"K" Style Thermocouple

[steelandcanvas]

My Hewitt EGT gauge with the typical "K" style thermocouple, quit giving me readings last week. But before I tear into this to try and figure out what went wrong, I thought I would ask you guys about the easiest way to troubleshoot the problem. I have physically traced the wire from the probe through the grommet in the firewall, to the gauge and found no damage. I am thinking of disconnecting the thermocouple wire from the gauge, connecting my DVM on the Millivolt scale, starting the engine and see if I get a voltage and it increases. If the voltage is there and increases, the gauge is bad...if there is no voltage present, the thermocouple is bad, correct?

 

[Scar59]

That's basic trouble shooting, hopefully your DVM is sensitive enough. Most likely the probe.

 

[MtnSnow]

I have yet to have a EGT Gauge fail but I have had multiple probes fail over the years.

 

[steelandcanvas]

Hewitt has a lifetime warranty on their probes, so I may be in good shape.

 

[The HUlk]

Thermocouples are a resistance type circuit. To test them you would disconnect both leads and measure resistance(OHMs) across the leads. If there is only one wire then measure to ground(outside of thermocouple body). The resistance should change in conjunction with a change in temperature.

 

[Welder1]

I would call Hewitt and ask them best way to troubleshoot the gage.

 

[tim292stro]

Thermocouples are a resistance type circuit. To test them you would disconnect both leads and measure resistance(OHMs) across the leads. If there is only one wire then measure to ground(outside of thermocouple body). The resistance should change in conjunction with a change in temperature.

No. Thermocuples are not resistance type, they are voltage GENERATORS.

The "type" of thermocouple is the combination of the two types of metal each leg of the thermocouple is made of when these two dissimilar metals are welded together (usually with a carbon rod), they create a thermocouple joint. This joint will create a voltage in relation to the temperature the joint is exposed to, which can be measured at the other end of the wire.

Since each joint of a dissimilar metal creates a new temerature measurement point, all of the extension wire and the socket where the thermocouple plugs ito the gauge or Analog to Digital Converter (ADC) must be made out of the same "type" material as the sense junction. for those who are familiar with electronics, you will already know that a circuit must be a complete loop, so there will always be at least two places where each type of dissimilar metal will have a junction - the sense junction and the measurement juntion, what is called the "cold junction". the tempeature of the cold junction is often easier and safer to measure with other means, so it is measured externally to compensate for the voltage generated at that junction (leaving only the measurement junction voltage, and thus the tempsrature at that junction).

Basic check you can do for your thermocouple is continuity. Check that there is continuity between both sensor terminals, and no continuity between either terminal and the engine. If the sensor circuit is open your thermocouple is bad.

If you want to rule out a gauge, k-type thermocouples are very comon, just pick one up and wire it in. Analog thermocouple gauges are self powered by the thermocouple voltage, digital (ADC type) are powered by the vehicle ignition circuit.

 

[The HUlk]

No. Thermocuples are not resistance type, they are voltage GENERATORS.

I learned something today, never knew about the voltage potential phenomina. I have aways used an ohm meter to check for continuity and assumed it was a resistance change in the metal causing a change in the temperature reading rather than a voltage change causing an apparent resistance change by way of ohms law.

 

[tim292stro]

That's a different type of sensor, where the temperature coefficient is allowed to affect the resistance (either higher resistance with higher temerature, called positive temperature coefficient [PTC] - or lower resistance with higher temperature, called negative coefficient [NTC]). There are also temperature sensors based on transistors that will allow passage of more current when they get warmer, these are comon in integrated circuits, where adding one "temperature sensing juntion" to a multi-billion transistor part doesn't cost anything.

Just as you can generate a tiny low current signal to drive a gauge, the HDT shelter heaters use a thermopile to charge a battery from the heat of the burner - it's enough power to run the fan, fuel pump, and control box after charging the battery.

You can build these yourself or buy things powered by this effect such as stove top heat-powered fans, or USB charging backpacker stoves.

 

[74M35A2]

I work with K type TC's 8 hours a day. Here is your chart. Read the very short script at the top otherwise the chart won't make sense. Then, scroll down to "K" type. With this, you will be able to tell not only which is working and not, but also their accuracy. You should have 2 wires coming from the thermocouple. If you are using a DVOM, it really doesn't matter which is which on connectivity, as it will just read a negative value. it does matter which is which when connecting to the gauge display. When doing this, don't have your skin pressing against each DVOM probe, as that may allow a very small amount of current to bleed off and cause inaccuracy. So, at the same time, if you feed this amount of mV into your gauge, you can see the value your gauge should display. You really can't connect traditional wire to extend TC wire. It will affect the accuracy. TC's, a thicker one may have a longer life, but will be slower to react to temperature change due to its thermal mass, and vice-versa on a thinner probe. Just FYI. http://www.ni.com/white-paper/4231/en/#toc2On