Notes on Equipment:
It is important to note that the design has been driven by the need to MEASURE SMALL CHANGES in conductivity, rather than accurate absolute values. Thus the presence of the two sensing cells, rather than a reactive reference.
Good accuracy can be achieved with (much more expensive) commercial instruments, although at some cost since platinum electrodes are required to give accurate values & good sensitivity/stability at the high conductivities considered here. The kit presented here could easily be calibrated for absolute accuracy (which would be moderate) but this has not been done.
The apparatus is very much DIY as will be seen in the following photos. [Equipment Design & Photos] The aim was to get a working prototype system & maybe smarten it up later. Almost every item was purchased from RS Components & a few are discussed below. The prototype now works very well. I hope to produce a PCB quite soon.
The thermostatic box is an ordinary sealable sandwich box (10 cm x 15 cm). The polystyrene was from received packaging & the main block hollowed out to take the box.
The fan, which cost about £12, is 4 cm in diameter, 1W, 12V. This gives an air circulation rate of ca. 1 per second. Some work was done on the flow geometry & this showed that the two thermistors (10k) [one for control, one for sensing/display via Dr DAQ] are best placed, close together, about 1.5 cm from the fan on the outflow side, at the central axis of airflow.
The lamps (2 x 10w, 12V; 1 x 35w, 12V) are mounted approximately at the end of the flow baffle, in a triangular format to get maximum interaction with the air flow.
The thermistor for observation/recording of the temp inside the chamber is connected to one of the 12-bit EXT connections of the DR DAQ, using the 5V supply as the voltage source.
The Pico Technology DR DAQ at £90, is much the most expensive of the items used in construction. It has proved invaluable during development as it acts as a scope and a pH meter as well as a data logger.
It would have been good to be able to use a mobile phone with a simple 12-bit A/D converter but this combination does not appear to be available in any simple format yet. Android does not seem to support scientific programs, so data analysis would anyway have to be done on a Tablet or PC.
The RS commercial carbon electrodes described previously on this site have now been replaced with DIY LOW COST electrodes. This saves ca. £50 on the cost of the apparatus.
The 10-turn pot (& digital readout) used to balance the bridge can be replaced with a simple pot. The balance position is not used in analysis of the results. However it is reassuring to see that the balancee position is near the mid-point of the pot - otherwise something is amiss!