The PIVlab result shows a lower flow rate than the actual measured flow rate

[Soccer Travel]

Hello, I got a good result from your advice last time.

The last problem remains, but there is a difference in the value of the average flow rate obtained by PIVlab than the actual measured flow rate.

The actual measured fluid is approximately 0.43 m/s, but the average flow rate obtained with PIVlab is approximately 0.2 m/s.

The delta T of the image pair used is 1.925 ms.

Please give me proper advice.

thakyou.

[thielickeoptolution]

Then your delta t or your reference distance are wrong... Please double-check (also with an oscilloscope and a photo-diode or LED inside the laser sheet).

[Soccer Travel]

Thankyou for your reply.

The actual length is the length and pixel measured by putting the checkered paper in the duct, so is there a better way?

And the actual delta T is 1.925ms, but the calibration's time step is not to enter delta T?
I'll be waiting for the reply.

Thankyou!

2024년 7월 15일 월요일 오후 7시 47분 26초 UTC+9에 thielickeoptolution님이 작성:

[William]

Maybe you have the raw data and the calibration image?

PIV cannot give wrong velocities. Only when the calibration is wrong it will give wrong results.

[Soccer Travel]

Thank you for your answer.
Instead of the current calibration image, we put the checked paper in the measurement area to reflect the pixel and the actual length.
Can you recommend another method instead of this one?
I look forward to hearing from you. Have a nice day

thankyou.

2024년 7월 17일 수요일 오후 10시 18분 30초 UTC+9에 William님이 작성:

[William]

Your data looks fine, I get the same result. But PIV does not have a bias of 100%, this will only happen if the timing  / scaling is wrong.

[Soccer Travel]

How should I modify timing/scaling?
I can't get a clue....

2024년 7월 18일 목요일 오후 5시 55분 7초 UTC+9에 William님이 작성:

[thielickeoptolution]

Validate the timing with an oscilloscope and a photo diode / LED.

[thielickeoptolution]

Does your calibration image and the PIV image have the same resolution? Maybe you apply pixel-binning somewhere...?

[Soccer Travel]

hello!

This is the result of checking with the oscilloscope.
Initially, delta T was set to 1s to verify that it appeared as 1s on the oscilloscope.
Lasers 1 and 2 are connected through channels c and d, respectively.
We proceeded by setting delta T at delay intervals of c, d  in the pulse generator.
In the oscilloscope below, the grid is set at 1s intervals.

[Soccer Travel]

[Soccer Travel]

The image below is when delta T is set to 2ms.

On the oscilloscope, one grid is set to 10ms.

So if we set delta T to 2ms,
In PIVlab, I thought that even if timestep was 2ms, it was not a different value from the actual one.

The resolution of the calibration image is no different.
I look forward to your advice.
Thank you for your kind reply.

[William]

Hi, thanks for this update. But I can only repeat: " Validate the timing with an oscilloscope and a photo diode / LED."

You are triggering the flashlamps of your laser. Ideally you would trigger the Q-Switches. When you trigger the flashlamps, then there can be additional errors. To be sure, you need to measure the light emission from the laser. This can easily be done by putting a LED or photodiode inside the laser sheet, then connecting it to your oscilloscope. The LED / Photodiode will emit a voltage when it is hit by the laser pulse. Then you can measure the delay between two pulses.

Another question: Has the calibration image be taken under the exact same conditions as the PV images? Have you aligned it with the laser sheet? Have you moved the camera after calibration?