Best 3-in-1 Temperature, Pressure and Humidity Sensor

I’m hoping to find a 3-in-1 sensor with good accuracy: temperature error of less than 0.1K and everything else < 3% that would work well with an Arduino or any microcontroller for that matter. I am aware of the BME280 from Bosch but the accuracy is not good enough.

If there aren’t any 3-in-1 sensors that have this high of accuracy, are there any 2-in-1 sensors perhaps? Then I could have the 2-in-1 and the other sensor connected to an Arduino. I believe that will also work well.

Any insight is greatly appreciated.

If there is a 3 in 1 sensor that matches those specifications, it will be VERY expensive. Look at professional instruments and let us know what you find.

For example, this "high accuracy" thermometer (+/- 0.05 degree C) costs $875.

Do not confuse precision or resolution and accuracy. Thermometers with resolution 0.1 degree C typically have error bars of +/- 1 degree C.

jremington:
Do not confuse precision or resolution and accuracy. Thermometers with resolution 0.1 degree C typically have error bars of +/- 1 degree C.

Thank you for the reply and insight. That sensor is just a thermometer so I wouldn't be interested in that but I did really mean an accuracy of 0.1K and not resolution. I need precise measurements. Before posting I did already find for example:

AM2315 - Encased I2C Temperature/Humidity Sensor : ID 1293 : $29.95 : Adafruit Industries, Unique & fun DIY electronics and kits (humidity and temp)

AMS 5915 - Board-Mount Pressure Sensor with I2C Output (pressure)

but due to space constraints the AM2315 would be too big. I thought about removing the plastic and just having the electronics being bare to the room elements but not sure if that will work out well. These two sensors fit also the accuracy requirements well.

As a professional, I have trouble believing the specs in the AM2315 data sheet and would be interested to see how measurements made with it would compare to a professional, calibrated instrument.

The temperature accuracy specification for the AM2315 is maximum error +/- 1 degree C. The AM2315 data sheet does not define the meaning of "typical" error. And, it does not even state the maximum error for the humidity sensor.

So, if you really do need the accuracy you state, it would be foolish to trust that sensor, or one like it. It would have to be checked against a professional instrument.

Very interesting. I'm new to this so I really do appreciate your professional input. Ok so let's say I scrap the 3-in-1 sensor idea. Do you have any recommendations on temperature, humidity and pressure sensors? They need to be small (preferably just a small component that I then with wires connect to an arduino outside of the housing unit they will taking measurements in). Thanks in advance.

I'm happy to advise, but I use professional temperature and pressure instruments, which incidentally need to be calibrated regularly (e.g. annually or better) if they are to be depended upon.

So, I need to understand if you are serious about your specifications. Is this for scientific research? Why do you want sensors to be combined, which severely limits your options? Please describe your application. Terms like "pressure sensing" are too vague to be useful.

I know little about consumer grade humidity sensors, but understand that they have a bad reputation for drift, exhibit very poor and/or sudden changes in calibration and generally need to be reconditioned or recalibrated if they even once encounter condensing levels of humidity.

Hopefully others will chime in too.

Yes this is serious. We are building a new research lab and we have noticed that the instruments/devices that we use heavily depend on temperature, pressure and humidity. So the measurements of these three aren't per se for research but rather to understand how the devices (that are for research) are dependent on the these three variables in the hopes to better regulate all three of them (or perhaps improvise how to improve the values).

The devices that we need are placed into a thermic controlled housing and the space in there is tight hence why we would like PCB sensors. But I'd prefer if we're going to roll out this small project, we should do it right and look for optimal sensors.

This isn't supposed to be a crazy project where we invest a lot of time. It's more on the lines: purchase the sensors, get them running on an arduino, write a small script for the incoming data on a computer and plot the data.

Thanks again for your time.

I would go with industrial grade, NIST-traceable calibrated sensor suppliers like Omega, Thomas Scientific, Honeywell, etc. Many have computer interfaces or SD card storage, so no obvious need for an Arduino.

Edit: Arduino would be fine for an application like monitoring the output of this modestly priced, guaranteed NIST-traceable humidity sensor from Omega. I have no connection with Omega, but in the lab use several different types of sensors I've purchased from them. and am satisfied with the quality and price.