So, my LPS33HW sensors arrived today. They are ic2, 3.3v, I can immerse them in 50cm of water depth (which is all I am trying to do), everything I could want but hoo boy are these things tiny. I mean, I knew they were small, but I didn't quite get how damn small.
The pads are spaced .6mm - .35mm of pad, and .25 gap between the pads. I mean - I knew I'd have to get the loupe out, but holy soldering irons, Batman! This is going to take some really, really fine wire.
Good choice of parts.......will surely test your skills.
You will need to get some 30 or 32 AWG wire, pre tin the ends, use a very fine tip, flux (never enough in the core of solder) and a very steady hand. Pre tin the pads as well. Hold the part in a small vice/helping hand and apply as much good luck as you can. If you were here I could have it done for you in a few minutes but we all need to learn.
Once done, clean the flux off with alcohol and then I would test it and then encapsulate the wires and the PCB only in epoxy, the wires will break off otherwise. DO NOT get lux of alcohol in the sensor it self. Use a cotton bud/cotton tip dipped in alcohol to do the cleaning, this will limit the amount of alcohol around the sensor.
There are other ways of doing it but without $'00 worth of tools the above would be the best.
PS. If the joints don't look the best apply more flux and reheat.
Your enemies at that scale are surface tension, and hand muscle tremors. Everything wants to join in one big clump, and the iron seems to have a mind of its own. Plan on talking to yourself a lot, make sure there are no sensitive ears around...
windoze_killa:
encapsulate the wires and the PCB only in epoxy, the wires will break off otherwise
All good tips - thanks. The message I'm getting is it's possible, and the only way to do it is to just do it. I'll head out to Jaycar at lunch for the wire and some flux, and attack the tip of my iron with some fine grit wet-n-dry.
The unint is square, with each side holding 3 or 2 pads. I'm thinking of mounting wires on kapton tape with appropriate spacing and a little bit poking out, and soldering 3 at a time with a single swipe.
I'm dangling the sensors in a fermenter. With one sensor near the top and one near the bottom of the fluid, I should be able so see the specific gravity change over time. When the SG stops changing, the beer is done.
So - not soldered to a PCB, they'll be soldered to a length of wire with 4 conductors (power, ground , SCL, SDA) weighted at the lower end (need a food-safe weight). So, a blob of epoxy, and then perhaps a blob of silicone over that to keep the alcohol from attacking the epoxy.
Next issue - the wire. I'm thinking of using shielded 3-conductor audio cable (ground, left earbud, right earbud, microphone), but perhaps that would be too much capacitance between the shielding and the data line? May have to just try it and see, maybe tell everything to use the slowest ic2 setting.
And after that - a little programming. Once done, I should see a website on my home network with a nice graph of the SG and temperature (because why not?). Depends on how sensitive and noisy these barometers are.
Although the data sheet says "water resistant" it does not state that the device is intended for immersion in water. I take "water resistant" to mean that it can be rained upon, once in a while.
In any case, connections to the device will corrode in very short order if it is powered up while under water, unless conformally coated. The problem will be MUCH worse if there are salts in the solution, as in a fermenter.
jremington:
Although the data sheet says "water resistant" it does not state that the device is intended for immersion in water. I take "water resistant" to mean that it can be rained upon, once in a while.
Fingers crossed. The sheet says that they are filled with some sort of gel and can be exposed to chlorinated water among other things. The main place you'll find chlorinated water is in a swimming pool, so I'm taking that to mean that an intended application is swimming pool filtration systems. They'll also need to be washed with detergent and sanitised with a phosphoric acid solution.
Interesting project. I guess you're going to use the pressure reading to indicate the SG. Will changes in atmospheric pressure be enough to disturb your calcs? How are you going to make use of two sensors? will you use the difference?
I've pondered the idea of using a pressure sensor in the headspace above the brew & somehow monitoring the production of CO2.
Will you monitor temp as well? Will you record? That would be interesting.
For a DFN style package like this I'd use PCB, solder stencil, solder paste, and a reflow oven or hot-air gun.
Some can solder by hand to these packages, but its a trial of skill. Reflow is childs play with the right
equipment and gives very professional results.
My reflow oven is a converted ultra-cheap sandwich toaster oven, and it produces great results.
There's a $12.50 breakout board for this sensor, if that packages OK for you and you're willing to eat the previous cost as a learning experience of sorts.
PCB, reflow, and conformal coating or some sort of enamel over the exposed conductors. Without coating, there is no hope of that holding up underwater.
Nup. Can't do it. Not even close. I think I had three wires attached at one point, but that's it. The slightest hand-shake, the tip brushes one of the other wires, and off it comes.
PullUp:
There's a $12.50 breakout board for this sensor, if that packages OK for you and you're willing to eat the previous cost as a learning experience of sorts.
It doesn't look like I can submerge this. There are a number of other sensors, but they supply a voltage out, not ic2. The ADC on an analog read pin will not be accurate enough.
I tried using an A/D chip hooked up to a voltage out pressure sensor, but just too much noise.
Oh, hang on. STEVAL-MKI183V1 . A breakout board that exposes the 'pipe' of the sensor (which the adafruit does not do), so I might be able to mount it on the outside of the fermenter and connect it with a small hole in the barrel. A bit bigger than it needs to be ... an option might be to scrape the traces to expose them and solder to that. The rest of the board could be cut ... then again, I could just use as-is. I't not going to be that big. Put it in some heatshrink - maybe just get some of that food-safe silicone mould stuff and dip the whole thing, being careful to keep the sensor port clear.
