LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Message channel behaviour in IMAQdx

Highlighted

I have a sensor with an integrated GigE Camera from Photonfokus (MV1-D2048-3D03) which is based on a Pleora iPort NTx-Mini OEM module.
To get the configuration of the sensor the serial bulk communication of the Pleora iPort modul is used.
One way of this communication (camera TX) makes use of the message channel and GenICam Events.
As far as I can see the message channel is opened by IMAQdx when I start the acquisition and closed as soon the acquisition is stopped.
In my case I need the message channel over the whole time the camera session is opened.

 

If I use GEV-Player I can see, that the message channel is opened as soon as I connect to the camera.

 

Is the point where the message channel is opened configurable somehow in IMAQdx?
Is there a way to open the message channel manually?
Is the behaviour dependant on the XML camera description?

 

Many thanks for advice.

Tom

 

Photonfokus MV1-D2048-3D03
Pleora iPort NTx-Mini OEM Modul
LabVIEW 2015 SP1
IMAQdx 15.5
NI GigE Vision Adapter
Embeded Controller NI-PXIe-8135

0 Kudos
Message 1 of 4
(622 Views)
Highlighted

Oh, boy, a Forum Post involving both Message Channels (which I assume are Messenger Channel Wires) and IMAQdx involving a "foreign" Camera that may or may not be "validated" by NI for compatibility (as far as I know, Axis and Basler cameras are "supported").  Sounds like a situation that I've used before (including IMAQdx Events).

 

And such lovely VIs to examine and critique (not!).  What a disappointment -- I might have learned a thing or two, or I might have been able to teach a thing or two ...  I may check back in a day or two -- my recommendation is that you attach enough code that we can open your attachment and have few "missing VIs".  Note that if you need to attach more than a few VIs, the easiest thing to do is to compress the folder holding the Project and attach the resulting .zip file.

 

Bob Schor

0 Kudos
Message 2 of 4
(577 Views)
Highlighted

Bob, thank you for your time and your answer.

 

My question is about the "internal" behaviour of IMAQdx, the message channel is part of the communication between IMAQdx and the camera.

 

I have added an example with lot of comments to make clearer what I’m looking for.

 

I know that this is a very specific question and that it is difficult to use the example because of the hardware setup.
But my hope is that somebody with more insight in IMAQdx can give me a hint. Maybe I have overseen something in the documentation and there is a quite simple solution for this problem. Maybe there is an undocumented or hidden feature which controls the behaviour of IMAQdx regarding the message channel. Maybe a slight modification in the xml description file of the camera makes IMAQdx work as desired.

 

A short comment on the camera from Photonfocus:
The camera is listed in NI’s Industrial Camera Advisor and works quite well with IMAQdx.
I use the camera with a modified XML camera description file. The modification allows IMAQdx to know how to decode the bulk-specific GenICam Events. This change is validated with NI’s "CameraValidator" (see attachments).

 

Attached you find the Serial Bulk Example.vi, the Serial Bulk Example.icd file with the used camera configuration, the customised XML camera description file and the compatibility report.

 

Thanks for any hint towards a solution.
Tom

0 Kudos
Message 3 of 4
(553 Views)
Highlighted

Wow, and I thought I was having GenICam/IMAQdx problems!  [My camera has two sets of APIs from the vendor, an "old one" (only in C) and a newer one (in C++ and other stuff), with partial GenICam implementation.  Trick is, the most useful modes are only implemented through the low-level non-GenICam calls, and when I ask for help, the C++ code they provide is wrong (they don't understand the "bit" arrangement of the registers in their cameras, not a "byte-endian" problem, but a "bit-endian" issue].

 

Sorry, I don't have much helpful to add.  What I've found helpful is to "find someone" who has some experience that overlaps with mine -- in the case of my camera, it was finding someone with more C++ experience with whom I could test the Vendor's code and realize where/how registers were "locked", verify the "bit-endian" problem, and realize how to better map IMAQdx to get at the camera's innards.  Know anyone in your area with camera experience?  Sometimes a field-trip is invaluable ...

 

Bob Schor

0 Kudos
Message 4 of 4
(546 Views)