Measuring temperature - looking for rocketworthy thermometer

[Funkworks]

I want to measure the temperature as function of altitude, known as the lapse rate in aviation. Measurements would be taken on the way down under normal parachuting conditions, so sampling rate doesn't have to be terribly high. Depending on cost, I'll consider different rates, accuracies, and perhaps real-time feed to a laptop. But yeah, mostly something small, light, with a high sampling rate and logging.

Surely plenty of people have done this before, so what products have proven themselves well-suited for rocketry?

 

[Nytrunner]

Thermocouples are lightweight, although the DAq to go along with it may be bulky

 

[UhClem]

Even though you will be measuring on the way down you need something that is small enough so that it has a very short thermal time constant. A thermistor is perfect and is an absolute sensor. You will also need an ADC with high resolution since the changes will be small.

Omega has a wide selection of sensors.

 

[cerving]

The best way to go would be an analog thermistor coupled with an ADC. The digital ones don't have the same response rate... depending on how accurate you need it to be over time, that may be a big deal. The ones built into baro sensors tend to have a really slow response rate, so if you're looking to do an accurate pressure altitude correction (using a known baro reading at the launch site and a responsive temperature correction during flight) you're gonna have to go analog.

 

[Speaknoevil]

Stratologger CF/SL100 have temp measures built in.

 

[Funkworks]

Stratologger CF/SL100 have temp measures built in.

So does my EasyMini, but I also agree with:

... The ones built into baro sensors tend to have a really slow response rate ...

Also: not yet sure where to find a satisfactory ADC (analog-to-digital circuit) for either thermistor or thermocouple.

 

[OverTheTop]

Betatherm do some good high-accuracy thermistors, with fast response times.

 

[BEC]

If I wanted to do this, I'd use an Eagle Tree data logger and one (or two or three) of their temperature sensors, which are just thermistors). The data are then downloaded after the flight into a Windows computer and can be viewed, plotted and even "played back".

https://www.eagletreesystems.com/index.php?route=product/product&product_id=54

I've used several generations of these in RC airplanes with good results. Even thought it's set up with the ability to measure high currents, you can power one with a small battery (say a 300 mAh 2s LiPoly) and have it do its thing.

My most complete installation of an Eagle Tree logger is in my Mountain Models Switchback Senior. It was very useful in gathering much of the data in this article: https://www.rcgroups.com/forums/showthread.php?740824-Mountain-Models-Switchback-Senior-Review

 

[jlabrasca]

https://www.arborsci.com/pages/pocketlab-voyager

Looking at the specs, the include a data rate fro pretty much everything but the thermometer.

a very short thermal time constant

How fast is fast enough, I wonder.

A sensor with even a small thermal mass is going to take some time to come into equilibrium with the ambient. The fastest sensors I've used for thermal physics labs have response times measured in tenths of a second. These sensors are small, nevertheless, for temperatures resolved to the degree (two sig figs) they typically take more than a second to settle down.

https://www.pasco.com/products/sensors/temperature/ps-2135

 

[Adrian A]

Fast times are important, but in case it hasn't been clear so far, the sensor has to be sticking out into the airstream and not inside the av-bay or even attached to a fin, or else you will be taking the temperature of what it's attached to and not the ambient air temperature.

 

[K'Tesh]

Ya know... the sensor package for the tornado rocket used by Reed Timmer might be of interest to you. I have ChasinSpin's contact info somewhere if you're interested. However, they're not cheap (IIRC they are >$100 each).

 

[Speaknoevil]

Gotta love purple PCBs!

 

[g.pitts]

Gotta love purple PCBs!

OSH Park "trademark"! Not literally, of course.

 

[Funkworks]

[...] https://www.eagletreesystems.com/index.php?route=product/product&product_id=54

I've used several generations of these in RC airplanes with good results. Even thought it's set up with the ability to measure high currents, you can power one with a small battery (say a 300 mAh 2s LiPoly) and have it do its thing. [...]

Yes! I will seriously consider this! Didn't realize RC enthusiasts would have a need for sensors.

https://www.arborsci.com/pages/pocketlab-voyager

How fast is fast enough, I wonder.

