Smoothing Thermocouple output

[jucole]

I like the look of that thermo-cal unit although im always a sucker for new tools.

I took a nice picture of it this morning for you; I bought it on Ebay very cheap (~£15) 2 years ago when I started the controller.  The great thing is most dmm meters read k-type temperature, so checking it's ok is very easy.  The thermocouple shown in the pic is one for a kiln; which is a thicker than normal type with small protective ceramic tube coverings.

I considered using the ad595 but they are really expensive for my liking although they do provide a no-nonsense signal.

Yes, they are very expensive;  I bought a couple of them on Ebay from a guy who was selling some old ic socketed temperature boards, they were £5 a board, so was well worth it.

Damn if you dont do electronics or maths, what hope is there for the rest of us when you do!

no, sadly not a chance! i'm pretty sure I'm dyslexic.  When you guys read datasheets - i just look at the pictures ;-)

[Apothus]

Thanks for the pciture, i will keep my eyes out for one of them on Ebay!

Truth be told most of the time i try to short cut and just look at the pretty pictures then get something spectacularly wrong

[robrenz]

@jucole,  how does that thing compensate for the thermocouples that form between the binding post  material and the various thermocouple wire materials?

[SeanB]

Built in cold junction compensation. Nice unit, i had to use the good old fashioned analogue versions with spanwires and a thermometer built into the panel , though you could get by with ignoring it for EGT monitoring, as you could not read a 25C difference anyway at 1500C.

[jucole]

@jucole,  how does that thing compensate for the thermocouples that form between the binding post  material and the various thermocouple wire materials?

As SeanB mentions there is probably a nice temperature sensing device that measures the binding post temperature.  I'll have to crack it open tonight and have a look; There is an option to output a simulated temperature based on either the internal reference temp or an external temperature reference.  You see similar units fairly often on Ebay;  although these units are old - they still work well and at a great price!   

I'm at the leading edge! of trailing technology! ;-)

[SeanB]

The other method I had was to take the multimeter, set to resistance and screw the test leads to the terminals, and adjust to 16 ohms with the compensation pot ( assuming the indicator was working) so that it would read correctly. Then start the engine and see if it ran up to operating temp in the first minute. Worked because the indicators were calibrated at a certain current from the TC for full scale and the TC was designed to work into a certain resistance. Cold junction compensation was done by setting the pointer to a hair right of zero. Only time you would find the sparkies calling you to a wiring fault, they could never figure how it worked without power, so I often was there replacing 2 pins and 2 sockets in the shell, the electricians doing the rest, but I had to do the 2 TC wires and have it signed off.

More important was the big red light that said vacuum low, as this was generated from a pressure bleed off via a venturi. No suck - no combustion pressure and you were losing the big fan keeping you cool.