Home brew Exhaust gas temperature sensor

[elnav]

I have several exhaust gas pyrometers  in my junk box.  Trouble is none of them have a working sensor.  It seems most pyrometer sensors fail after a while.  I have the correct thermocouple wire  but  so far have not had any luck  fabricating a durable  sensor. Has anyone  done this and succeeded  in making one that stood up to more than a few hours of use?

[elnav]

Where  on an air cooled engine  is the best location to mount a temp sensor? 
If no sensor is available as an option what is the best way to fabricate  something home brew.
Most  water or oil temp sensors  look to be too big  to be suitable even if the temperature range is suitable. 

[Westcliffe01]

Have you looked at omega.com ?

The platinum based PT100 style probes are about $70. http://www.omega.com/ppt/pptsc.asp?ref=PR-13&ttID=PR-13&Nav= You can use an ohmmeter to read them because the resistance changes from 100 Ohm linearly as the temperature changes.   

Alternatively, Type K thermocouples are cheaper http://www.omega.com/ppt/pptsc.asp?ref=GJMQSS&ttID=GJMQSS&Nav=, but since they produce just a tiny voltage, you would need a meter designed to read them, which is an additional expense.  The good thing is that replacing the element that goes in the exhaust is relatively inexpensive and the fact is that they do wear out (but that takes many hours, not minutes).

[elnav]

Ooops! : :-[Guess I should explain in better detail. I am familiar with the technique  described by Jens and have used that  in the lab on previous occasions   And I have  the Omega catalogue. 
I also have the old SS thermo wells from failed  sensors. Apart from them being too large physically, it seems a shame to just throw out something that could still work.  What I have not yet figured out is how to repair the thermo couple  and bed the wires in the well so it is sealed in that white ceramic stuff you can see,

The  wires  used are actual thermocouple wire ( usually #18 ga  or maybe #20 )  but  welding the wires  leaves them with exposed ends  that could short out to the grounded shell of the well.   and the typical  sensors used on Catapillar or Cummings engines are  kinda big for the smaller engines I had in mind. 
Yes I know I could buy ready made  probes from Omega  but being  temporarily financially embarassed  I simply haven't the $$ to go buy something.    Hence my attempt to roll my own.

[elnav]

[/Ok, one last possibility .... Sodium silica .... I do not know if this would be effective as an insulator but it's cheap. The problem I have with potting things is that it really throws the response time of the thermocouple for a loop because of all that mass/quote]

REPLY
Seeing as how the original probes  have been potted  I don't think  thermal response time  is a major problem.  All the pyros I have seen tend to respond within a few seconds  to added loading as throttle is opened up. 
Probe failure tends to be vibration induced fatigue  where the wire breaks from flexing, If a coiled spring is used to support the wire the break is at end of spring if no spring then the break is right where the wires exit the thermowell casing. either way  when the meter is mounted  remote from engine  and probe is on engine  there is vibration.  I do have  barefoot thermal probes for my test meters but then there is  mass in the manifold  or the clip used to hold probe to exhaust manifold. 

[billswan]

Guys

Have always wondered if a thermocouple out of an older furnace might work? They can be bought new for less than 10 $ and are used to hold open older gas valves. Always thought about trying one, they glow red hot when they are in a pilot flame and give off about 10 milliampers give or take when red hot. Usually last for many years to!

Billswan

[elnav]

And they are all jacketed in copper. The copper would work harden  from the engine vibration in short order.   

[rcavictim]

Isn't a thermocouple two partners that share the infliction of thermopiles? 

That burning sensation's gotta hurt!   For the record, I have no experience in such matters which might best be filed under 'occupational hazards of voluntary lifestyle choices'.

And now back to to the actual topic.  Sorry I thought this was funny and an opportunity to whore my post count.

[billswan]

Thanks guys now I know 

Learn something every day.

Billswan

[elnav]

On the subject of thermo piles.  The question was asked on another forum.  Does a 'thermocouple'  in a furnace produce current or voltage.  I among many others  thought it produced voltage. The correct answer is more current than voltage.  It was pointed out the solenoid valve had few turns of large diameter wire not many turns of thin wire to create enough flux to open the valve.   I didn't quite believe it so I dismantled a gas valve to find out. sure enough  the coil had few turns of relatively thick wire.        

A couple of  years back I read  that one truck company was offering a electrical generator  of 1kW output  that consisted of a exhaust stack  section surrounded by "thermo couples" whic hI now realize should have read "thermopiles".   Hmmm?  Maybe a means to make a non rotating start battery charger if  a belt driven alternator for a slow diesel is not practical?

   

[XYZER]

Does a 'thermocouple'  in a furnace produce current or voltage.  I among many others  thought it produced voltage.    

I think it is called EMF.
http://en.wikipedia.org/wiki/Electro-motive_force
they use 2 disimular metals for wire.

[elnav]

As explained by the wikipedia file  a thermopile is more pf a Peltier- Seebeck  effect
EMF is also used to denote Electro Motive Force.
The  thermopile used in a furnace  probe or the truck exhaust pipe generator  is direct heat  to electricity conversion  without any movement of wires in a magnetic field like that used in an alternator. 
The thermopile is recovery of waste heat  whereas converting rotary motion into electrical energy is not.
In its native form  the Lister or Changfa  does not include any charging source for an electric start.  That was my point.  The exhaust heat first passes through the thermopile generator then to the heat exchanger to recover more waste heat.   

[rcavictim]

I suspect a Thermopile is merely a stack or 'battery' of series'd Thermocouples arranged to produce electric current with voltage as a DC generator intended to drive a DC load.  I saw a military unit on sale at a hamfest in the US that ran on a bottle of propane and made a half kW or so and produced nominal 12 VDC.  They are not particularly efficient.  They require one side of the disimilar junctions to be heated and the other side kept cool.  The unit I saw used cooling fins ala large finned heatsinks to do the cooling in air with convection currents or possibly blower fan.  I forget, that was many miles ago.

[rcavictim]

I suspect a Thermopile is merely a stack or 'battery' of series'd Thermocouples arranged to produce electric current with voltage as a DC generator intended to drive a DC load.

You would be wrong if you suspected that ..... chances are that the item you saw at the ham fest was based on a Peltier effect device.

Jens,

Really?  I am aware that Peltier devices work as DC generators when heated on one side and cooled on the other but I also know that any devices I've seen are manufactured using low temperature eutectic solders and they simply cannot handle much heat without simply melting and falling apart.  They certainly cannot handle a propane flame.  Do you know of Peltier cells that are made to handle higher absolute temperatures for power generating applications?  If they exist maybe they would be attractive for concentrated solar power generation instead of Sterling engine technology.  Normal Peltier cells are way cheaper per watt than present day solar photovoltaic, so even if the Peltier was way less efficient compared to the ~15% of photovoltaic, it might make economic sense.

Funny, I was looking at Peltier cells yesterday and thinking about this very thing!

[Lloyd]

Here's how to use a thermo-couple in chp.

Lloyd

http://www.cyclonepower.com/works.html