A sensor with even a small thermal mass is going to take some time to come into equilibrium with the ambient. The fastest sensors I've used for thermal physics labs have response times measured in tenths of a second. These sensors are small, nevertheless, for temperatures resolved to the degree (two sig figs) they typically take more than a second to settle down.

https://www.pasco.com/products/sensors/temperature/ps-2135

I guess 1 second would do. To verify accuracy, I could compare the measured temperature when it lands, with the actual ground temperature from a ground thermometer. Pasco is a great source, never would have thought of them.

Fast times are important, but in case it hasn't been clear so far, the sensor has to be sticking out into the airstream and not inside the av-bay or even attached to a fin, or else you will be taking the temperature of what it's attached to and not the ambient air temperature.

Measurements will be on the way down so I think that can be solved pretty easily, e.g. outside a bulkhead vs inside the ebay.

Ya know... the sensor package for the tornado rocket used by Reed Timmer might be of interest to you. I have ChasinSpin's contact info somewhere if you're interested. However, they're not cheap (IIRC they are >$100 each).

Why yes. Yes it is of interest to me. I miss electronics and need to start packing more of it in my projects.

 

[K'Tesh]

Ya know... the sensor package for the tornado rocket used by Reed Timmer might be of interest to you. I have ChasinSpin's contact info somewhere if you're interested. However, they're not cheap (IIRC they are >$100 each).

Why yes. Yes it is of interest to me. I miss electronics and need to start packing more of it in my projects.

PM me your email address and I'll see if I can make the introduction.

 

[Funkworks]

If they're selling it, wouldn't they have a spec sheet? Website?

 

[jlabrasca]

I guess 1 second would do. To verify accuracy, I could compare the measured temperature when it lands, with the actual ground temperature from a ground thermometer. Pasco is a great source, never would have thought of them.

What would you be measuring if the sensor has to be resident at one temperature for ~1 second for it to come to equilibrium, but the temperature of the sample is changing continuously?

You will get data, to be sure, but I am not sure I'd know how to interpret it. Temperature near apogee might be a reliable measurement -- but teasing out the temperature versus altitude during the descent seems like its going to take some math.

I am guessing that it is a problem that has been solved, though.

 

[BEC]

Yes! I will seriously consider this! Didn't realize RC enthusiasts would have a need for sensors.

Those of us who were electric airplane enthusiasts back when that was hard to do wanted all kinds of data on how our aircraft and their power systems performed. There are a couple of graphs in the Switchback Senior review I linked to from a bench top test rig that I also used when I was evaluating power systems....

 

[K'Tesh]

If they're selling it, wouldn't they have a spec sheet? Website?

I was in communication with them and Curtis Brooks about their collaboration with Reed Timmer. The subject of the price came up. I don't know if the sensors are available for general sale. Hey... ChasinSpin is a member of the forum here. I have a design for a tube launched version of the tornado rocket that I proposed to Curtis and to Reed, but they seem to not be interested in it. I'm actually pretty worried that he's going to get himself electrocuted when a lightning bolt hits his launch rod. The design I came up with was designed to launch up to three sensor packages with different descent rates to give them multiple information streams on any successful deployments, and preserve the CG/CP of the original design.

 

[Sooner Boomer]

A lot depends on how fast your rocket is coming back down. If you're coming down (relatively) fast, a thermistor would be best. I've had my eye on these for a similar project: https://www.bgmicro.com/ACS1280.aspx Very low mass, inexpensive.

Adafruit has a bunch of temperature sensors and interface boards (thermocouple). They have digital temp sensors so you don't need an ADC. They're slower responding than the thermistor, so the rocket would need to come down slowly.


https://www.adafruit.com/category/56

 

[dmo]

If you want to do this, you should take a look at how the professionals do it. They use an RTD sensor. Basically just a piece of Platinum wire in a low mass, thermally isolated frame. Here is a peek inside a modern, hackable, radiosonde:



You can buy used RS41 radiosondes on e-bay and there is information on the WEB for tracking/modifying them.

 

[OverTheTop]

RTDs work well but rely on quite low resistance wiring in the sensing circuit. Make sure you get that right or you will have offsets to deal with.

 

[RoadRunner]

I was thinking of trying these sensors from tiny circuits.

 

[cerving]

Most of the digital temperature sensors such as the ones on those breakout boards are designed for high/low temperature alerting, i.e. your notebook computer is getting too hot, or for low-response environmental monitoring. Their response rates are either not specified or are not very high. You're better off with a separate thermistor bead and a DAC if you need a fast response rate.