how to use the proper interrogation area

[Vipferncb]

hello, I am beginning to use PIV and I use PIV to measure jet flow that has a velocity of about 10 m/s but the result is not good, Here is my video. 

https://drive.google.com/file/d/1NOx_5IQcQ4nc3cyONY8COsciIoz660Zz/view?usp=sharing    

Thank you in advance

[William Thielicke]

There seems to be a bug with some image preprocessing settings when using color video files.

Please disable the "auto contrast stretch" and set the min/max to 0 and 1. Then the standard settings should already give ok results. However, the image data is not really optimal for performing PIV.

[William Thielicke]

This bug will be fixed in the next release.

[Vipferncb]

Thank you very much for your help, can I ask you one question? 

The sentence that you said '  the image data is not really optimal for performing PIV.' it means like video quality, amount of smoke, the intensity of the laser, etc. are not proper to use PIV right? and what is the biggest scale of the picture that is proper to use PIV method?

Sincerely

Thanks for your suggestion in advance.

[William Thielicke]

Hi,

yes all that is not optimal. The camera is bad (high compression artifacts and low resolution). The seeding is way too dense and not covering the whole image. You shouldn't heat up oil to create a seeding, but you should shoot pressurized air through a nozzle to create larger particles. The laser sheet (is it actually a laser?) doesn't seem to be thin enough.

Here are some of our products that solve these issues:

https://www.optolution.com/fileadmin/user_upload/Geraetetechnik/PIV/Seeding_Generator_Datasheet.pdf

https://www.optolution.com/fileadmin/user_upload/Geraetetechnik/PIV/Light_Sheet_Optics_Datasheet.pdf

https://www.pco.de/fileadmin/user_upload/pco-product_sheets/FL_PCOPANDA26DS_V101.pdf

https://www.optolution.com/fileadmin/user_upload/Geraetetechnik/PIV/SimpleSync_Datasheet.pdf

https://www.optolution.com/en/products/piv/

[William Thielicke]

Maybe tell us what equipment you are currently using.

[Sachio Isaiah Otsuka]

hello, 

if the calculated pixel displacement is 10px, then by the rule of thumb, the interrgoation window size must be greater than of equal to 4x the pixel displacement, which means the window should be greater than 40px by 40px. However, when selecting the multipass feature in PIVlab, what should the lowest size be for the final pass? Will i get bad results if the last window size is less than 40px?

thank you

[William Thielicke]

Hi, These rule of thumb are ok as a start, but they do not 100% apply to multipass deformation algorithms. For the last pass, there should be around 5 particles per interrogation area. The passes before can be much larger, also larger than 4x the displacement. There still doesn't seem to be a good rule, most is based on trial and error. Just always keep in mind: the larger the interrogation area, the lower the noise.

[Sachio Isaiah Otsuka]

thank you very much!

[Sachio Isaiah Otsuka]

Good day, Thielicke
Thank you very much for your help. We have finished our experiment for our thesis and now we are finishing up our final manuscript.
Would it be alright if we could cite you in our final paper? 

Regards,
Sachio

[William Thielicke]

Yes 

[Sachio Isaiah Otsuka]

Thank you very much!