Acquisition rate is half of trig rate, why?

[TiTou]

Hi all,

I'm having a "funny" issue!

I have a Basler A504k camera with a PCIe 1433 and I'm using an external trig source.

In MAX, when I set external trig acquisition with frame size 1000*128 pix and set the external trig source at 3900Hz, MAX says the display rate is around 500Hz and the acquisition rate is 3900Hz, exactly what I expect.

Then I go to LabVIEW (zip attached contains a simple VI + llb to communicate with the camera) and try to do the same, he acquisition rate is only 1950Hz, then if I change the frequency of the external trig source to 7800 I get 3900Hz frame rate in my VI.

I'm assuming there is a setting either on he camera or the framegrabber does that but I can't find it.

Running Win7 64bit + LabVIEW 2011 SP1f1 64bit, latest VDM + VAS in 64bit as well.

Any clue?

[BlueCheese]

Any reason you appear to be configuring the frame rate limit on the camera even though you are triggering it externally to drive the framerate? I wuld suspect that if you limited it to the exact same rate as what you are triggering it, if the external trigger is coming even a nanosecond faster each period than the camera's configured rate it will cause it to skip every other frame as it would violate the limit you configured.

Eric

[TiTou]

---

@BlueCheese wrote:

Any reason you appear to be configuring the frame rate limit on the camera even though you are triggering it externally to drive the framerate?

---

I only set the frame rate when I set the camera in free-run mode. When I set ExSync (= truggered) mode I'm skipping the "set frame rate" command.

Now, how would the camera behvae if I first set it in free-run mode with a frame rate of 500 fps and then set it to triggered mode without changing the frame rate and set my trigger source higher then 500?

If in that case the frame rate would be used as a limit that would be a sly feature that I would expect to find in the documentation of the ExSync mode.

I've run the test it doesn't work like that. I've set ExSync mode and a 3950 frame rate, I still a frame rate of half the trig rate 

Any other idea?

[TiTou]

Humm, I've been testing more and here's what I'm seeing :

- Trig source at 2500Hz or lower, acquisition rate in LabVIEW is the same as the trig source.

- Trig source at more than 2500Hz, acquisition rate in LabVIEW is half the trig source, and from here withtout stopping the acquisition, i f a lower the trig source there is an hysteresis, I have to go as low as 1500Hz on the trig source to have my acquisition rate equal to my trig rate.

In MAX, during the acquisition I can set whatever frame rate I want between 1Hz and 4000Hz, MAX acquires at the same rate as the trig.

Why can MAX do it fine and my LabVIEW code not?

To confuse thins further, my LabVIEW code does the acquisition fine when I'm on Win XP...

[TiTou]

How much fun...

Before my coffee break, I thought I would automate a little test routine to graph actual acquisition rate vs trig rate.

Here's the test routine :

And here's the result trig rate on X axis and actual acquisition rate on Y axis :

So if I want to be able to acquire between 3 and 4 kHz I have to trig at twice the acquisition rate I want, but if I want a less than 2 kHz I have to trig at the rate I want.

[S.Schulte]

Hello TiTou,

why are you using a trigger frequenz from 3 or 4 kHz?

The aquisition rate of the Camera is only 400-500 fps.

Is there a special reason?

Regards,

Basti

[TiTou]

Sorry I thought I mentioned that the camera's max framerate depends on the number of lines per frame, if full frame (1280*1024) max is 500Hz but if only 128 lines (like I do), max is 4kHz.

[Christian Mergl]

TiTou,

are you still working at this issue?

I'm just wondering if you could measure the acquired FPS using "IMAQCalculateFPS.vi" located under [LabVIEW]\examples\IMAQ\IMAQ High Level.llb

Christian

[TiTou]

I'm currently off work because of a dislocated shoulder but I'm still looking for a solution (there is a support request open Reference#1859641).

I have no objection to try any idea you might have but in what way this VI is different from what MAX displays - in fact I believe MAX uses this VI and I also believe that the code I posted above uses a similar way to determine acquired FPS.

I should be back to work next monday to try this anyway.

Thanks for the suggestion.

[Christian Mergl]

Unfortunately I don't have access to the SR.

Please post back as soon as you tested it. I should have a 504k somewhere, so if necessary I can setup a system to test this behaviour.

Christian