Experimental Physics and
Industrial Control System

<[email protected]> · Fri, 28 Nov 2014 16:35:06 +0000

Hi Mark,

Just tried I15's camera, and I get the correct behaviour from ExposureTime and it seems to be producing frames with noise on it. I'm not running jumbo packets, and the camera is on the network with other devices. It does look like we have different cameras though... I've attached the screenshots of the EDM screen in case that helps. We have different camera names and resolutions, so there may well be other parameter differences. Have you tried running the makeDbAndEdl.py script as shown in the aravisGigE documentation?
http://controls.diamond.ac.uk/downloads/support/aravisGigE/0-1-15dls6/documentation/doxygen/index.html

Can you get data from any GigE Vision viewer? (We use Stemmer Imaging's common vision blox for testing our cameras before integrating them into Aravis)
http://www.commonvisionblox.com/

Also, which version of aravisGigE are you using? The one on GitHub?

Thanks,
Tom Cobb

> -----Original Message-----
> From: Mark Rivers [mailto:[email protected]]
> Sent: 28 November 2014 15:38
> To: Cobb, Tom (DLSLtd,RAL,TEC); Thompson, Jon (DLSLtd,RAL,TEC);
> Pedersen, Ulrik (DLSLtd,RAL,TEC); Rees, Nick (DLSLtd,RAL,TEC); Pratt,
> Linda (DLSLtd,RAL,TEC)
> Cc: [email protected]; 'Perceval Guillou'
> Subject: RE: Photonic Science X-Ray sCMOS camera system
> 
> Thanks for the help.  Using the fonts.list that Nick sent me (I think
> it is the same as the one Linda sent as well) fixed the edm display
> problem.  I don't get any messages about missing fonts, and the button
> and menu widgets now display the text correctly.  The text entry
> widgets are still too small, but Tom sent me a patch for that which I
> have not yet tried.  I've attached the screens I now get.
> 
> However, I am still having the problem that the data frames from the
> Photonic Science camera are all zeros, as evidenced by the statistics
> plugin.
> 
> Percival from Photonic Science suggested that this could be because I
> don't have jumbo frames enabled.  I don't seem to be able to enable
> jumbo frames on this computer, and I am not sure the GigE switch I am
> using supports them either.  Do you folks use jumbo frames with the PSL
> camera?  The parameters GevSCPSPackeSize is set to 1500, which should
> mean that it is not using jumbo packets?
> 
> Thanks,
> Mark
> 
> 
> -----Original Message-----
> From: [email protected] [mailto:[email protected]]
> Sent: Thursday, November 27, 2014 3:14 AM
> To: Mark Rivers; [email protected];
> [email protected]
> Subject: RE: Photonic Science X-Ray sCMOS camera system
> 
> I would have suggested the same things that the tech talk people have.
> Arial seems a particularly problematic font so you could try
> 
> sed -i 's/arial/helvetica/' *.edl
> 
> Also, there is a local mod in our edm to make the textupdate widget
> more space efficient which I've attached. Without it the textupdate
> widgets on this screen are pretty useless.
> 
> Thanks,
> Tom Cobb
> --
> 
> +44 (0)1235 778582
> 
> 
> > -----Original Message-----
> > From: Mark Rivers [mailto:[email protected]]
> > Sent: 26 November 2014 21:30
> > To: Thompson, Jon (DLSLtd,RAL,TEC); Cobb, Tom (DLSLtd,RAL,TEC);
> > Pedersen, Ulrik (DLSLtd,RAL,TEC)
> > Subject: RE: Photonic Science X-Ray sCMOS camera system
> >
> > I got the EDMDATAFILES path set, and now PSL_SCMOS finds the embedded
> > display.  However, as I posted a few minutes ago to tech-talk, the
> text
> > entry widgets and button widgets don't display any text.
> >
> > Any idea what is wrong?
> >
> > Thanks,
> > Mark
> >
> >
> > -----Original Message-----
> > From: Mark Rivers
> > Sent: Wednesday, November 26, 2014 1:01 PM
> > To: '[email protected]'; '[email protected]';
> > '[email protected]'
> > Subject: RE: Photonic Science X-Ray sCMOS camera system
> >
> > I managed to get edm running, though with lots of errors about fonts,
> > etc.
> >
> > I have attached the screen shot of PSL_SCMOS.edl and
> > PSL_SCMOS_features.edl.
> >
> > Any clue why my data is all 0?
> >
> > How do I fix the font problems?
> >
> > What do I need to do to fix the white area on PSL_SCMOS.edl?  I
> imagine
