Following the confirmation that exposure to the ambient IR of the sunlight was going to cause issues with our IR detectors, I found a potential work around in the form of the CNY70 which effectively performed the same task as our current sensor setup but had it's photodiode contained in a leaded case which would, supposedly, help prevent ambient IR from saturating it.
Unfortunately testing revealed that this would not resolve the issue, especially in bright light. The same issue persisted using the same test setup as we did in our group post and turning the sensor away from the light source only provided a minuscule increase in sensitivity, approximately 40mV.
After confirming IR sensors would be non-viable, a set of HC-SR04 Ultrasonic Sensors were ordered for testing. It should be noted that the voltage requirements for these sensors are different than those of the IR sensors. For this, a 5V regulator will need to be implemented to ensure the voltage supplied to the sensor is sufficient. From a previous project, I have a set of 5V regulators that can accept an input between 1V and 5V to produce a 5V output at up to 480mA which will more than meet our requirements.
The HC-SR04 requires a pulsed input to activate the output for echolocation with a pulse width of no less than 10us. A 40kHz signal is then produced by and sent out, the echo pin will then produce a signal with a pulse-width that is proportional to the distance detected i.e. the shorter the distance, the shorter the pulse-width produced at the echo pin.
Initial testing showed a similar result could be achieved wherein a longer pulse at up to 0.5 seconds would result in the echo pulse creating a signal if nothing was detected close to the sensor and would result in no signal is something was detected. Effectively creating an active low sensor. Which option is best will be decided later this week. Show below are examples for each condition.
Lastly for this week, a baseline test of the system was run to confirm all components and code were working under operating conditions off of the PWM charge controller, battery, and solar panel and left from 12:00pm to 5:00pm. Nozzles were attached to each of the vent visors (minus adhesive for the time being). Testing on 3/4/2023 was done keeping track of the temperature in the front and back seat to get an average internal temperature as well as weather conditions, battery level as seen at the PWM, outside temperature, and temperature at each of the four car windows for comparison against the monitoring code written by Ian.
After the baseline test was done, several issues were noted on the UUT. One, the rear window nozzles extend past the vent visor creating a risk of rain - to remedy this the fans and need to be relocated and a set of more shallow nozzles created to prevent those issues. Two, the Thermistor on window 4 appeared to be damaged, likely caused by unintended stress on the wiring pulling against the adhesive mount. Three, according to the monitoring code written by Ian, Thermistor 2 was not providing a signal to the Pico resulting in the inability to change cooling modes throughout the day resulting in a lower efficiency from the UUT - this is either because Thermistor 2 was pulling data from the damaged Thermistor on Window 4 (wired incorrectly during setup) or there is a disconnect somewhere between the thermistor and Pico. Troubleshooting will need to be done this week to determine the root cause.
Temperature data, despite the issues, seems promising. For comparison, temperatures were monitored on 3/5/2023 for comparison with no cooling.
The baseline test showed that an average of 21.8°F of cooling was attained between a cooled and uncooled vehicle under similar testing conditions. It should be noted that on day 2, the overall cloud coverage was greater. Throughout the day, the battery level rarely dropped below 14.4V and only did so when the power provided by the solar panel was not enough to keep the battery fully charged. Checking against our engineering requirements, on Day 1 of testing the maximum ambient temperature reached 89°F meaning that the maximum allowable temperature (at 130% ambient) was 115.7°F. The maximum singular temperature recorded (due to brief sun exposure of the sensor) was 106.5°F with an average of 97.8°F got the internal temperature throughout the day, or to put it another way, the average internal temperature was only 9.9% hotter than the outside ambient temperature.
The next test, once changes are made to rectify issues found this week, will run for upwards of 8 to 10 hours simulating a day out in the sun after a short drive.
Comments