Looking at the spec sheet for the LM35 shows that you may need some additional circuitry if you want to use the 6008 to measure the output. I've never used the LM35 but a quick google circuit search shows that most circuits seem to have the output terminal shunted to ground through a resistor. This would probably be for reading with a DMM or other high impedance input. That brings up another concern in that the 6008 input impedance is spec'd at 144 KOhms. Not very high at all. You may need to add a buffer of some kind like an op-amp voltage follower to the output of the LM35 and possibly the resistor to ground as well. This would help eliminate the effects of the measurement circuit; either the 6008 or a DMM.
Well, I haven't got it to work =( Have you? Please let me know... I think the resolution using 6008 is to bad with thermocouple. I connected the cables to the analoge input but it will not work.
The 6008 is not a good choice to measure a thermocouple. A K type thermocouple has a sensitivity of around 40uV/C. The +/-1 V range on the 12 bit 6008 means that the sensitivity of the 6008 is about 490uV, and its accuracy is only 37.5mV. In addition, to take a thermocouple reading, you must have the cold-junction temperature. Unless you will put the junction in a bath at a known temperature, you will need to measure the temperature of the bath.
