Guys,
I was at work talking with the head research engineer at our facility. We manufacture Thermistors, which are temperature sensitive resistors. Ok here is what transpired:
He had a couple of automotive parts on his desk that I was ripping on him about saying that he didn't even know how 1/2 of them work. Him and I are good friends, so we are always jerking each other around. Anyways I commented to him that we should find a way of manufacturing Wide Band O2 sensors for a cheap price. After explaining to him how they work and the benifit from having them to tune with and also the explination of how many individuals would buy them if they were a reasonable price ~150 bucks or so. He became very interested and started asking more questions. Well here is what we got to. I also mentioned that many of our components are temperature sensing equipment and that maybe a EGT (exhaust gas temp) sensor would be more up our alley. Well he about jumped out of his seat and began explaining that this coming week they are to build 10 prototypes of *very high* response, very accurate heat sensors (good from 25*c to 1000*c). They are experts at developing devices with very accurate results and very quick time response. Our cars probably have several in them already, ie, manifold temp sensor ect. He explained he would be very interested in building 1 more of these sensors for me to put in my exhaust and have the electronic techs build me a way of recording data just as our scanners do. I explained that we should get the sensor installed and get it logging, then take the car to a dyno with a WB o2 sensor and chart the exhaust gas temp along with the WB reading and try to relate the two back to Fuel ratios for a tuning device. I realize that guys already use EGT devices for that exact purpose, but perhaps as we spoke we could provide this tool at a very low cost. This would allow us all to have a very accurate way to tune our cars. Just as a WB o2 sensor would. Obviously there would be other issues that would need to be oversome such as placement in the exhaust and what not, but its still incouraging.
We are going to chat with my boss to get the final approval of the testing and then we will get the sensor installed and logging equip developed and then I get to take my car to a dyno several times to get some runs at the companies expense! Can't beat that! I will let you all know how things unfold.
Later.
Jesse

 

On the EGT side, you have to stress to him how tought of environment a car is, lots of vibration, and carbon generated. An they can be asked to run at 1900dF for prolonged periods of time. Also, there is a standard of about .125 for drilling a hole and using a hose clamp to keep on in. There is no such thing as too responsive.

What would be 10x as nice, IMNSHO, is a good MAT sensor. If you took and showed him a MAT, and couple get one that was a bolt in, and reacted faster that would be neat.

There are some problems with the way GM does thigns, and with a good MAT and code changes, you could really have some neat stuff.

 

Grumpy,
I believe the sensors they are developing are for Mack Trucks. They need the sensors for measuring exhaust temps after the Turbos. I can't get much more detailed than that, but they are suppose to be, from talking with him suppose to be the quickest responding sensors of their type. Mack has been there several times discussing the parameters the sensors would need to exibit and now they are ready to build some. I will get one of them. Therefore he is well aware of the enviroments they will be exposed to. He just became very interested in the performance market that might exist that they may have overlooked. Hey who knows, maybe I could become part of the marketing scene. They could help build and pay for my vette to be running in the (oh say 9s) and I could travel around to the circuits promoting the system. Sounds like a nice enjoyable job to me!
I know you are much more knowledgable than I am on these items, perhaps you could give me some direction on reading material for EGTs that exist now and any other comments you may have. I would also be willing to talk to them about letting you trial one as well. From reading the boards, you are a very knowledgable individual. This I would obviously have to clean with them, but I think it would be more than possible, since once they hit the market either for this application or mack, they could be reverse engineered. Although taking it apart would yeild very little information, since the material ingrediants that make up the thermistor are sintered together and materials change phases and chemically bond and also burn out. Therefore a person would be left with many pieces of the puzzle missing. Well I have to go now. Let me know your thoughts. I think it has great potential, but you would know better than I.
Thanks Grumpy.
-jesse

 

For the EGT sensors, I doubt you need something too exotic like Pt-Rh junctions. These should be available with SS or Inconel sheaths down to 1/16" easy, but 1/8" would be nice & durable (just a little slower). Maybe just a K-type couple would be OK, and they're cheap. They output mV, and there should be myriad devices for display/logging/amplification. I know the rice burner guys have all kinds of options like that.

Key for response time is being small. Just have to balance that with durability and manufacturing tolerances.

Now for the intake temperature thing, yeah that would be good too, especially since the gas temperature varies so widely and quickly depending on boost, MAP, etc. Expansion and pressurization can change gas temperature instantaneously, and a MAT sensor which weighs in at several ounces is not appropriate. Here, since the temps are pretty mild and the atmosphere is relatively forgiving, you should be able to use a RTD-type thermistor. These are a Pt wire which varies its resistivity as a function of temperature. A current is applied and voltage measured through a divider, and there you go. Voltage as function of temperature. Much like the MAF sensors. A BUNCH of equipment uses this principle for analysis of all kinds of things, particularly thermal conductivity of fluids / gases (TCD chromatography). I've used these for some REALLY sensitive measurements in some fluid dynamic work, and they are great. Even just used raw Pt wire as a COMBO heating element / thermosensor / surface catalyst for chemical reaction catalysis studies.

Some of these sensors are VERY small and give temperatures within 0.1F. Capture variations even smaller with great repeatability.

Check this link out:

http://www.omega.com/temperature/tsc.html

These guys, and others, make thermal sensors for the process industries, where environments are MUCH harsher than automotive, and an adaptation would be very cool. Maybe this helps some, I don't know.

I have had K-type couples fail at temps around 1000C (1800+F) even with Inconel sheaths, but that was in a condensing HCl environment after a few hundred hours, which isn't real easy on SS alloys (Chloride stress cracking), and the couples were only 1/16" diameter.

A rice-burner friend of mine installed an EGT into his exhaust manifold (kinda meaty) using a NPT/compression fitting, but for headers or something just a clamp-on 1/8" drilled through with some sealing dope would probably be fine.

Bruce, I bet with a couple Swagelok fittings you could get a bolt-in MAT that'd work like a dream. Maybe even with a little extension inside to be placed, say, at the intake port or behind the throttle body. Should be standard 3/8" NPT. Reduce to 1/8" tube, drill-through, and pass your probe through. What's the temperature-volt-resistance behavior of the MAT you have in mind? Should be able to match something up with a little Pt coil on the end of a stick which has more like a 1/10th-second response. Jesse, maybe you guys are already there.

For what it's worth,

-Craig