AD8495: thermocouple interference

Document created by analog-archivist Employee on Feb 23, 2016Last modified by analog-archivist Employee on Feb 23, 2016
Version 2Show Document
  • View in full screen mode

My data acquisition system has 16 K-thermocouple channels.  For each channel, I
intend to use separate AD8495.  The thermocouple tips will NOT be insulated and
can touch the metal body of the vehicle.  My question : If two or more
thermocouple tips touch each other (through vehicle body), will it cause
problems in readings? Do I need to go for isolation?

 

So I’ve discussed this with a couple other engineers now and understand better
the source of error with a bare wire TC in contact with a chassis, etc.  If the
chassis is floating (or has a floating voltage) and the TC tip is in contact
with it, it becomes a source of common-mode voltage carried by the TC.  So
simplistically, the voltage measured on the TC would be this common-mode
voltage + errors + the actual TC temperature measurement.  The errors are a
result of the common-mode voltage interacting with the 1MOhm resistor to ground
that we recommend, which produces a small current… This current through the
wire resistance of the TC generates a small error voltage.  This error is
dependent on the wire resistance and common-mode voltage.

With TC tips touching each other, or chassis at the same time, should not
affect the measurement…. This is because the TC’s temperature measurement is
relative to the temperature difference between the measurement and reference
junctions of that TC. 

So after having understood these effects better, TC tips touching is ok, but
the customer would need to evaluate how much error y they could have dependent
on the lengths and wire resistance of the TC leads, and any possible voltage on
the chassis.

If the customer would know that the chassis is always grounded, then they don’t
need the 1MOhm resistor shown in the attached drawing (Figure 1).

Outcomes