Pressure Sensitive Paint [PSP] - An amazing tool for aerodynamicists!

Рет қаралды 3,703

Күн бұрын

This is a compilation of my learning process as I used Prof. Keisuke Asai's Pressure Sensitive Paint [PSP] formulation to perform aerodynamic experiments in the Slanted Cylinder model at subsonic conditions. Paper coming up!
Check out the papers by Prof. Asai's group for details on the working principle, processing and the technique:
A review of pressure-sensitive paint for high-speed and unsteady aerodynamics (2008) - JW Gregory, K Asai, M Kameda - doi.org/10.1243/09544100JAERO243
Evaluation of the Characteristics and Coating Film Structure of Polymer/Ceramic Pressure-Sensitive Paint - Y Sugioka, K Arakida, M Kasai, T Nonomura, K Asai, Y Egami, K Nakakita - doi.org/10.3390/s18114041
This is a paper I wrote on the usage of this paint on one of our aerodynamic models:
journals.aps.org/prfluids/abs...
Link to NASA Ames PSP article:
www.nasa.gov/feature/pressure...
www.nasa.gov/feature/unsteady...
Recipe:
- 10mg of Pt(II) meso-Tetra(pentafluorophenyl)porphine - CAS no. 109781-47-7 - orders.frontiersci.com/Orders/...
- 100mg of Isobuthyl Metachrylate Polymer - www.tcichemicals.com/US/en/p/...
- 900mg of TiO2 nanopowder:
30nm - www.us-nano.com/inc/sdetail/604
15nm - www.us-nano.com/inc/sdetail/1...
- 10ml of Toluene - www.mcmaster.com/3190K653
Timestamps of the video:
0:00 Intro
0:39 Luminescence
2:10 Luminescence quenching with oxygen
2:40 Comments on my questions about the fluid variables being measured
4:01 Engineering a paint nanostructure
5:31 PSP Recipe
7:10 Paint application and mistakes to avoid
7:42 Correct application of the paint
8:04 More mistakes
8:20 Bad paint job vs good paint job
8:56 Details on the aerodynamic testing
10:27 Post-processing by simple gaussian filtering
11:04 Dynamic characterization with a shock tube
11:52 Dynamics extraction with POD
14:36 Thanks for watching/Outro

Пікірлер: 32

@funlearneat 2 жыл бұрын

You are the only person who has explained this topic. Thank you.

@delanehood4784 2 жыл бұрын

Outstanding video! As an Aerospace engineer undergrad, love to see your videos.

@coleburnette1597 2 жыл бұрын

Incredible video, thank you for posting!

@lilolail1582 Жыл бұрын

That’s really awesome video with great details Tons of thanks to you

@kadealey1565 3 жыл бұрын

Absolutely beautiful video. Thorough, yet concise.

@CristinaRodriguez-zn9ew 3 жыл бұрын

Once again, great explanation of a very novel technique and the additional processing with it. Your tips to avoid mistakes are extremely helpful. Thanks for that! Looking forward to your next videos on the topic!

@fzigunov 3 жыл бұрын

Thanks, Cristina! I'm glad you found it useful!!!

@SKKarthick12 7 ай бұрын

Hey nice video man! you always surprise me!

@fzigunov 7 ай бұрын

Thanks!! I saw your talk at APS on the lobed nozzle, very nice work!

@ZitaoJiang Жыл бұрын

Very informative video! I have a qeustion about the UV-LED: Is there any special requirement for the lighting system?

@fzigunov Жыл бұрын

Glad you found it useful! Usually you want a deep violet or UV LED exciting about 370-420nm. The LED needs to have a fairly high power to excite the paint enough so your camera can see the signal. It can also be a laser if you want. As an example, you can use the CBM-120 series from Luminus Devices, they have an assortment of UV and violet high power LEDs. Regarding whether it needs to be pulsed; not for this technique. There are some PSP variations such as the "lifetime" technique that use a pulsed light source, but for the one described in the video you can use a continuous light source. Also, remember to long-pass the camera view with a red filter to remove the blue reflections. Edmund Optics sells some of these filters (they call them "machine vision filters") with M52 threads for commercial camera lenses. One that we used for PSP is the EO stock number #54-764, works pretty well. Hope this helps!

@restmusic7417 2 жыл бұрын

A very interesting video. What software did you use to overlap the two images (I / I0) and how?

@fzigunov 2 жыл бұрын

Thanks! We use Matlab, but any data analysis package (like Python or others) could also do the job.

@fzigunov 3 жыл бұрын

[Update] Our collaborators (Prof. Nonomura and his group) now have their recipe published on Sensors and Actuators A! Please, if you use this recipe for your experiments, cite this paper: www.sciencedirect.com/science/article/pii/S0924424722007750?via%3Dihub Summary of the recipe in case you just want to make it: - 10mg of Pt(II) meso-Tetra(pentafluorophenyl)porphine - CAS 109781-47-7 - orders.frontiersci.com/Orders/WebPlugin/ProdDetail.aspx?cat=PtT975 - 100mg of Isobuthyl Metachrylate Polymer - CAS 9011-15-8 - www.tcichemicals.com/US/en/p/M0086 - 900mg of TiO2 nanopowder: 30nm - www.us-nano.com/inc/sdetail/604 15nm - www.us-nano.com/inc/sdetail/13037 - 10ml of Toluene - www.mcmaster.com/3190K653 Costs for one mix (~11ml, enough to cover 150x150mm area) as of 2021: PtTFPP: $285/250mg - *$11.40/mix* IBM: $17/25g - *$0.07/mix* 30nm TiO2 nanopowder: $45/25g - *$1.62/mix* Toluene: $14.23/qt - *$0.15/mix* Total cost of the mix: *$13.24*

@JW-kl2wd 3 жыл бұрын

Thanks for the amazing video, can you please make another video and cover the calibration of the paint?

