Retired Product

This product has been retired from our catalog and is no longer for sale. This page is made available for those looking for datasheets and the simply curious.

Creative Commons images are CC BY 2.0

Description: MAXIM IC MAX6675 Cold Junction Compensated K-Type Thermocouple to Digital Converter. This simple 8-pin SOIC IC attaches directly to any K-Type Thermocouple and interfaces via a SPI read-only interface.

Works great with 8-pin SOIC to DIP adapters.

  • 0.25C Resolution
  • 0C to +1024C
  • Minimal 2-wire digital interface


Recommended Products

Customer Comments

  • Gnnh. I needed one that goes past 1700C for accurate temp measurement for melting titanium.

  • This is a great chip for thermocouple applications. I had used this years ago for a project. I had documented my work with it, which some may find helpful. Check it out here:

  • I was really interested in the tutorial link, but it is broken!

    I found another one here, in case Michael’s tutorial is permanently gone:

  • So….Is there any reason we couldn’t use an E-type thermocouple, and use the reported Celsius value to backcalculate the real Celsius value?

    Maybe I don’t follow exactly what’s going on.

  • would this work with Serial Multiplexer Breakout - TS3A5017
    (sku: BOB-08970)
    Has anyone tried this? I would like to use 4 thermocouple probes, but avoid expense of 4 of MAX6675 chips.
    Is there a BOB board available for this chip?

    • In theory I think may affect the cold junction compensation a little, as would having long tracks between where the thermocouple terminates, and where the IC is located (as it measures temp on the IC to compensate)
      I see no reason why it would cause a problem,
      in particular the amplifier is likely Hi-z, and the source impedance of the thermocouple is low, like 1R ~ 10R depending on the particulars.
      offsets introduced by using the switch (if there are any) would be the thing to look out for.

  • Has anyone tried using these on a shared SPI bus?

  • There’s a nice Arduino Max6675 Library available to really simplify reading temperature data from this amplifier.

  • Could I use a multiplexer with one 6675 to read from multiple thermocouples, or do I need a 6675 for each thermocouple in my project?

    • No, Use separate chips and use common ck and DO lines and then have multiple CS lines. Works great for me.

Customer Reviews

2.5 out of 5

Based on 2 ratings:

Currently viewing all customer reviews.

2 of 2 found this helpful:

Handy little chip

I’ve had good luck using this chip, particularly when I compare it to the AD595-AQ.

What I like:

  • Handles amplification and ADC on the same chip. I don’t need to worry about analog signal routing, or any of the complications of ADC accuracy on the µC.
  • Really simple pinout, therefore a simple board layout.

What I don’t like:

  • The MAX6675 Arduino library isn’t suitable for state-machine-based µC applications. I needed to write my own low-level SPI routines due to both timing and SPI compatibility issues.

ADC timing gotchas, :

  • Be aware that since the ADC conversion starts upon taking CS high, it might be wildly out-of-date if, say, you’re only reading the MAX6675 every minute or so. Toggling CS low, then high can get around this.
  • If you take CS low during a conversion, it’ll stop the conversion that’s already in progress. I seem to recall it’ll just return the previous value in this case. So, don’t read faster than about 4Hz.

I’ve used this successfully in a project which uses:

  • 5V supply
  • ATMEGA328P-AU Microcontroller
  • Other SPI peripherals on the same bus
  • Maximum 4Hz read rate

Worked for a while...then self-destructed!

Not sure if I just got a bad part or if there’s a larger quality issue here, but I attempted to use one of these devices in an oven control with a standard type K thermocouple. Power was 3.3V.

The device worked perfectly for several hours including multiple power cycles for testing purposes. After those few hours (fortunately during a test cool-down cycle) suddenly the temperature readings went to zero, and on subsequent power cycle the chip literally exploded (!) destroying the attached ARM controller circuitry in the process (probably due to shorting out and applying either 0V or 3.3V at high current to the delicate SPI pins). I have no idea how this is possible besides a manufacturing defect; the 3.3V supply was rock stable throughout and the ARM device on the same supply didn’t blink until this chip exploded.

I have been in electrical engineering/embedded systems for well over a decade so I know all about ESD precautions, etc. This is the first time I have had a non-Chinese sensor chip do anything like this!