Achievement
Feb. 24th, 2009 10:24 pmIt shouldn't have taken me a week and a half to make this work but I have finally managed to make a thingy. It is, to give it its electronic name, a non-retriggerable one-shot with duty cycles of 5 and 420 ms. What this practically means is that the Gilson liquid handler can talk to the Ocean Optics spectrometer. This is useful because the liquid handler can be programmed to move around so that it holds a couple of fibre optic ends up to either side of a series of cuvettes with coloured stuff in, and repeat this over time, and also close one of its electrical contacts when it moves into place over a cuvette; the one-shot is connected to the contact so when the contact connects it sends the one-shot's output high for 5ms, which is connected to the Ocean Optics and makes the Ocean Optics take an absorbance spectrum of the coloured stuff so I can find out how much it's changed colour. It has to be 5 ms because if it's longer than about 10 ms it might make the spectrometer take more than one spectrum, which would spoil the results because I wouldn't know that it had, and if it's much less than 5 ms the software won't notice than the pin went high. And it has to then be non-retriggerable for about half a second because then the contact on the Gilson has to open again and that causes a glitch down the input which makes it take another spectrum and again it shouldn't do that or it'll mess up the numbers. But this all now works and I get one spectrum per cuvette reading, each of which is saved with an automatically incremented number, and then I get an Excel sheet with a bit of VB in it to load up all the spectrum files in the right order and put them in eight separate graphs and pull out the particularly interesting results at 602 nm and put them all in a table together.
And that, dear readers, is how you get robots to do your PhD for you.
And that, dear readers, is how you get robots to do your PhD for you.
no subject
Date: 2009-02-25 01:23 am (UTC)If I'd had a robot to do my PhD, I might have had better results...
no subject
Date: 2009-02-25 05:09 pm (UTC)no subject
Date: 2009-02-25 08:42 pm (UTC)