MAX31855 component: fails to return temperature and faults "7"

01 Sep 2023 21:40 - 01 Sep 2023 21:42 #279615 by PCW
According to your halscope plot, Its outputing data so that does not
seem like the output has been damaged (its probably clamped to
diode drop above 3.3V by the MAX38155)

Also the halscope plot shows the same repeated data
(if the halscope display was adjusted, you could read the actual data)

Its also a bit odd that inverted data is used. Have you tried
non-inverted data?
Last edit: 01 Sep 2023 21:42 by PCW.

Please Log in or Create an account to join the conversation.

14 Apr 2024 17:40 - 15 Apr 2024 16:47 #298239 by jacczy
Hi travis036,
I have the same problem with max31855 as you have described. Did you figure out what was wrong with your setup?
My hardware is: mesa 5i24, custom bit file (4 step motors, 2 encoders for spindle and spindle cooling test, modified wj200 vfd driver to support Omron MX2/1kHz vfd, ATC, ect... still under construction)
This is my first post here, so hello to every one.
Last edit: 15 Apr 2024 16:47 by jacczy.

Please Log in or Create an account to join the conversation.

27 Apr 2024 20:11 #299205 by travis036
Hello jacczy,
i have all but abandoned using the max31855 directly through my Mesa board, in favor of working on building a Arduino to read temperature, and communicate back to LinuxCNC via Modbus... i just have not gotten around to building it yet.

i have confirmed that the max31855 board will communicate with an Arduino. and it reads temperature correctly.
i have also confirmed that the Arduino can communicate via Modbus with LinuxCNC.

my task will be to build a small enough module that i can mount it on the machine, supply it with 5Vdc, and run a Modbus wire-pair (RS485 RTU) back through the machine, and route that to my LinuxCNC computer. my prototype worked, but was big and ugly.

rather than a official Arduino UNO board, i will be using a self-built with a ATmega328PU, the necessary clock and power filter components, a MAX485 chip, 120R termination resistor, and the MAX31855 board. bare-minimum circuits, and i may possibly POT the circuit in epoxy, where it will ride on the machine, but have not decided on that yet. makes it rather hard to change for a future update...

i was pondering the idea of how to send the temp, a floating-point number over Modbus. i think it allows for an 0-255 integer (byte var), but it occurred to me i can just send the whole-number. i hope my spindle does not reach 255C, as it is only rated for a max of 180C. then, as i am from USA, i use F temp units, and this can be converted once in LinuxCNC from C to F units. it just allows more temp range when transmitted as C units.

Please Log in or Create an account to join the conversation.

Time to create page: 0.105 seconds
Powered by Kunena Forum