Hi, I've been running some test on the ZMOD4510 and O3 concentrations never went up from 0, so there's probably something I'm doing wrong...
Our application has to get O3 concentration every 20 mins. The implementation of the firmware is done in a way that performs the reading like it's suggested in the Programming Manual, (table 4), runs the first algorithm calculation, sleeps for 20 mins and starts from step 5 of the program flow (start measurement).
According to table 4, the 2s delay is needed between measurement and status register reading, but is there any timing needed between algorithm ending and starting a new calculation? Can the sensor perform the warmup routine without needing a read of the status register every 2s or so?
Thanks in advance.
Hi,
Thanks for the request. Sorry for the late reply. We are glad to assist you. I have couple of questions.
1. Please tell the which MCU and Compiler you are using.
2. How long you have you tested our sensor and also mention the firmware version that you have been used.
3. End application of your project.
Our ZMOD4510 has a hard-coded stabilization time where the output is suppressed. OAQ 1st gen stabilization will take 1 hour and for OAQ 2nd Gen stabilization will take 30 minutes. Our AI machine learning algorithms are designed for a specified sampling time (+/-5% tolerance allowed). If the sampling time is changed, the algorithm output result will not be accurate, therefore we don't recommend changing the sample interval and also, we cannot reduce the warmup time. When zmod restart, it has to redo the warmup process.
Regards:
R.k
Hi! I'm using an ESP32S3 as my MCU and I've been testing the ZMOD with the 2nd gen algorithm. I've been polling every 5mins in 24hs, monitoring the values of the Rmox, O3 and if it left it's warmup state (which it didn't, but I did get Rmox values over I2C, so there isn't a communication issue).
The polling algorithm I've used is the same as the one in the programming manual, with the added step of sleeping for 5mins AFTER results have been logged, NOT between starting the measurement and processing the results (which does have the 2secs delay as the manual suggested).
Is it necessary for the algorithm to start the measurement right away, after results have been processed? Do you keep the warmup time/flag in the internal registers of the sensor, or is it something that's being tracked in the MCU which could be lost when it goes to sleep?
Thanks for sharing the details. No, we have to strictly follow the time interval of 2s. If you sleep for 20 mins, our algorithm cannot report a better value. Our sensor and algorithm are not designed for such "on-demand" application.
Raja.