I looked at a few spec sheets and they state that the Impeedance of these devices are less than a half an Ohm so I don't believe that this poses a prblem.
Have you tried using very large temperature changes to see if there is a change in the inputs?
Do you have more than one sensor you can try? Do they give the same results? If they do I would suggest that it's not the sensors either, and we know that the device works fine also.
So my last thought would be that because the 6008 is a low cost DAQ device and only has 12-bit resolution (and the spec sheets state that the output will only change 10mV/°C), the 6008 isn't the best device to use for temperature measurement and in reality, you may only be able to detect a temperature change of around 2-3 degC due to the large quantisation error.
Please get back to me if you still require some help
Just wondering if you have managed to get anywhere with this?
Did anything I said help?, Post back if you need more help
Thank you for your replies.
I think it was an impedance issue alright between the LM35 and the DAQ. What i did was incorporate an CA3240 op-amp as an emitter follower in to my arrangement (ie output of lm35 into the op-amp, then output of op-amp into the DAQ). This worked properly and perfectly thank god
I've been working with the LM35 and 6008, and found the same problems as you did, I eventually resorted to using differential mode for connection, with good success.
I'm now trying to configure an emitter follower using an LM358 (dual op amp, single power supply package) so I can use single ended terminal configuration, and am getting badly quantized values read in at the 6008 (ie code width of around only 0.01V), although I can see with the meter the 'correct' voltage is being presented to the 6008. I'm not sure if this is still an impedance match problem, but this time between the op amp and the 6008.
Can I just check the configuration you're using for your CA3240? I assume you're connecting the Vout from the LM35 to the non-inverting (+) input of the op-amp, then linking the amp's output back to the inverting (-) input with a wire. I'm running the amp from a single rail supply of around 5V. To my mind, this should provide the lowest output impedance possible, which should be the best situation to present to the 6008.
Please post back if you're using another configuration, or a resistor feedback network, I would be really grateful!