FAST RESPONSE TIME TEMPERATURE SENSOR RECCOMENDATIONS

Hello,

I am working on a project that requires finding a temperature sensor with an accuracy close to +/- 0.02C. I've been searching on Mouser and Digikey and have found multiple that have achieved this requirement; however, most of them have a response time/time constant of 0.5s - 1s. For this project I must find a temp. sensor that has a response time in the ms range (something like 50ms or less). The reason for this is because I need to have this sensor sample at a rate of 20Hz or higher.

Any recommendations????

Josh

MLX90614

jpasion22:
I am working on a project that requires finding a temperature sensor with an accuracy close to +/- 0.02C. I’ve been searching on Mouser and Digikey and have found multiple that have achieved this requirement

I doubt ±0.02C accuracy will be easy. Or do you mean resolution.
Post links to the sensors.
Leo…

I am working on a project that requires finding a temperature sensor with an accuracy close to +/- 0.02C. I've been searching on Mouser and Digikey and have found multiple that have achieved this requirement; however, most of them have a response time/time constant of 0.5s - 1s. For this project I must find a temp. sensor that has a response time in the ms range (something like 50ms or less). The reason for this is because I need to have this sensor sample at a rate of 20Hz or higher.

Unrealistic. Do you know what you're writing about? What kind of material are you testing that changes it's temperature in 50ms? Although this is not impossible, it's very unusual to need an accuracy of 0.02°C in such conditions. So I get the impression that the specifications aren't reasoned.

Response time constant and settling time to 0.02C aren't the same thing, note...

More information about what you are trying to measure please?

jpasion22:
accuracy close to +/- 0.02C
sample at a rate of 20Hz or higher.

Seems beyond the possibility of anything I can find.

You can get speed with a non-contact sensor, or accuracy with a contact sensor.
Maybe use both?

FantomT:
MLX90614

Thank you for the reply. Will look into that sensor.

pylon:
Unrealistic. Do you know what you're writing about? What kind of material are you testing that changes it's temperature in 50ms? Although this is not impossible, it's very unusual to need an accuracy of 0.02°C in such conditions. So I get the impression that the specifications aren't reasoned.

Appreciate the reply. This project that I am working on is a CTD device that can be lowered down by a drone. Our project sponsor gave us these requirements in hopes to create a device that was high in accuracy and low in cost($200-$500). Preferably we would want to use an platinum RTD but since those are generally slower to respond we are also looking into NTC Thermistors. In addition, we are willing to sacrifice accuracy to obtain a sensor with high response time.

Hopefully that clarifies any confusion.

MarkT:
Response time constant and settling time to 0.02C aren't the same thing, note...

More information about what you are trying to measure please?

I failed to mention this earlier but my team and I are working on a CTD device that can be lowered down by a drone. The sensors we require are a temp. sensor, pressure sensor, and a conductivity sensor. We are good with the pressure sensor and cond. sensor. but are having trouble finding a temp sensor. Our project sponsor gave us these requirements in hopes to create a device that was high in accuracy and low in cost($200-$500). In addition, the sensors must be able to operate at a maximum depth of 30m. Also we are willing to sacrifice some accuracy for faster response time.

Wawa:
I doubt ±0.02C accuracy will be easy. Or do you mean resolution.
Post links to the sensors.
Leo..

Lol I'm serious that is the accuracy that I'm looking for.

Here are some sensors that I have found on Mouser. Note that all these sensors have a slow response time.

https://www.mouser.com/Sensors/Temperature-Sensors/_/N-7gz50?P=1z0vvke

ol I'm serious that is the accuracy that I'm looking for.

I don't think so.

I quote the datasheet of one of the more expensive sensors in that list: "±0.3°C or 0.6% of temperature, whichever is greater, R0 ±0.1% trim (optional)". Even the 0.1% R0 will be difficult to get, you also need an ADC with such an accuracy (which means you need a voltage reference of that quality) and at the end you get +/- 0.3°C (after calibration!) which is only slightly better than a pretty standard DS18B20.

So why do you think an accuracy (and not resolution) of 0.02% is needed and why do you think you need to have a response time of 50ms? I'm not aware of a sensor that would be only nearby that values and almost in your budget, I doubt that it exists even if money would be no question.