Multiplexing the MAX31855?

[Grimace06]

Hey Guys-

I'm looking to monitor 4 K-type thermocouples and output the data to a screen. I've been able to interface one MAX31855 with no problems as an exploratory exercise. I understand that it would be VERY easy to do 4 TC's using 4 MAX31855 boards with a common clk. I'd like to know if it would be possible to use 1 MAX31855 in order to read each of the 4 TC's and store the values in a variable for display later on.

Can you use some sort of shift register IC in order to feed different TC leads into the MAX31855 at a set clock rate? Has anyone done this before or should I spend the 60 bucks and get 3 more MAX31855 boards?

I'm just a lowly Mech E with sub-premium coding skills, please help!!

-KDG

[adafruit_support_bill]

The output of a thermocouple is in the microvolt range. You can't switch them with a logic-level device like a shift-register. For accurate readings, It needs to be connected directly to the inputs of the amplifier (Max31855).

[Grimace06]

Thanks for the quick reply! I understand the ADC limitations of other devices, but it seems like there should be some way to interface four thermocouples using one MAX31855 chip. If there was a way to switch the input delivered to the chip at the same clock rate of sample by the Arduino then you could have a seemingly infinite amount of channels!

My problem is that I am unaware of a solution to achieve the input switching on the thermocouple end of the MAX31855 board. Is there a way to make that happen, or is the voltage so small that it really is the best plan to have a dedicated amplifier for each TC?

Thanks again!

-KDG

[adafruit]

you must use one amplifier per thermocouple

[bbarenz]

They can be multiplexed quite effectively using a CD4066 Bilateral Switch . I recently designed ( and am using) a circuit that has 4 inputs with a common - . The TC + are switched on the CD4066. I'm using a microcontroller to switch the input and take a sample. There is a downside in that the sampling rate is reduced. The fastest sample rate is about 100ms per sensor. Some of this is due to the MAX31855 averaging the samples together which takes time to clear out. You maybe able to switch the power to the MAX31855 to reset it to increase the sample rate but I haven't tried it. If you sample too fast you will see some jitter between all the values when one changes. Schematic below.