Post Go back to editing

LTC2983 development board, occasional problems reading the K thermocouple

Category: Hardware

I have the DEMO CIRCUIT 2026C connected to 2209A LTC2983 board and that is connected to the 2210A experimental board.  I have the system setup as follows:

I have the input of the thermocouple grounded and its connected to either a K-TC simulator and sometimes to a test block which I can heat up.

What I am seeing is that the K thermocouples readings sometimes are not correct.  Please see the Excel spreadsheet showing the captured files.

Most of the time the program is reading 149C.  In cells 1672, 1669 and 1679 are 45.7C, 117.4C and 139.6C.  The value of the channel 4 (the CJT) does not change and does not appear to be the source of the error.  This represents 3 errors in about 35 minutes, the other 2100 reads appear to be accurate.  In this run, all the bad values are somewhat bunched together.  Most of my readings show the problems scattered and occurring once every 10 minutes.  

I disconnected the power cable to my PC and let it run in battery mode.  I also added a large capacitor across the Vdd/GND terminals in order to smooth out any ripples.  The Vref looked to be a good 2.5V.  

I do not understand why I am getting these errors.  If I cant get it to work with the demo board, I will have problems when I try to layout a board.

Any help would be greatly appreciated.XLSX

Top Replies