IR tire temps vs. temperature at the core

[Nihal]

I was wondering how well realtime IR temp sensing on tires works. I've always thought that the temperature of the tire at the carcass/ inside the rubber. How big would temperature gradiant be from the outside of the tire to the carcass? Is the gradient constant slope? I'm interested in running Ir temp sensors for tire and camber testing etc. How well does the core of a tire retain the heat from a run?

[terra_dactile]

nihal,
i dont think its a question if it will work or not but how do you plan on buying these sensors, i looked for a while and good sensors cost 600 $ and more and you would probably need three per wheel to be able to see the change between in , mid and outside of tire.

if you find cheap sensor keep us notified

Jude Berthault
ETS Formula SAE
Steering System Leader
Brake System R&D

[B Hise]

It would be much "cooler" to have a high res IR camera. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Bryan

[B Lewis @ PE Engine Management]

Try looking at IR thermocouples like the ones from Omega http://www.omega.com/Temperature/pdf/OS36SM.pdf. They are reasonably inexpensive (<$100). Cheaper than that, try hacking a IR temp gage made for cooking (<$30)

[95M3Racer]

Recent RCE has a nice article on the IR hi res camera's, amazing stuff. On the fly IR tire temp sensors will be ok for comparison while testing, but if you want data from the time immediately after testing has been performed, core tire temp is still what you want to see what the tire temps actually are. But nothing so far can beat these HI res IR cameras...they can really show some amazing stuff.

[Nihal]

I was thinking we would only need sensors with about a 1-2 degree accuracy. I think it would really show what kind of temperatures the tires are seeing mid corner and roughly how the tire is rolling over/deforming. IR cameras would porbably be very expensive (and delicate). Didn't UWA do something like this last year? I'm sure other teams have tried it. How did it turn out?

[Kevin Hayward]

Nihal,

UWA have not used IR Cameras (well at least as of 2005 US Comp). I'd be really interested to find out more about them.

If you can find the money to get IR sensors for mounting to the car then buy them. They are invaluable. There are some cases where the surface temps do not give a good correlation to the carcass temp, especially when the wheel is sliding. However even in that case the information is valuable in driver training and diagnostics. (i.e. wheels locking up under braking shows extremely well)

To get a good idea of the correlation when the car comes screaming in after a run quickly get a carcass temp and compare to the data traces.

I can almost guarantee that if you buy the sensors you will not be dissappointed. At a worst case get 3 sensors and switch them around the car. You might even be able to get away with 2 if you run middle and inside but definitely not as good as running 3 per wheel.

UWA had 6 sensors and they would get switched around. However seeing as though the car is basically symmetrical we tended to do most of our tests with one set of 3 on a front wheel and the other set of 3 on a rear wheel. Most of the time they were on the same side.

Hope that helps

Kev

OptimumG

[Chris Boyden]

How about a wireless pressure and temperature sensor mounted on the inside of each tire?
It should be fairly easy to find a cheap and well calibrated temperature sensor on a chip.
Monitoring pressure real time would be a plus as well. Many car manufacturers today are working on ways to integrate pressure sensors into the wheel/tire assemblies to tell the driver a tire is getting low....
3 chip based temperature sensors bonded with rubber cement would be a lot cheaper than IR sensors. You'd be measuring the backside of tire and not the contact surface.

TPM scheme:
http://ww1.microchip.com/downloads/en/AppNotes/00238b.pdf

Debugging such a scheme would be some work...but interesting.

An NTC termistor could be used in order to handle the high temperatures.

[apruz]

There are many different factors regarding how the tire temperature is measured and how accurate it can be. I am replying to this thread because I am the sales manager for a Racing Company based near Detroit, MI. We happen to offer special deals for formula teams. We are talking on the range of $100 per sensor, sometimes even less. Anyone with questions should contact me directly. Thanks.

[Ekasey]

We make a nice IR tire temp sensor, terminated with an autosport the FSAE price is 420.75

Still not cheap but we may be able to bring the cost down more by replacing the autosport with a DTM or some other cost effective connector.

if anyone is interested please e-mail me
casey.heerman @ efitechnology.com


Originally posted by terra_dactile:
nihal,
i dont think its a question if it will work or not but how do you plan on buying these sensors, i looked for a while and good sensors cost 600 $ and more and you would probably need three per wheel to be able to see the change between in , mid and outside of tire.

if you find cheap sensor keep us notified

Jude Berthault
ETS Formula SAE
Steering System Leader
Brake System R&D

[Poe]

digging up a 2 year old thread for a sales pitch....followed by another in just minutes! http://fsae.com/groupee_common/emoticons/icon_rolleyes.gif

[Neil_Roberts]

Here is an example of the things you can learn from IR tire temp sensors:

http://thinkfastengineering.com/2012/08/solo-heat/

The surface is where the grip happens, so that's what I'm most interested in. Core temps are of primary interest to the tire manufacturers because they need to monitor the margins to blistering and carcass failure.

[Jersey Tom]

Originally posted by Neil_Roberts:
The surface is where the grip happens

Can't say I agree with that.

[rrobb]

Can you elaborate Tom, or is it top secret?

[Dash]

rrobb,

I'm not a tire expert, but it depends on the makeup of the whole tire.

