Current News
Status, June 9th, 2005
WFI takes Image of SN 2005cf.
See
Rochester Super Nova Site
1. SN2005cf data taken from WFI FPA
Status, May 18th, 2005
WFI Returns.
The Wide Field Imager was packed up yesterday (Tuesday) and is now on its
way back to Siding Spring.
It wil be pumped and cooled tomorrow (Thursday and should be available for
observing by Friday evening this week, May 20th.
As reported below - there are now 7 of the 8 detectors working.
No further work will be attempted on the Focal Plane in the forseable future
unless a replacement detector of suitable scientific merit can be found
as a replacement for CCD-06.
WFI would have to be brought back to RSAA again for this work to be undertaken.
Status, May 11th, 2005
Red Herrings.
After efforts this week to operate the other output amplifier on WFI
CCD-06, it now appears that this CCD is in fact non-functioning.
A close inspection of the associated Flex cable for this device revealed
cracks on the circuit where this connects to the pins of the CCD connector.
This break looks to have begun - probably when the cable was first installed
in 1998/9. It is thought that the recent failure of the camera may have exacerbated
the final failure of the cable.
This failure in the cable may then have been responsible for the damage to the op(A)
amplifier. op(B) may also have been affected by this - as it too is connected
to some of same signals as op(A).
A replacement flex was installed on Monday in the hope that this would allow us
to operate op(B). After effort spent in modifying the DSP code to permit the other
corner to be read, it was clear that the DSP code design would not allow us to easily
do this. So yesterday we swapped over all the connections on the external socket -
which effectively allowed us to operate op(B) from the op(A), original DSP code.
This did not result in any clear improvemt to the images you see below. The CCD-06
image looked to be a low-gain, noisy frame - as we had before.
One final test was performed this morning to verify there was not a problem with
the Video hardware - I swapped the CCD-06 and CCD-07 video channels over - the
result was the the image from CCD-06 moved to the channel for CCD-07, no
improvement was seen.
In addition, it now looks as if what we are really seeing on CCD-06,
appears to be a cross-talk component from one of the other CCDs!!
See the image below-
-NoVid_Conx-ico.jpg)
1. 8-element pin-hole image, No Video
connection to CCD-06, showing X-talk from CCD-04
So, to test this, I have just read-out this channel without a CCD
connected and have seen a faint pin-hole image. On blanking off the
pin-hole to CCD-04, the top left detector, the faint image on CCD-06
disappeared.
This type of cross-talk between detectors and their associated electronics
is well known and indeed the video processing channels for these 2 CCDs,
4 & 6 are within 2cms of each other in the ARCII (Slave) controller.
It therefore seems likely that this type off cross-talk also exists
between similar pairs of detector in WFI, eg between 5 & 7 on the same
controllers.
It has therefore been decided to abandon any further work, as it now
appears likely that CCD-06 has suffered due to the flex cable failure.
It is not clear from any information I have whether op(B) ever worked,
indeed 1 of the 2 of the potential replacement CCDs sent to us by Gerry,
has only one working amplifier. So it may be that op(B) never worked, or
if it did, also failed at the same time as op(A).
My thanks for help with all this work go to-
Gerry Lupino for help with spares for testing and discussion
about the WFI FPA
Barry Burke (of MIT) who designed the CCID20 Detector which
is used in WFI
Mark Downing for help with the original DSP code for WFI
Mat Doolan and Helmut for mechanical support for the test system.
Hans and Andre for electronics technical support
Mark Waterson & Mike Blacksell for useful discussions and help
with the connector swap
Col Vest and Robbie for help with the WFI camera
Peter Walsh for getting WFI here and taking it back next week!
Status, May 5th, 2005
More indication that CCD-07 is not dead
The Image below is a ~5sec exposure of a set of 8 pin-holes taken with
the Wide Field Imager.
As you can see, the remaining detector - CCD-07 - is showing more signs of life,
given the longer exposure. The 2 frames were taken from amplifiers A, then B
after installing extra signal paths for this. The problematic Bias level still
persists and hence these frames show what amounts to a low-gain, noisy image.
Cutting the common RD connection to the op(A) & op(B) sides of the chip,
on Monday, when the camera has warmed up will hopefully restore the operation
of the op(B) amplifier.
_Aftermods-ico.jpg)
1. 8-element pin-hole image, op(A)
_Aftermods-ico.jpg)
2. 8-element pin-hole image, op(B)
The 2nd image is further confirmation that removing the commoon RD connection between
the op(A) & op(B) will restore the detector's operation.
It will then be a matter of assesing whether the imaging capabilities of the device
are in fact useful scientifically.
Status, May 4th, 2005
Some indication that CCD-07 is not dead
The Image below is an exposure, ~0.5sec of a set of 8 pin-holes taken with
the Wide Field Imager.
As you can see, the remaining detector - CCD-07 - is showing signs of life!
This was achieved by disconnecting the Bias supply to the op(A) amplifier,
the one believed now to be faulty (though I am confirming this with Barry Burke
at the moment.)
If you squint at the image - you will just be able to make out features in the
defective CCD - similar to the ones in the working CCDs.
-ico.jpg)
1. 8-element pin-hole image, showing CCD-07 is responsive
This image indicates that it is worth attempting to remove the connections to
the op(A) amplifier so that the fault does not propagate to the other side.
