Thermocouple
GrahamCousens
Drew University
in Cyton
Hello. I'm looking for a way to use a K-type thermocouple to monitor nasal temperature. Has anyone tried this before or have an idea how to do it? Is it possible to attach an i2c thermocouple amplifier (e.g., MCP9600) to the Cyton board and feed the signal into one of the front-end channels? Thank you.
Comments
Graham, hi.
Since temperature changes so slowly, there is little advantage in having it tightly coupled to the EEG data stream. An easier approach would be to use a separate Arduino, attached through a 2nd serial port.
https://www.google.com/search?q=read+temperature+arduino
This arduino would periodically read the temp, then output as an ascii text string on the serial port.
Regards, William
There is this other type of three wire analog temp sensor:
https://learn.adafruit.com/tmp36-temperature-sensor/using-a-temp-sensor
In that case you could connect Vdd (3.3V) and ground to the sensor, and run the analog output to one of the 3 Aux channel analog mode pins on the Cyton. Then apply the conversion formulas shown. The Aux data is accessed by turning on Analog mode.
https://docs.openbci.com/docs/02Cyton/CytonSDK#board-mode
William
Thanks so much WIlliam.
I'm using a Max31855 thermocouple-to-digital converter, which works well on my arduino. I can't quite get it to work on the Cyton board.
The converter has the following pins, which I've connected to the Cyton board in this way:
Max31855 Vin not connected
3vo to Cyton VDCC
Ground to Cyton ground
Do to Cyton D11
CS to Cyton CS2
CLK to Cyton CLK
I'm recording with digital read on, and am hoping find some evidence of temperature samples in the datafile.
Graham, this is the data sheet,
https://datasheets.maximintegrated.com/en/ds/MAX31855.pdf
This is an SPI bus device. You will need to connect up the other SPI lines, and use another chip select. Then you will also need to mod to firmware. Sounds like a lot of work.
The Cyton analog and digital modes are for reading a single input pin in either analog-voltage, or digital-level (0V or 3.3V).
William
I don't want a lot of work.
Is there a way to get the temperature readings from the arduino into the data file collected by the Cyton? One possibility might be to use a DAC breakout on the Arduino and feed the voltage signal into an analog channel?
This was the previously mentioned 3 wire device: Vdd, Gnd, analog output.
https://learn.adafruit.com/tmp36-temperature-sensor/using-a-temp-sensor
I noticed that, but the issue is that I need to position it in the nose to measure faint changes in respiration. The thermocouple is a good sensor for this purpose. I really appreciate your comments. Very useful.
You could see if you can find an "interface chip", that plugs to a thermocouple on one side, and outputs an analog voltage on the other.
I guess you are looking for size and form factor. But the Adafruit TMP36 is not that large. Could be mounted on some type of carrier / holder under nose.
If you are really looking to track breath movement, I believe there are some techniques to do that.
https://www.google.com/search?q=arduino+breath+sensor
I'm going to try the way that involves a lot of work. Is there sample code on reading from and SPI device through the OpenBCI platform?
https://docs.openbci.com/docs/assets/CytonImages/OBCI_V3_32bit-Schematic.jpg
Left column mid page shows accelerometer SPI wiring. You can search firmware for how it is read. You need to use a separate chip select line from any other device.
Ok. Thanks William.
Conventional way to read respiration reliably, is with a chest strap that varies resistance with the the amount of stretch. I believe these are also sold on the web. Sounds more comfortable than a thermocouple sticking up the nose.
https://www.google.com/search?q=respiration+chest+strap+arduino
That's true. However, this particular application requires temperature (as a proxy for airflow) measurement from different locations within the nasal cavity. This might differ from chest strap output right after swallowing, which is the period we're interested in.
Thanks again for your help. I'll let you know how it goes.
FYI: An analog thermocouple amplifier (AD8495) worked best for this application, where the highest degree of temporal resolution is required.