneath start to melt.

_________________
65Org16:https://github.com/ElEctric-EyE/verilog-6502

I've been building a reflow oven at my local Hackspace using a £25 Argos toast oven and a bit of electronics. We salvaged a solid state relay from an old laboratory sample robot.

_________________
Andrew Jacobs
6502 & PIC Stuff - http://www.obelisk.me.uk/
Cross-Platform 6502/65C02/65816 Macro Assembler - http://www.obelisk.me.uk/dev65/
Open Source Projects - https://github.com/andrew-jacobs

Yes, at that point you're looking for a rework solution. I think reflow makes sense for the initial soldering. I can imagine you might only reflow one or a few SMD parts and then do the rest manually.

For hot air rework, I think there are special nozzles for heating just the pins of a QFP type package. You are meant to make a thermal shield to protect other parts on the board.

I have a cheap hot air gun that I use mostly for desoldering large devices. I just use a round nozzle, and aim it on the center of the device until it gets hot enough to come off. I then usually throw the device away, and put on a new one. When I first got the hot air machine, I tried soldering stuff, but I abandoned that plan quickly. Getting an even temperature just above the melting point, only at the right place is difficult. And with lead free soldering, there's not much room between the melting point, and the point where stuff gets overheated.

Depending on what it is, you may not need to worry about it too much. At a previous job in the mid-1980's I sometimes did thermal scanning with an infrared microscope on power transistor dice, and actually saw transistors operating (not just in storage) at over 350°C, dissipating hundreds of watts in a package the size of my fingernail.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Yes, proceeding manually sounds viable to me (reminds me of the darkroom)

That crayon which indicates the reaching of a critical temperature looks good: all you then need to manage is the ramp-up rate. Can probably assume it's linear, and adjusting it by adding thermal mass is a good trick. Alternatively, just cycle the power at some estimated duty cycle.

Also important is a fairly steep cool down phase, so adding a fan to the oven is probably a good idea. A faster cool down rate yields a better metal structure in the solder joint. Of course, you don't want temperature shocks either, so it should be fast but controlled and evenly distributed.

This is all very good information.

So, if one were using a hot plate, that should be removed in the cool down phase preferrably without opening the door. Also, the fan should be sucking hot air out? I would assume a type of sliding door before the fan too.

_________________
65Org16:https://github.com/ElEctric-EyE/verilog-6502

When the solder is still liquid, you'd need a fan so you can cool without disturbing the board. Sucking the air out would probably be good, although it would require a fan that can withstand the heat. Blowing cold air avoids that, but you need to be careful not to blow cold air on the board directly to avoid temperature shocks. As soon as the solder has solidified, you can open the door to speed up the process.

That's what we do

_________________
Andrew Jacobs
6502 & PIC Stuff - http://www.obelisk.me.uk/
Cross-Platform 6502/65C02/65816 Macro Assembler - http://www.obelisk.me.uk/dev65/
Open Source Projects - https://github.com/andrew-jacobs