diy infrared brake temp sensor

[ihasnostang]

I'm looking at adding a IR brake temp sensor which will be datalogged with an Arduino during track use and overlaid onto track footage. Here is the 800C sensor , which is 0-5V. i have been exploring places to mount it that put it within 50mm of rotor. I printed out the sensor and the sight cone that is 20mm diameter at 50mm away. Currently the easiest place seems to be mounting to the dust shields with a Z bracket as shown. There should be enough room in the cavity area of the dust shield to put hardware. I modeled my bracket to 2mm thickness, and the closest sizes on sendcutsend are 1.9mm, 2.6, 3, 3.6.

Looking for feedback as i am a bit concerned about vibration/ resonant frequencies. I recently installed the ducted belly pan and have the vorsclag deflections on the way as i have a track day at road america coming up.

Sponsored

 

[wingnutt]

good on ya… but curious, what is the issue for the data you are attempting to collect? Or is it just for GP? (General Purposes for the civilians in the crowd)

 

[ihasnostang]

just for the heck of it, seems like something not many amateurs do. and just to clarify, i would only be doing a sensor on one of the front wheels.

 

[HeelToeHero]

Why not mount a bracket under the caliper bolt. That looks like a perfect spot.
That dust shield will be too flexible.

 

[Optimum Performance]

You would be better off with a flir camera. I have lots of experience with infrared sensing and it's extremely easy to get bad data. Old school temperature crayons are far more accurate compared to a sensor. Flir cameras constantly recalibrate and you can film in real time. This is why pro race teams still rely on temperature paint. Even in a lab or factory race team setting with very expensive sensors there is still error. They use that data corrected by lots of brake dyno time in a lab environment. The sensors are using embedded in the pad to collect more consistent data.

It will be a great experiment and look forward to your results. I would strongly suggest over-laying the results with data from actual in contact temperatures because those sensors are heavily influenced by radiant heat so the data is dirty.

 

[tosha]

You would be better off with a flir camera. I have lots of experience with infrared sensing and it's extremely easy to get bad data. Old school temperature crayons are far more accurate compared to a sensor. Flir cameras constantly recalibrate and you can film in real time. This is why pro race teams still rely on temperature paint. Even in a lab or factory race team setting with very expensive sensors there is still error. They use that data corrected by lots of brake dyno time in a lab environment. The sensors are using embedded in the pad to collect more consistent data.

It will be a great experiment and look forward to your results. I would strongly suggest over-laying the results with data from actual in contact temperatures because those sensors are heavily influenced by radiant heat so the data is dirty.

This, and even Flir can be very finicky - try to accurately measure a temperature of glass or mirror with it.... or any reflective surface, which rotor surface is.

I have not tried any recent Flir cameras, but the one I tried was a $10k model and it still required post processing in the software and an old scool taped thermocouple was a very useful reference point.

 

[Optimum Performance]

This, and even Flir can be very finicky - try to accurately measure a temperature of glass or mirror with it.... or any reflective surface, which rotor surface is.

I have not tried any recent Flir cameras, but the one I tried was a $10k model and it still required post processing in the software and an old scool taped thermocouple was a very useful reference point.

The software is pretty good though. I used to make presentations with both actual images and temperature data of systems and it gave really good visual data to communicate points. After struggling with compiling data from non contact flow meters FLir was a life changer lol.

 

[tosha]

After struggling with compiling data from non contact flow meters FLir was a life changer lol

Oh absolutely, that IR/visual picture overlay and ability to tune settings afterwards was a complete new level. Solar farms acceptance testing, was a great learning experience, so much we could do with these cameras.

 

[ihasnostang]

Incase anyone didnt look at the datasheet, is it possible that this sensor is tailored for a narrower use case than say a general use household IR gun? maybe its a crapshoot

99% emissivity
8-14 µm wavelength.

https://www.flukeprocessinstruments...-center/infrared-technology/emissivity-metals


izze racing
"An emissivity of 0.55 and 0.85 is a good starting point for cast iron / steel and carbon rotors,
respectively.
o The exact emissivity of cast iron rotors is not constant and depends on many factors,
such as: rotor temperature, oxide layer growth, surface roughness/grooves, pad
material, arrangement of holes/slots, and rotational speed. Generally, the emissivity
will increase with temperature; accordingly, an emissivity of 0.50 to 0.60 is a
recommended starting point for rotor temperatures greater than 400˚C. It is the user’s
responsibility to calibrate the sensor if utmost temperature accuracy is important."

 

[ihasnostang]

Why not mount a bracket under the caliper bolt. That looks like a perfect spot.
That dust shield will be too flexible.

i took a look at your suggestion, it seems like it would of been a good place but it looks like that bolt is exactly the length of the caliper boss (or from what i could tell by the reflection off the rotor as i did this with the car on the ground). is 2 or 3 mm less of thread engagement really going to make a big deal?

all; i asked about the sensor accuracy out of the box on a iron rotor, they said " the accuracy would be read as 3% of Full Scale. "

 

[HeelToeHero]

i took a look at your suggestion, it seems like it would of been a good place but it looks like that bolt is exactly the length of the caliper boss (or from what i could tell by the reflection off the rotor as i did this with the car on the ground). is 2 or 3 mm less of thread engagement really going to make a big deal?

all; i asked about the sensor accuracy out of the box on a iron rotor, they said " the accuracy would be read as 3% of Full Scale. "

You can source a longer bolt. Just make sure it is the same design and material grade. I wouldn't reduce thread engagement, especially with the aluminum calipers. You could do a thread shear calculation to be sure but there doesn't appear to be an abundance of engagement to begin with.

 

[paulm1]

sometimes simple is better. https://kennybrown.com/collections/...s-technologies-gb7000-brake-temperature-paint

 

[Optimum Performance]

sometimes simple is better. https://kennybrown.com/collections/...s-technologies-gb7000-brake-temperature-paint

We agree.

https://www.opmustang.com/store/p42...ake_Rotor_Temperature_Indicating_Paint.html#/

 

[Skye]

sometimes simple is better

Your idea reminded me of F1 cars and the use of temperature-sensitive stickers around the car, to be checked during test season, pit stops, etc. There is at least one like product, for brake calipers:

https://www.summitracing.com/parts/wil-400-15671

Could be useful during reviews while stopped.

 

[Optimum Performance]

Your idea reminded me of F1 cars and the use of temperature-sensitive stickers around the car, to be checked during test season, pit stops, etc. There is at least one like product, for brake calipers:

https://www.summitracing.com/parts/wil-400-15671

Could be useful during reviews while stopped.

We also agree

https://www.opmustang.com/store/p114/Genesis_Caliper_Temperature_Label_250F_-_554F.html#/

https://www.opmustang.com/store/p232/OMEGA_Temperature_Label_170F_-_240F.html#/

https://www.opmustang.com/store/p115/OMEGA_Temperature_Label_250F_-_320F.html#/

Sponsored