Just made a batch of 2 TUTs for Howard Rush. I thought I'd toss out some pictures.
This is a revised Howard-style TUT. I had made some changes to the schematic of the big TUT that I wanted to retain, and I wanted to get all of the parts onto the top of the board for ease of assembly -- which I have, at the cost of leaving off the USB connection. Howard doesn't use it anyway, so it's just extra weight, cost and trouble.
First is a picture of my reflow skillet. $5 skillet from Goodwill, $20 thermocouple meter from eBay, and a sheet of glass from a friend who was moving. Getting the time/temperature profile right takes some attention, because it's not automatic. It's also a bit slow going from preheat to the liquid solder stage -- I'm just living with that part.
Second picture is of the TUT with the parts placed on the solder paste, but not cooked yet. I sprang for a solder stencil for this one, and I'm glad I did -- with a stencil I can do proper reflow, which the packags for the accelerometer and gyro demand. It's also quicker, which is nice.
The red circle is a little pile of solder paste. I like to put some on an open spot on a board, as a check that the thermometer is correct, and to get an idea of how even the heating is. And -- the paste's binder melted when it should have, at around 150 degrees C, and then the solder melted when it should have, at around 220 degrees C (it's lead-free solder, if anyone's paying attention).
Third picture is the two TUTs in the skillet, cooling down after hitting a high temperature of 243 degrees C. It's foggy because I put a plate of glass over the skillet so it'll heat up evenly (and enough!), and the glass is a bit dirty. I had some reflow failures on earlier TUTs (Howard, if you're reading this, this is probably why you have a bad gyro on your remaining TUT). It turns out that I wasn't soaking them at high temperature for nearly long enough -- you're supposed to hold them in the liquidus phase for 60 to 120 seconds; I had been just barely hitting the point where the solder balled up and then saying "oh no, I'm gonna burn it!"
Fourth picture is the two new TUTs plus the one I've been doing software development on (the new circuit called for some tweaks to the software). There's one through-hole part on the board -- the connector on the right. I haven't been able to find an equivalent connector in surface mount, dangit.
Bon appetit!