The good news: I spent much of my time in lab watching the Eurocup this past week. First Spain barely beat Portugal by PKs, and then Italy defeated Germany to reach the Finals. On Sunday, Spain demolished Italy 4-0 to take the title.
The great news: I kind-of-sort-of blew something up in the lab.
I had wired the "switch" circuit I mentioned last week (i.e. the opto-isolator+triac setup that separates main voltage from control) on a breadboard, and it was running just fine. So I soldered it onto a circuit board to make it more permanent. It took some tries since the wiring was a bit tedious, and one of the triacs was actually fried. Finally we connected the switch to the wall, all one-hundred twenty Volts and fifteen Amps, and nothing terrible happened. But the moment Ross turned up the voltage to the circuit (which simulates the voltage provided by the temperature controller), the circuit jumped off the table a few inches, let out a great big "bang," and filled the room with smoke and the nice burning smell of electronics. I could hardly control my elation. It was a success. My summer experience was fulfilled (I am only half joking). Ross had said on the first day, "If you don't set fire to this lab by the end, you're doing something wrong." Well, the prophecy had been fulfilled. On the clean surface of the Maple-wood table I had left a mark for eternity: a black smirch that stuck out like a sunspot on steroids.
So that was the highlight of the week, and perhaps this internship.
Anyway, there was something wrong with the board I used. We're not quite sure, but the metal tracks between the triac and resistors were completely vaporized. Maybe a short somewhere. So we used a different kind of board, and this time around the test worked, although uneventful.
The MUX chips also came in, so I got to fiddle with those. It's actually pretty straight-forward. Since I ordered 8:1 16-pin DIP chips, that meant that there were up to 8 inputs being selected for a single output. The number of selector pins is log base 2 the number of inputs, which is 3 (because 2^3=8). So there are 3 selector pins, which the manufacturer had labeled A0, A1, and A2. To select which of the 8 pins you want to readout, you consult the truth table on the datasheet. So for example, if you want to readout the LM35 input on S1 there is a corresponding binary code e.g. A0=1 A1=0 A2=0. In other words, A0 is connected to +5V and A1 and A2 are connected to ground. Then you can read a voltage from the output which has a linear relation to temperature, i.e. 0.208V=20.8 degrees C. Each readout has a unique "code" attached to it (101, 110, 001, 010, etc.). That way when I connect it to the ADC and the computer, I can change the combination of ones and zeroes to get a readout for whichever sensor.
I also drilled the rest of the holes for the LM35s since we got the new 43 drill bits and 4-40 taps. And I spent the rest of the time tapping/threading the screw holes, which is monotonous, tedious, and menial, and I thoroughly enjoyed it ("It's good for the soul" -Ross). I only broke one tap this time. It was not too bad since I could watch the football match the same time.
On Friday I just started to attach all 32 LM35 temperature sensors to a copper plate so we could calibrate them. Now soldering multiple wires onto the little electrical leads should be an interesting task. It should be fun.
No comments:
Post a Comment