Experimental Physics and
Industrial Control System

Mark Rivers <[email protected]> · Thu, 31 Mar 2016 11:56:00 +0000

I have been working with this person at Tsinghua University to solve a problem with a Point Grey camera.  We have been communicating privately since their original tech-talk message.  Starting last night all of my messages are bouncing on their end.  So if anyone else at Tsinghua is reading tech-talk please let this person know there is a problem with their e-mail.  And this is the reply I am trying to send them.

**************************************************************
Hi,

OK, those ImageJ pendIO errors are what I wanted to see.

The problem is that you need to set the environment variable EPICS_CA_MAX_ARRAY_BYTES to be somewhat larger than the maximum image size you want to send across the network.  Your camera is 1288 x 964 monochrome.  That is 1241632 bytes.  You could set EPICS_CA_MAX_ARRAY_BYTES to 2000000 or even 10000000.  Don't make it much bigger than is needs to be (like 100 MB) because it will waste memory.

That needs to be set in both the IOC process and the ImageJ process.  So your should set it in your shell login script.  You will need to restart the IOC and ImageJ after setting it.

All of this is discussed in my online documentation here:

http://cars9.uchicago.edu/software/epics/areaDetectorViewers.html#Troubleshooting

and in the Configuration section of this document:

https://github.com/areaDetector/areaDetector/blob/master/INSTALL_GUIDE.md

Mark

________________________________
From: 彭毛毛 [[email protected]]
Sent: Wednesday, March 30, 2016 9:12 PM
To: Mark Rivers
Subject: RE: RE: RE: RE: RE: The camera cannot work when I start the IOC

Hi , Mark !
    I found it's OK when I set Packet size to 1440 and also the GigE packet delay has been changed to 800.
In the IOC ,There are errors only when init it . The errors are like this:

# Records with dynamic enums need to be processed again because the enum values are not available during iocInit.
zdbpf("13PG1:cam1:Format7Mode.PROC", "1")
st.cmd line 84: Command zdbpf not found.
dbpf("13PG1:cam1:PixelFormat.PROC", "1")
DBR_UCHAR:          1         0x1
# Wait for callbacks on the property limits (DRVL, DRVH) to complete
epicsThreadSleep(1.0)
2016/03/31 09:31:49.113 13PG1:cam1:FrameRate devAsynInt32 process error
# Records that depend on the state of the dynamic enum records or property limits also need to be processed again
# Other property records may need to be added to this list
dbpf("13PG1:cam1:FrameRate.PROC", "1")
DBR_UCHAR:          1         0x1
2016/03/31 09:31:49.611 13PG1:cam1:FrameRate devAsynInt32 process error
dbpf("13PG1:cam1:FrameRateValAbs.PROC", "1")
DBR_UCHAR:          1         0x1
dbpf("13PG1:cam1:AcquireTime.PROC", "1")
DBR_UCHAR:          1         0x1

However ,the whole software works fine .
In the ImageJ , the errors are here
[cid:467acfab$1$153ca718225$Coremail$pmm15$mails.tsinghua.edu.cn]
But every time I closed the image showing window ,it will open again .So I suppose ImageJ should have received the picture.
I have set shutter and Gain to manual in the pointGreyProperties screen and exposure time to 0.1 anf Gain to 0. I also changed the Maximum to many values from 50 to 10000 ,but the image is still dark . Does it possible that the arrays IOC outout is all zero ?
--

祝好！

彭毛毛

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System