For the Poll: Please select ALL options you would be interested. It's obvious anyone would want pay be less for the same item. So if you might be interested at $300 but would prefer to pay $200 please select both options.


Last year I built an IR Temp sensor array data logger for my old FSAE team. I am currently refining it some and was curious if there was any interest in the community to make a few production versions.


Specs:
-Arduino Powered
-4qty of 3 sensor arrays in an enclosure (12 sensors)
or 4qty of one sensor enclosures (4 sensors)
-sample rate of 2Hz (should be more than enough for dynamic tire temp readings)
-SD Card slot for logging to .csv file
-LCD display for live temperature views
-plug and play wiring


Notes:
-Enclosure has mounting holes for universal application but you will need to make you own brackets
-due to the nature of the sensors requiring a line of sight to tire, the set up will not be water proof
-I can make enclosure variations based on tire width for optimal sensor positioning ie: 205, 225, 275
-sensor can also be used to monitor brake rotor temps or any other surface temperatures


Potential analysis that the data can be used for:
-determination of optimal tire pressure based on even temp gradient
-effects of camber changes on inside to outside temp gradient
-effects of toe on rate of temp increase
-monitor heat cycles of tires
-study affects on tire temp of different tire compounds and tire sizes
-Validation of tire modeling software
-Advanced analysis can be done when combined with other sensor such as steering angle inputs, slip angle sensors, etc.
-Here is a neat analysis and write up: http://thinkfastengineering.com/2012/08/solo-heat/

Pricing Estimate: Unfortunately, these sensors are not cheap. Their price also fluctuates greatly based on bulk ordering (hence the wide range of projected cost). I cannot determine an exact cost yet but based on expected materials cost

$180-$300 for 12 sensor kits
$100-$200 for 4 sensor kit.

The only similar product I have seen is this: http://www.opti-grip.com/optigrip/Opti-Grip.html
-$169 for 4 sensors without logging
-$449 for 12 sensors without logging (requires a data logger with 12 free analog channels. Analog resolution < digital)


Please provide any feedback for additional features that you would like to see in a product such as this if you are interested.


Mods: If I am not allows to post something like this please remove this post. I am not selling any products, just trying to survey the market to see if there is interest. If there is I would happily become a official vendor here. Thanks!

Very cool idea, one way to justify the costs of an IR setup would be that it could effectively replace the purchase of a probe with the added benefit of being able to be datalogged/viewed on-track. The 4-sensor kit would be radical for things other than tire temps, such as brake rotor temp... I voted for the 4-sensor version between $100-$200, and the 12-sensor version at $200; don't know if I'd spring for it if closer to $300.

One question - since you mention they can't be sealed enough to be waterproof - how susceptible are these to damage from dirt/debris/fluids?

edit: another question - what is the +/- accuracy and is there a max temperature these will read?

-Ryan

I read your tech article on brake cooling and instantly thought this would be perfect for you. Sending you a PM.

One thing that needs to be taken into account is the emissivity of aluminum is low which will throw off the IR sensor (they are calibrated for 1.0 emissivity, which is the coefficient of a pure black body). I have gathered brake rotor temp data on our FSAE car successfully and it seamed reasonable, but a study of it's accuracy on an aluminum body needs to be done to take into account the emissivity difference.

this is taken from the data sheet: (380C = 716F)
As far as debris go we ran it a few FSAE test sessions without any issues. I am looking into clear protective covers as well. If I were to sell this, they will be thoroughly track tested before release to make sure it is not an issue.

Hmm, would it be possible to envision spitting out the data to a CAN bus in some way to make it work with MS/TunerStudio or other CAN able logging system?

I had a hard enough time when I was using MS+VVTuner that there were two separate logging systems (and that's only in a Dyno setting), so I'd assume that you would benefit the same way in the pits by only needed to look at one log file and have all your data there.

But I have no clue if it's easy to just cram more parameters into a CAN bus (is just unique names enough)?

This Device->CAN->MS->BT->Shadow Dash/Logger ?

Just a quick search hit
Hack your vehicle CAN-BUS with Arduino and Seeed CAN-BUS Shield

If you can make it output 0-5V per sensor, I think most people would go for it. The MS3 has enough logging capacity for this.

How close to the tire do they have to be?

READING THE DATA SHEET: Spot diameter plot only goes up to 30 (mm?) So like 1.25"?

One picture from a discussion of the Opti-Grip
https://www.miataturbo.net/attachmen...ine=1413461135
It also shows what's needed for the front tires (they need to follow the wheel).
Add some metal on top of them and the sensors are shielded from (most) harm. Additional shields in front and below and they will live longer in the gravel as well.

Hmm, I can't find the price (if you have to ask...), but these look real nice.
Products » Sensors » Temperature sensors » Tyre temperature - Racing - Race Technology Ltd - Automotive Technical Excellence
One CAN interface per sensor strip make it quite general.
http://www.race-technology.com/wiki/...yreTemperature

Just for inspiration of course.

Here is the first prototype from last year's car. The enclosure will be similar to the flat section parallel to where the tire is, but a little more compact in both width and height. The brackets were all 3-d printed and attached via the caliper mounting bolts. You can also spot our rotor IR temp mounting location in this pic. For a 3 sensor array, the front will require an upright mounted bracket to follow the steering angle changes. The rear can use a chassis mounted setup if total suspension travel is low and care is taken into the mounting location. For a 1 sensor front setup up, a chassis mounted bracket can be used if centered properly.


https://www.miataturbo.net/attachmen...ine=1411103077



Yup, if this thing takes off a CANbus module can be added to output to a MS or any other datalogger. I've had some CAN bus experience and am pretty confident I can implement it.

