Above. V3 is based on a lab quality cuvette holder from a surplus HP8452 spectrophotometer. Notable additions include electronics carrier L brackets hand made from 3mm plastic material. After hooking up and with bench testing using the RB to RN calibration obtained by Yang et.al. and computing pH from RN, this unit shows sampling variability of about 3 measurement units which translates to an uncertainty of about .002 pH units. At least from this stand point of repeatability, a +- .01 pH unit is possible with the arduino uno based v2 controller. We want to get this in the lab asap for calibration and this provides the baseline for going forward with a DIY make of a cuvette holder – the next one being V4.
2.3 Bench tests
- The blue 434 nm and green 578 nm LEDs are both down on brightness as measured in the blank shot.
- Sample to sample variation translate to ± .002 pH units or better. This meets the ± .01 pH unit target requirement
- dye takes a while to “settle down”
- ambient temperature of the holder/LED/TSL257 carriers strongly affect readings
Construction notes: 24 may 2017
Cuvette holder: reposition the LED holder for a shorter path length (44 mm) and increase LED intensity. The path length provides a means of tuning intensity readings. Shorter path lengths increase intensity due to beam spread with distance
Controller change notes:
moved green LED to Arduino uno pin 9 and reflashed- determined the problem is not the pin but the 1k trim potentiometer.
bypassed 1kR led series resistor for green 578nm. 1kR remains for blue 434nm.
add two 4-40 bracket stabilizer screws on TSL257 board
wiring harness wiring change to match wire color to led color : blue to blue, green to green.
pot adjustments are too sensitive. green pot adjustment is difficult – ok now, fix needed
optics signal cable male header pins move – OK now, fix needed
Important factor in testing. LED light output decreases with increasing temperature.