> > I need to specify a path for EDM to search for an included file?
> >
> > Thanks,
> > Mark
> >
> >
> > -----Original Message-----
> > From: Mark Rivers
> > Sent: Wednesday, November 26, 2014 12:35 PM
> > To: '[email protected]'; [email protected];
> > [email protected]
> > Subject: RE: Photonic Science X-Ray sCMOS camera system
> >
> > Hi Jon, Tom and Ulrik,
> >
> > I've got the PSL x-ray eye in my office, and running with the
> > aravisGigE driver.
> >
> > I am having some trouble however, and wonder if you might offer some
> > advice.
> >
> > When the IOC starts up I get the output shown in iocStartup.txt.
> >
> > I get a few warnings/errors before iocInit:
> >
> > 2014/11/26 11:58:10.345 asynPortDriver:createParam: port=ARV1 error
> > adding parameter Status to list 0, parameter already exists.
> > Done.
> > pthread_attr_setstacksize error Invalid argument
> >
> > and a number after iocInit:
> >
> > 2014/11/26 11:58:12.007 asynPortDriver:drvUserCreate: addr=0, cannot
> > find parameter Global_Shutter
> > 13ARV1:cam1:Global_Shutter_RBV devAsynInt32::initCommon drvUserCreate
> > 2014/11/26 11:58:12.007 asynPortDriver:drvUserCreate: addr=0, cannot
> > find parameter Image_count
> > 13ARV1:cam1:Image_count_RBV devAsynInt32::initCommon drvUserCreate
> > 2014/11/26 11:58:12.012 asynPortDriver:drvUserCreate: addr=0, cannot
> > find parameter Global_Shutter
> > 13ARV1:cam1:Global_Shutter devAsynInt32::initCommon drvUserCreate
> > 2014/11/26 11:58:12.012 asynPortDriver:drvUserCreate: addr=0, cannot
> > find parameter Image_count
> > 13ARV1:cam1:Image_count devAsynInt32::initCommon drvUserCreate
> > 2014/11/26 11:58:12.017 asynPortDriver:drvUserCreate: addr=0, cannot
> > find parameter ADC_range
> > ...
> >
> > As soon as the IOC starts up if I start acquisition it does start
> > acquiring frames at 7-8 frames/sec.  I have attached a screen shot of
> > ADBase.png in this state. You will see that although the requested
> > exposure time is 1 second, the readback is 0 seconds.  In this state
> > the frames contain all zeros, as proven by the statistics plugin.  I
> > can change the data type from UInt8 to UInt16 and the statistics
> plugin
> > sees the change, but the values are all still 0.
> >
> > I believe there is a bug in the AcquireTime record.  If I set
> > AcquireTime=1.0 I see the following with asynReport:
> >
> > Parameter 91 type=asynInt32, name=ExposureTimeRaw, value=1000000,
> > status=0
> >
> > If I try to acquire it never gets any frames in 5 minutes.
> >
> > However, if I set AcquireTime=0.001 I get the following:
> >
> > Parameter 91 type=asynInt32, name=ExposureTimeRaw, value=1000,
> status=0
> >
> > and I get exactly 1 frame/second.  AcquireTime=0.0005 then I see 2
> > frames/sec.
> >
> > So it seems like there is a confusion between ms and microseconds
> > somewhere.
> >
> > But even when I set an AcquireTime of 0.005, so I get 5 second
> > exposures the counts are always all zero with a radioactive source on
> > the front of the detector.
> >
> > When the system is acquiring frames at 1 Hz I did an asynReport on
> the
> > aravisGigE driver.  This is attached as asynReport_1Hz.txt.
> >
> > Any idea why the data are all 0?
> >
> > I only have the ADBase.adl screen, I don't have a GUI for any of the
> > aravisGigE specific parameters yet, since I am not running edm.
> >
> > Thanks,
> > Mark
> >
> >
> >
> >
> > -----Original Message-----
> > From: [email protected]
> > [mailto:[email protected]]
> > Sent: Friday, November 21, 2014 3:04 AM
> > To: [email protected]
> > Cc: [email protected]; [email protected]; [email protected];
> > [email protected]; [email protected]; Mark Rivers;
> > [email protected]; [email protected]
> > Subject: RE: Photonic Science X-Ray sCMOS camera system
> >
> > Hi Mark,
> >
> > We are indeed using the aravis gige driver for this camera.  It
> > basically works but there is an odd issue when starting an
> acquisition.
> > There is a random delay before the first image that appears that can
> be
> > up to a minute.  I have not yet had chance to do much investigation
> but
> > I did do some wiresharking.  It looks like the camera starts
> > immediately but the aravis library throws some images away.  I15 are
