Does anybody know how long after I apply solder paste to a board I have before it goes bad and can't be reflowed? I've read about printing and abandon time but that's related to the stencil, and see numbers for things lie overall shelf life, but can't find out how long the paste is good for once it's on the board.
I've got to make 15 boards that take about 40 minutes each to populate (384 SMD components) and am trying to plan out if I can print, print, print, place, place, place, bake or if I need to print, place, bake, print, place, bake, etc.
It tends to "dry-out" fairly quickly. I don't think that's a problem once the parts are placed... it's still going to melt & flow (or re-flow). But it should be in it's paste-state when you place the parts.
If your oven is automated or you set timers, you can get started stenciling and populating the next board while the previous one is baking.
Helps to have two people. We did a run of 50 boards of similar complexity, by the end we were down from 50 to 30 minutes placing the parts
Pre-organizing the parts is also a good idea. We made color coded component maps, and laid out all the parts in matching color coded circles on sheets of paper. Get a big pack of colored markers.
Many pastes list how long they are good for. As I recall, some of the fluxes, etc. are somewhat evaporative, so the best approach IMO is to get organized, have everything handy, and start by stenciling, then placing, then baking. I do my boards individually, that is I do not try and economize on the bake cycle by running multiple boards behind each other. A toaster oven heats quickly...
Thank you for that link. My method so far has been:
preheat oven for about 3-4 minutes as I'm finishing up with placing the SMD parts
place board in oven
Bake for 4 minutes at full blast
Turn oven off and let it sit for 1 minute
Open door and start work on the next board, so it cools over about 20 minutes
I'm using lead-free solder and have done about 20 boards this way. The solder gets shiny between 3:00 and 3:30, and once in a while I use a piece of temperature crayon:
So far everything I've done this way has worked; the only problems I've had are a couple of FT232RLs with bridges (I think if I went to a 3 mil stencil that would solve that problem).
I do need to build something to control the oven -- I was thinking of something very basic with just 2 buttons (preheat for 2 min, bake for 4 min), no other interface.
I've read a few web pages that say this is all that's necessary, but it looks like there's a lot of people making their own controllers too...so now I don't know what to do...I would appreciate any insight / opinions you have.
The timings we use are dependent on the board, but a typical run would be put the PCBs in a cold oven, set the thermostat to 300F, and let it run until you hear the element click off. Then set the temperature to maximum, run that for about 90 seconds, then turn off the oven and leave the door closed for about 2 minutes to cool down. Of course that's lead solder, and depending on how big and how many PCBs, timing can change by 200%.
I have the rocket scream controller + relay but I have yet to hook it up. One of the things I want to work on at the same time is to improve the shell of the toaster oven, current designs have gotten so cheap as to omit insulation in many areas, such as the back surface. I have the "right stuff" for the job, 2000+*F rated material, but I have to find the time to mount it.
Then, there is the question of how to hook up the Rocket scream controller + SSR, I am inclined to do it in an external metal enclosure that can act as a heat sink for the SSR as well. But if you go external, how to hook up the two elements inside the toaster oven (fan and resistance heaters) that the controller controls the temperature of? I'm currently inclined to use two small extension cords - cut & color code so as to make hooking the two things together relatively idiot-proof.