Basically to get "grip", you have to stretch the tire in some direction. You then get a reaction force that is very, very, basically summarized into F = K * x where K is the spring force and x is the displacement and F is the force or "grip" the tire can put out.

Rubber can be referred to as black magic, because it behaves in a very unlinear pattern when stress is applied ( stress - strain curve is not straight ). Adding in temperature variations make this unlinear curve change shapes. Hotter generally means softer, and colder generally means harder. Really cold tires can go through a glass transition and become a brittle plastic. Really hot tires can become goop. There are then chords and stuff running inside this rubber to give it extra strength and shape. This whole big mess of stuff working together is whats going to give you grip.


Hope that shed a little light and didn't confuse you to hell and back.

[rrobb]

Thank you for your response.

Given the nature of my question, it was right for you to assume that I was looking for some tire basics.

I was looking for more of an understanding as to how temprature effected components other than the compound.

My assumption has always been that the belts and such are realtively immune to temp, while the compound is highly sensitive to it.

Given his background, and the nature of his response in this thread, I was thinking that Tom might be alluding to something along those lines that might be intersting.

[Dash]

I figured you meant you wanted a good explanation, but figured I could take a stab if you just wanted a little bit of info.

I'd like a good explanation myself. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Jersey Tom]

Without getting much into it, as Dash mentioned and is available in the public domain (see the work of Persson)... in-plane (lateral, longitudinal) tire forces are not generated by a Coulomb friction model as we are taught with a block on an inclined plane in physics class, where the surface is really the only thing that matters.

As an aside, with whether or not the belts and things are temperature sensitive... everything in a tire is surrounded by "compound." There's tread rubber, there's rubber around the belts, the plies, the bead, you name it. The whole thing is a composite structure of a fabric or steel reinforcement within a rubber matrix.

I'll leave it to you all to consider how that reinforcement might be loaded, to what extent the surrounding matrix's mechanical properties are temperature sensitive, and what effect that would have on things.

But don't spend too much time on it, probably more important things to focus on. Just some academic food for thought when it's 2am and you're sick of welding terribly fit frame joints, take a break, have a beer, and contemplate the universe.

[Moop]

I would think that the temperature of the rubber on the surface would influence how much grip can be generated, as all the force has to pass from the road to the tire and hence through that rubber on the surface. In fact, the TaMe tire model(which has achieved a level of correlation that I think is pretty impressive) uses a surface friction coefficient dependent locally on contact pressure and temperature to calculate the lateral and longitudinal forces.

I would also think that as the core rubber heats up, it'll become more soft and compliant, so you'll have to stretch the tire more(or put it at a higher slip angle) in order to generate the same amount of grip(assuming that the tire surface temperature stays the same through the whole ordeal).

Am I missing something here?

On a side note, has anybody had any luck chasing down the paper by Fevrier and Le Maitre "Tyre Temperature Modelling: Application to race tyres"? Supposedly, it has more details on the model. I spent a good while trying to find it purely for interests sake. I even got a guy from the school's library to help me look for it, and even he couldn't get a hold of it.

[Jersey Tom]

My understanding is Tametire considers the bulk temperature of the entire tread - which should be a good clue. Been a while since I've gone through any documentation on it, so I could be mistaken.

In any event the bit about all forces having to be transmitted through the surface of the tire, to me comes across as still assuming that the tire/road interface is 2 dimensional - it isn't quite. It does have depth, and tire forces are not just a function of traditional friction. As I said earlier I think Persson's work goes over this well, as does Pacejka's (different aspects of it). Both readily available in the public domain.

Beyond the more academic side of things there are practical challenges to IR measurement, including varying emissivity of the tread surface.

Without a doubt there's plenty of hand waving and mixed beliefs when it comes to this topic. Some of it is arguable one way or another based on individual's own anecdotal evidence... some is flat out myth. Suffice to say that in my opinion IR / surface temperature measurement is far from the holy grail or "the answer." May work acceptably well for some people's requirements, and completely not for others. Like anything in engineering I'd say you have to know the strengths and limits of any data or process, or you can get yourself in trouble. Almost more important to ask yourself, "What CAN'T this tell me?" and not just "What CAN this tell me?"

Really not much black magic to this stuff, just an issue of (a) the public domain information available [of which there's a lot] isn't obvious to find unless you know what to look for, (b) concrete data is hard to come by because of IP and competitive advantage, which leads to (c) massive amounts of conjecture and speculation which once repeated enough times is assumed to be fact.

My 2 cents on it anyway. Individual opinions and experiences may vary!

[Moop]

Tametire uses a 1D diffusion equation through the tread thickness, but apparently also somehow has the temperature used to look up the local friction coefficient changing through the length of the contact patch. The paper I mentioned supposedly has more details, hence my trying to track it down. Danny Nowlan references it in an '09(I think it was at least) issue of racecar engineering, so it's obviously available somewhere, I just haven't been able to find it.

I think I found what you're referring to by Persson, he's got a good number of papers on the subject. I most certainly don't need more to read at this point, haha, but I'll try to give it some priority in my gigantic list of things to read.

What work are you referring to by Pacejka? My lit search didn't pull up anything by Pacejka relating to temperature. Unless you're referring to SWIFT?