> > using the camera as a x-ray eye that runs continuously so this does
> not
> > bother them, hence the slow progress resolving this.
> > Jon
> >
> > On 21 Nov 2014 08:36, "Kleppe, Annette (DLSLtd,RAL,SCI)"
> > <[email protected]> wrote:
> >
> > Hi Mark
> >
> >
> >
> > Jon Thompson installed the photonic science camera for us in EPICS.
> He
> > knows which files we are using to run it and so I have copied him
> into
> > this email.
> >
> >
> >
> > Best regards,
> >
> > Annette
> >
> >
> >
> > From: Mark Rivers [mailto:[email protected]]
> > Sent: 20 November 2014 17:29
> > To: Kleppe, Annette (DLSLtd,RAL,SCI)
> > Cc: Cobb, Tom (DLSLtd,RAL,TEC); Pedersen, Ulrik (DLSLtd,RAL,TEC);
> > 'Ghose, Sanjit'; Dooryhee, Eric; Lewis, Wayne; Daniel Brau
> > ([email protected])
> > Subject: Photonic Science X-Ray sCMOS camera system
> >
> >
> >
> > Hi Annette,
> >
> >
> >
> > I received a copy of your message to Sanjit Ghose below.  I have
> > offered to help Sanjit get this detector working with EPICS.  I was
> > expecting that I would need to test it with the areaDetector Photonic
> > Science driver (ADPSL).  Photonic Sciences says that the camera is
> > GigEVision compliant.  Does this mean that at Diamond you are using
> the
> > aravisGigE driver to control this camera, and not the ADPSL driver?
> >
> >
> >
> > The aravisGigE driver comes with files called PSL_SCMOS.template,
> > PSL_SCMOS.edl, and PSL_SCMOS-features.edl.  Are these the files you
> are
> > using with this camera?
> >
> >
> >
> > Thanks,
> >
> > Mark
> >
> >
> >
> >
> >
> >
> >
> > From:
> [email protected]<mailto:[email protected]>
> > [[email protected]]
> > Sent: Thursday, May 01, 2014 10:08 AM
> > To: Ghose, Sanjit
> > Subject: RE: Information: X-ray eye camera from Photon Sceinces
> >
> > Dear Sanjit,
> >
> >
> >
> > we are very happy with our Photonic Science camera and I can
> recommend
> > it.
> >
> >
> >
> > I’m on beamline I15 (Extreme Conditions). I15 is a 3.5 T wiggler
> > beamline with a Si(111) DCM. We have tested and will start to
> regularly
> > use the camera for direct x-ray beam focussing down to about 70 x 70
> > microns and, with a 2nd set of small KB mirrors, down to 4-10 microns
> > at energies of 20-80 keV; maximum beam size we offer is around 5 x 5
> > mm. We haven’t had the need  and chance yet to use the Photonic
> Science
> > camera as a detector for diffraction but we do regard it as a spare
> > detector.
> >
> >
> >
> > Details of our Photonic Science X-Ray sCMOS camera system
> >
> > High resolution X-ray digital camera with direct coupled (micro)
> fibre-
> > optic input and cooled CCD. Camera features high Quantum Efficiency
> > (typically 55% @ 550nm), cooled CCD sensor with typical ∆T of > 40 C,
> a
> > range of standard input sizes, dual readout speed, on-chip pixel
> > binning, sub-area readout and up to 1920 x 1080 pixel resolution at
> up
> > to 16-bit digitisation. At highest quality setting the X-Ray sCMOS
> > produces images with the equivalent of 30,000 e- per pixel full well
> > capacity and greater than 14-bit S:N / actual dynamic range with 1. 2
> > to 2 electrons read out noise (less than 1 electron with
> interpolation
> > noise reduction).
> >
> >
> >
> > X-ray sCMOS system with 7.16 mm active input diagonal (micro) Fibre-
> > Optic taper
> >
> > · Active input area of 6,24 (h) x 3,51 (v) mm imaged on the sensor
> >
> > · Optical pixel resolution at the input of 3.25 micron (no binning)
> >
> > · Gadolinium oxysulphide scintillator layer density of 5 mg/cm2
> > optimised for resolution with x-ray energies of 5-60 Kev.
> >
> > · Input taper ratio of 1:2 (straight)
> >
> >
> >
> > Regarding your other questions:
> >
> > 1)      Photonic Science supports an EPICs driver for the camera. We
> > have the camera integrated into EPICS using the area detector
> interface
> > of EPICS.
> >
> > 2)      We have the software but we haven’t used it.
> >
> > 3&4 )We are happy with the dynamic range and efficiency at all
> energies
> > (20-80 keV) and it performs in that respect much better than the
> > indirect imaging systems we have / had.
> >
> >
> >
> > Hope this helps.
> >
> >
> >
> > Kind regards,
> >
> > Annette
> >

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System