@fzigunov 3 жыл бұрын

It's coming, but it'll take a while! We are currently upgrading the PSP calibration chamber in our lab, when that is done I'll make a video on it!!

@JW-kl2wd 3 жыл бұрын

@@fzigunov thanks for the reply, can you please recommend me some sources , that may helps me to implement the image rectification with MATLAB, because i have to analyse all the data and make like an evaluation software with MATLAB.

@fzigunov 3 жыл бұрын

@@JW-kl2wd For references, I recommend the papers I included in the description. Processing is as simple as a division and a polynomial fit in Matlab; so as long as you read the references you should be good!

@JW-kl2wd 3 жыл бұрын

@@fzigunov Thanks so much, i am gonna read it.

@jorgemunoz6973 2 жыл бұрын

Dear Fernando, thanks a lot for such interesting video, giving so much info and ideas to be done. Just a question, the most challenging is the camera set-up. Is it required any special quality of this high-speed camera?

@fzigunov 2 жыл бұрын

Hi, Jorge, thanks for watching! I didn't discuss the camera setup as it's mostly "point and shoot", but I know there's a lot of details to it that require experience and knowledge such as illumination, depth of field, aperture, color filtering, etc. Those are typically the same type of photography skills learned for other kinds of scientific photography, but here the goal is to maximize light captured while still capturing the model. In order to get good results, you typically need a camera with good dynamic range (i.e. bit depth), which will be more forgiving if your camera optics is not quite ideal. For high speed testing, 12 bits is typically used and for low frame rates 14-16 bits are common. We use Phantom v2012 cameras in our setup (12 bits) but I know those are expensive. Low frame rate cameras like the ones by Allied Vision (and others) are usually much cheaper, but you'll miss the flow dynamics. For illumination, we use a pulsed UV LED at 405nm, driven by a custom circuit. The basic circuit is provided in Wilert C et. al. (2012), "An assessment of high power light emitting diodes for high frame rate schlieren imaging" Hope this helps!

@jorgemunoz6973 2 жыл бұрын

Hi @@fzigunov , thank you very much for so insightful explanation. I apologize for the delay replying you, but I was checking the technical specifications of some cameras we had at the Lab... Indeed, we have a quite old camera, MotionPro HS-4, with ~5000 fps at 512x512, and an image bit depth of 24-bit? (sorry but I don't understand pretty well the latter...)

@fzigunov 2 жыл бұрын

@@jorgemunoz6973 Jorge, that camera should be plenty for these measurements. I suspect it's not 24 bits, though, but that's ok. Not sure if that's the part you didn't understand, but I'm talking about the effective dynamic range of the pixels in the camera. The maximum dynamic range possible is 20log10(2^nbits), but usually it's a little less. It's basically the ratio between the maximum signal you can get vs the minimum change you can detect. In any case, you should be able to get sufficiently good images with that camera, no problem.

@jorgemunoz6973 2 жыл бұрын

@@fzigunov Thanks a lot! Then, that sounds great!

@user-un1ws2qg6l 2 жыл бұрын

Thank you,your video is perfect. when i turn on wind tunnel the paint start to separating from surface and i know that i have to use of base coat but i don't know which base coat is suitable and can increase adhesive of polymer on surface. i use of PtTFPP=Platinum meso-tetra(pentafluorophenyl)porphin and FIB polymer(poly-heptafluoro-n-butyl methacrylate-co-hexafluorisopropyl methacrylate) or FEM polymer(poly-tifluoro-ethylmethacrylate-co-isobutylmethacrylate) thank you please guide me.

@fzigunov 2 жыл бұрын

We used the IBM polymer in this recipe, but I've seen recipes with the FIB polymer stick to Aluminum just fine. Is this a high shear flow like a M>2 supersonic flow? The paint presented in the video has been tested in M=2 flows with no shear degradation. For some of the polymers like the FIB you may need to bake the base coated part in an oven for 30 min, in some recipes. I don't know which formulation you're using, but I recommend you don't start coming up with your own recipe, as there's a lot of trial-and-error and the ones described in the literature are well-tested and characterized. Unless, of course, you have the equipment to characterize your paints (pressure chamber, shock tube, resonance tube).

@user-un1ws2qg6l 2 жыл бұрын

@@fzigunov thank you for your perfect answer.i changed the polymer and i want to use of PMMA or PMSP. do you know suitable basecoat for this polymer to increase adhesion?

@paille-boy 3 жыл бұрын

Hopefully there gonna be a more visible to eyes paint like this (so I can put it on my car)

@fzigunov 3 жыл бұрын

Hahaha I wish it was that easy!! I'm supersonic flows, you can see easily with your eyes, though. So you just need a supersonic car =]

@kavu4487 2 жыл бұрын

What materials is this model?

@fzigunov 2 жыл бұрын

Hi, KAVU, the model is made of 6061 aluminium. We've had success with 3D printed models (PLA), as well as acrylic as substrates for the PSP, in case you're interested.

@kavu4487 2 жыл бұрын

@@fzigunov That’s very helpful , I’m glad to hear from you!!!