search

paulvha / scd30_on_raspberry

Program to set and read SCD30 on a raspberry Pi
GNU General Public License v3.0
7 stars 2 forks source link

Question? #5

Open pi1993 opened 5 years ago

pi1993 commented 5 years ago

I need to make CO2 measurements for 10 diffrent spots and found your project. Thx you. I integrate it and try to test it. It's running, but when I compare the Temperature and Humidity Value, they aren't like my other measurment system. I used a SHT31 (Sensirion) ans saw some differents +-10 RH in H and +-2°C in T. My question, is this a problem from the software? Tomorrow we will take it to place, but i have a bad feeling with this drift...

Kind regards

paulvha commented 5 years ago

A difference in temperature is always. I have not seen a complete match with any project i have done between sensors. 10% humidity looks high to me. I have not seen such a difference. What is the other sensor? Same conditions?

Regards Paul

Outlook voor Androidhttps://aka.ms/ghei36 downloaden

From: pi1993 notifications@github.com Sent: Sunday, January 13, 2019 6:46:23 PM To: paulvha/scd30_on_raspberry Cc: Subscribed Subject: [paulvha/scd30_on_raspberry] Question? (#5)

I need to make CO2 measurements for 10 diffrent spots and found your project. Thx you. I integrate it and try to test it. It's running, but when I compare the Temperature and Humidity Value, they aren't like my other measurment system. I used a SHT31 (Sensirion) ans saw some differents +-10 RH in H and +-2°C in T. My question, is this a problem from the software? Tomorrow we will take it to place, but i have a bad feeling with this drift...

Kind regards

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://nam03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fpaulvha%2Fscd30_on_raspberry%2Fissues%2F5&data=02%7C01%7C%7C0c799911d60a423c5f8508d6797f0949%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636829983850211560&sdata=WNLbQ%2FYDooNSki%2B96un%2BpevZ42%2FaJbLM7Co3SiTqh%2FM%3D&reserved=0, or mute the threadhttps://nam03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAPRgvA6ely9glalrvZzdPb_7_rRm9SHBks5vC3DvgaJpZM4Z9SW1&data=02%7C01%7C%7C0c799911d60a423c5f8508d6797f0949%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C636829983850211560&sdata=IxnoCtzuHHFQ0IvtfhFCRZbvsB2DCSH2VpDI2kVYRkI%3D&reserved=0.

paulvha commented 5 years ago

Just returned home. I have reviewed the code to be sure. The number are directly coming from the SCD30.

I have started the SCD30 AND BME280 (Bosch) AND BME680 (BOsch), after couple of minutes running : Temperature : SCD30 ~23.02 C, BME280 ~21.53 C, BME680 ~22.83 C Humidity : SCD30 ~40.70, BME280 ~47.67, BME 680 ~46,61 %RH Another Humidity meter (external) is showing 46% RH at 20 C temperature So it looks indeed the sensor on the SCD30 is underestimating the humidity with 6%

There is a correlation between temperature measurement and Humidity, but on the SCD30 these are 2 different sensors on the board. A higher temperature leads to lower relative humidity readings The BME280 and BME680 carry the same sensors (the BME680 has a couple more), but still there is a difference. I have never made it to a matching couple.

Given the amount of heat dissipating components on the SCD30 one can expect a higher temperature. With the -o option: -o1 you can correct the Temperature over temperature of 1 *C. This will be sent to the sensor and it will adjust in 10 to 20 readings. This will increase the humidity a bit.. but not enough. Also keep in mind that the temperature and humidity sensors are on the other side of the board of the connector. For a better reading you have to put the board "up-side-down" e.g. the connector to below and actually a small fan should blow the air over the board.

NOw the good news... as long as the humidity is between 20 and 60%, there is hardly impact on the CO2 reading. And to what to expect as CO2 reading look at: https://www.kane.co.uk/knowledge-centre/what-are-safe-levels-of-co-and-co2-in-rooms

pi1993 commented 5 years ago

Hi Paul,

Thx you for your fast support.

I made more test in our lab in the school. We took two scd30 with the RPI in the oven and heated it up to 40°C. The RPI logged datas all 5 seconds. Also we made a referencmeasurment with a Thermoelement.

Thermoelement: 40.1 °C (+-0.1) SCD30: 40.0 °C (+-0.1) Oven internal measurment: 40:1 °C (+- ?)

It showed, that the sensors are working pretty good, but the Halogen lamp from the CO2 Sensor and the RPI are the heatsource. Also when I made measurments for a longer time, the SCD30 looked like a charching curve from a capacitar (Voltage). The endvalue was ~35 °C.

Also we made test with the reaction time. For that we took a coolingspray (CO2) and penetrated the sensor directly. The CO2 Value went straight up to 40'000 ppm and under 0°C the sensor showed no data availabel (like in the datasheet).

My next steps are like you said. Change the logtime from 5 to 30 seconds and maybe make a offsetcorrrection. But for better measurements, we will use other Tempertureelements. Maybe DHT11, because i have it arround me. Last step will be, install a ventilator. But i don't like to do that, because the RPI need the voltage. I think.

Thx you so much. your programm is amazing. I just made some little changes and it works pretty well.

Kind regards Pascal

paulvha commented 5 years ago

Good to see it works.