Reverant, I think this would be a good idea for the 4 sensor option but most people will not have 12 free 0-5V analog inputs available unless they have expanders. If I use the arduino mega I should have enough analog outs to convert the signal for use of an external data logger. I'll look into developing this. Thanks for the feedback.


it depends on the exact sensor selected since they have different field of views. but about 2 in from tire is a good estimation.

Thanks for the link. "If you have to ask, you can' afford it." lol

Here is a quick little demo video of the system in action.

Unfortunately, due to the amount of labor hours required and the lack of a water proof enclosure for the sensors, I will not be making production versions at this time. I will be doing more testing in the mean time and we will see how things go.

Very cool use of remote sensing. I think that the data would be great to have, but are surface temps a good representation for what is going on with the tire as a whole?

There is some debate on this. For the traditional method of taking tire temp data after a session, most believe using a IR non contact measurement is useless since the surface temp has cooled significantly from the cool down lap and time between stopping the car and the measurement. Therefore, a temp probe is better since the temp slightly under the surface will not have cooled off as much.

Taking live data on the track, IR non contact sensors is the only way (either an array of sensor like this setup or thermal imaging). You will be surprised at how much heat gets added into the outboard tires in a turn and how quickly it dissipates out on a straight. These types of dynamic changes can be shown and studied with this set up.

there is little debate about live tire temp data. What people question is an IR temp sensor after you pull back into pits (at which point using a probe properly is the right way). Live data is great and lots of pro touring cars use setups like these.

Bring something like this to the miata world is really nice, but it can be difficult to take full advantage of this type of data. Cool stuff though!

This is a great idea. Props given. IMO water-shielding doesn't seem like it should be a paramount concern as a DIY project, though I do understand the need to get wet-track data too. I hope you're able to push forward and make this happen despite the setbacks.

It would most benefit competitive race cars, and hopefully trickle down to HPDE cars via 949, TSE, et al experience on wheel rates vs. tire selection vs. tire pressure vs. alignment.

Personally, since my car is just an HPDE car, I'd be interested in a more ballpark setup to get tire pressures and camber about right, not squeeze out the last tenth of a second. As such, a $100 4-sensor setup sounds good to me.

Keeping the whole setup mobile is a big plus. Going from tire temps to under-hood heat-shielding reality check to brake temps would be very handy.

Having looked at those sensors for this purpose in the past, I'm guessing you're using the PWM interface(since you can't stick 12 on the same i2c bus)? What kind of speed are you getting with them?

Thanks! :)

The water shielding will also help ease my mind of rocks getting flung at the sensors. Maybe after extended testing with open sensors without any issues I will look past this concern.

I am looking to use it for simple camber and tire pressure settings at first too. Its also would be a useful tool to track max temp of tires in a session and heat cycles on the tires.

My old FSAE team is planning on doing more advanced vehicle dynamics analysis by syncing the data with a multitude of other sensor on their DAQ so it would be interesting to see what they come up with.

The biggest set back for a business case for these at the moment is the 3D printing aspect. My personal 3D printer is a reprap and is kind of high maintenance. If I had the stratassys printer I use at work this would be less of an issues because I can load up 10 or so enclosures in a single print over night and not have to worry about any issues. But that's a $30k printer that uses expensive material :( Also, there is a bit of soldering, and programming involved to build the kit and to make it worth my time kit price would be out of the budget of most.


All the sensors are on the I2C bus. You need to program a new unique address in place of the factory set one for each sensor to use multiple on the same bus. I think you can have 127 unique address for 127 total sensors all on the same bus. For 14 sensor I can read at over 10 Hz if I am just reading data and logging to SD. Once I add the LCD functionality and analog output it drops loop time to about 3-4 Hz. I currently have it set to 2 Hz which should be sufficient for most.

Nicely done! I know there are also challenges with i2c over "longer" distances and was one reason I was shying away from using that protocol.

IRN-RC - Motorsport IR tyre sensor, IR tire temperature, NASCAR sensor, multi-channel infrared - Texense sensors - By Texys

&

IRN8 C - Tyre temperature sensor, tire temperature sensor, multi channel, CAN, Motorsport, - Texense sensors - By Texys

A CAN transceiver is a basic requirement for a product like this.

Racepak is on the cheap end of motor sports sensors and their single IR sensor retails for $350. So for 12 thats $4200 + $1500 for the Data logger. The sensors you linked are from a higher end company and are probably only sold to top tier professional racing teams for $6-$10k. My goal is to develop a 12 sensor kit that the average Joe track day enthusiast can use for a budget of $300. This is easier said than done.

I have played with CAN communication and have working code to transmit all sensor values to be logged by another 3rd party data logger. The problem is, most of the big companies won't support a project like this. I've talked to Racepak and their VNET (CAN) network is proprietary so they can sell you their $350 a piece sensors. Aim won't support both and ECU via CAN and another device.

I can probably get code to log to the Megasquirt via CAN and might attempt this at some point. I have the frame ID, byte order, scaling, and offsets for the CAN messages all documented if anyone with Megasquirt Coding expertise want's to give it a shot.

Have you looked at the Race Technology DL1? I know they have a CAN interface, but it may have the same problem as the aim stuff. Race Technology is much better about allowing you to setup universal sensors instead of requiring proprietary Race Technology branded parts. I loved my DL1.

Also, thanks for posting a part number on those sensors. My boss has been encouraging us to start our own non-work-related arduino project with an offer to pay for sensors and parts, so I may try to play with this stuff if I can find time.

Thanks. I'll look into them. Almost every data logger has CAN support. It's just a matter if they are open source/ or advanced enough to let you configure your own CAN framce. Which most aren't and the average user has not need for this feature.

No problem on the sensors. I have quite a few brand new ones and will beat whatever price you find for them online if you need some. Let me know if you need any help with the coding too. There's a couple straight forward example code for the sensor in the arduino community.