Need to do some SMD soldering? No tools? No problem! Here’s a creative method that could be a handy tool to add to your belt: SMD soldering using hot sand.
[Oliver Krohn] recently released this little video demonstrating how to perform re-flow soldering using hot sand. He’s using a bunsen burner to heat up a ceramic pot of sand to use as a kind of hot plate. It seems to work pretty well, and it’s a very unique way of doing it — if you wanted to get a bit more technical, you could also throw a temperature probe in the sand to get a much finer heat control!
Of course there are lots of other ways of doing re-flow soldering, like using a re-purposed toaster oven, frying up some circuits on a skillet after you’ve had your bacon, or if you want to be fancy, you could even build your own toolkit for it!
Anyway, stick around for the epic video of SMD soldering on hot sand.
http://www.youtube.com/watch?v=OtrdsLLYAIM
Doesn’t that just get you pumped to do some soldering?
[via Hacked Gadgets]
I’d be concerned about potential impurities or possible contamination, but the idea has merit, and perhaps could be extended to something more homogeneous and neutral (perhaps sifted pulverized granite – or maybe salt?)…
I’m certainly interested. Kosher salt maybe? Table salt usually has iodine. What would the iodized salt do if you used it? Actually I never heated salt up that way, would it even work?
Optometrists sometimes use heated salt to soften eyeglass frames for fitting and adjustment. They use salt over sand because it doesn’t scratch the lenses. Not sure about the iodized vs non-iodized question though.
The equipment they use for this doesn’t get hot enough to melt solder but I don’t see why it wouldn’t work with a hotter heat source. NaCl melts at 800 deg C which is well above solder’s melting point.
Just use clean sand that’s been sifted to a certain size. The granules will be too big o rest on top of the board without being noticed.
I’ve used ordinary sand and gravel up to 800 degrees f with no problems, just rinse off the sand first.
For the (PID) temperature controlled version, you could dispense with the Bunsen burner and embed power resistor(s), 3D printer hot-end heater(s), or even a custom nichrome heating element in the sand along with the sensor probe(s).
This reminds me of how I once opened up a classical iron to see what was wrong, and I found inside it had a tube on the sole and inside that tube was sand and a nichrome wire, so basically the same concept, the idea I guess was to isolate the wire and stabilize it and to spread the heat evenly.
Same thing with old time porcelain fuses for mains, they also have a wire in it that melts at a certain amperage but it’s suspended in a small amount of sand.
This is genius! I have a leftover heating element that just found a home.
Only thing it really needs is some sort of stand-off to the side to rest your hand on to steady it, I could see the urge to rest the side of your hand on the lip of the hot plate as being almost automatic…
So, Everyone completely missed the point of this project?
I certainly do. Sand is an insulator.
This is called a sand bath, and it has been common for many thousands of years. “OMG LOOK AT THIS COOL NEW THING :D”
Anyways, the idea is that the sand is made of silicon dioxide {glass} primarily, and isn’t reactive towards your board. An even MORE easy homebrew way of doing this is to place a layer of sand, a quarter inch is usually plenty, in a frying pan and to place this on a stove or hotplate or what have you. Nice, even heat along your surface without fancy schmancy instrumentation. Just right when you want a quick little reflow action.
Not like there’s a shortage of even cheaper and equally bad soldering processes.
The ‘new’ part is using it to solder SMD.
People have used toaster ovens too for a long time before we started to reflow SMD in them.
First heating up the solder until all the flux is deactivated, and then trying to put the part on is a good way to get bad solder joints from all the oxides forming.
Seriously, it only takes a minute to hand solder an SOIC 16. First tack a corner pin, adjust, do another corner, adjust, and then do the rest. Especially with the really thin solder wire shown in the video, it’s really easy.
+1
Exactly. Interestingly enough I posted a comment on the Hacked Gadgets page about the lack of flux but they never published it.
It’s sort of useless to preach reason on this site. It’s a bad and a good thing at the same time.
I was really surprised that this video becomes so popular :D Had to solder this for a project and did not have a fine soldering tip so I had to improvise and tried it this way.
At home it would be easier and safer to use an electrical heat source. I do not think salt will work well, sand is fine.
However, I just started my YT channeI and I think I have more interesting projects than this ;) For example I made a so-called skinner-box for a diploma thesis in biology. Therefore I used a elotouch touchscreen and an Arduino Uno to registrate a chicken’s picks on the screen. I uploaded to videos regarding this project, but I promise I will do a third, better and more interesting video of this project to finalize it. In a couple of weeks.
>did not have a fine soldering tip so I had to improvise
A fine soldering tip is the last thing you want when soldering fine pitch parts.. flux is your friend. 5 minutes on google to find out about drag soldering would have saved you a bunch of time and given you a better result..
well, the whole process did not even take 5 minutes. I don’t have to do this regularly, but when I have to I will remember ;)
The whole process still requires keeping the board and parts around solder melting temperature far longer than you’d want, especially considering this board only has one part. What if you need to solder 15 parts ?
A better way would be to put solder paste on the cold board first, then add all the parts, and then set the board in the sand, and heat it just long enough to melt the solder. But really, unless the board has really tricky parts, or you need to make a dozen of them, hand soldering is usually easier and quicker.
I can see this process more useful for removing SMT parts with exposed thermal pads or very large leadless packages, but I won’t reuse the parts handled this way as they might be damaged by the prolong heat damage.
Why not sand in a skillet or lava on a radiator or hot rocks in a bathtub once you at it ?!?
One of the big thing that SMT introduced was mounting parts on both sides of the PCB because you don’t have pins poking parts under it! That’s not going to work here. :P
Probably also need a stead hand to remove the PCB as soon as soldering is done or you want the long wait for the hot sand to cool down!? The same steady hand could have done the soldering in the first place.
Hint: look at the reflow profile to see how long the actual time the part is exposed to *different* rising rate and peak temperature. Engineers don’t do this for no reason.
I found that using a switchmode circuit and a diode as a temperature sensor and holes in a 10 mm plate for low value round carbon resistors (as heaters) worked very well. The switchmode pulse width is wide when getting up to temperature, and narrows as the temperature needed 200C is attained. When a cold PCB is placed, the Pulse width widens to correct the temperature. Soldering was easy using flux and tin lead solder on individual components. temperature was not too high and the soldering iron provided the last bit of heat to solder. the components did not slide off as the plate temperature was low enough for the solder to be a solid. also there was no real burning of the hands as I made a surround pf hardwood on which to rest the hands. To make the PCB stay in place I used Cynoacrolate glue to glue 3X3 mm blocks to hold the PCB steady The flux did not oxidise and was easily cleaned.off.