The Reset Drains on these devices were made common in the early design and so
the only way to separate the output amplifiers is, to once again, remove the front
window to gain access to the internal panel where this can be done. The risk is low.
A few more cold tests will be made today and tomorrow and then WFI will be allowed
to warm up over the weekend so that the internal parts may be accessed on Monday.
From here it should be a simple matter to confirm by taking more pin-hole exposures
when cold, that the switch to the other amplifer(B) has now worked, having now
dis-connected amplifier(A).
If amplifier(B) proves unsuitable for sceintific use, WFI will be left in this state
and sent back to Siding Spring, pending a possible future change if a suitable device
is made avaiable.
Status, April 28th, 2005
Focal Plane Re-installed
The WFI FPA was tested at Room Temperature to investigate whether the
2nd of the non-working CCDs (CCD-06) could be brought back to life
using a new flex cable.
However, both this and the existing part, which was also tested, show,
that the output amplifier which is currently used (and is used on all the
WFI FPA CCDs), namely op(A), has indeed failed. The room temperature
test using independant DSP code shows that op(A) is non-working and op(B)
looks nominal. The image below (#2) confirms this.
Therefore WFI was put back together late last week and pumped ready for
a cold test, which was performed yesterday.
As you will see from the image here - the 8-element pin-hole test shows that
All 7 detectors are still functioning after removal of the FPA and the Room
Temperature test, 'on the bench', all of which is very re-assuring!
The following image again shows a ~0.5 sec exposure at 180K using an array of
8 pin-holes. Images show illumination changes as I didn't get the front
cover on uniformally (I am the shutter hardware in this instance!)
for the ~0.5sec exposure during the 5sec integration! WFI is therefore
being used again as the most expensive 8-element pin-hole camera in the world?
1. Another 8-element pin-hole image from WFI
As you will be able to see, CCD-06 is not imaging, the signal level on this frame
is at the bias level, about 450adu. The absence of a pin-hole image on this device
being due to a failed read-out amplifier. My room temp Split Read out of CCD-06
shows at least one of the amplifiers working - probably the one which is not used
by default and which has a question mark against its suitability in the DSP timing
code from the early 90's
It is therefore now felt that the 1st detector has started working due to
removal and re-installation of its flex cable.
This does not appear to have worked for the 2nd device - which at this
stage looks to now have a non-working op(A) amplifier.
We therefore plan to swap output corners over the next few days, by
installing new video wiring for that side. With a small modification
to the old timing (DSP) code we should then be n a position to finally
determine whether this device is useful scientifically or not.
The plan would then be to get WFI back to Siding Spring for about mid-May.
Status, April 14th, 2005
The Testing of WFI is progressing and as reported last week, the 1st,
(CCD-04 - top left hand) device which was removed and was working in the
test system - normally, is now back in the focal plane without any problems
arising in the installation!
Also a Room Temp R/O of this and the CCD which wasn't removed (CCD-06) indicate
that the former is working as expected whilst the latter is still showing only
1 amplifier working - which also may be nominal.
We now have the FPA back in WFI - it was pumped last night and the vacuum performance
is good. It is about to be cooled and tested to see what state we are now in.
Further testing will then be done using the test system hardware to try and read each
of the 2 devices independently of the WFI ARC control hardware. We think we can do
this by swapping the ARC II for ARC III timing, clock and video boards and using that
as a test bed with the devices operating normally in the focal plane.
1st WFI Science device Operating Normally In WFI
The CCD-04 device now appears to be working normally in the WFI FPA
The following image shows a 0.5 sec exposure at 170K using an array of
8 pin-holes. The small images are of the Electronics lab - all 7 the same
illumination changes as I didn't get the front cover on uniformally in the
0.5sec exposure during the 5sec integration!!!
WFI is therefore being used as a very expensive 8-element pin-hole camera!
1. 8-element pin-hole image from WFI
As you will be able to see - CCD-06 is not imaging, the signal level on this frame
is at the bias level, about 450adu. The absence of an pin-hole image on this device
may be due to a failed read-out amplifier. My room temp Split Read out of CCD-06
shows at least one of the amplifiers working - probably the one which is not used
by default and which has a question mark against its suitability in the DSP timing
code from the early 90's
I plan to try and swap the read-out port next to see if this remedies the fault. If it
does then that would be the end of the testing work - if it does not then I can try to
use my own DSP timing code to see if I can at least get back to the read-out performance
when tested at room temperature, with the FPA in a black box here in the clean room, see
following picture-
2. Room Temp Split Read-out of the WFI CCD-06
One unanswered question- why is the 1st device now working
removing and re-installing the device may have cured the problem,
I do not know!
Status, April 6th, 2005
1st WFI Science device Operating Normally
The WFI CCD-04 (Serial number 10-15-1) device has been tested and found
to be working normally in the test system dewar.
The figure below illustrates the spectral response for this device along
with what are thought to be reasonable errors on the measurements
Comparison data is also provided from the new E2V Imager CCD at SSO.
1. Spectral Response for the 10-15-1 WFI CCD.
The next two pictures show 1300 second dark frames, the first taken
here yesterday in the test system and following this an image provided
by Gary Da Costa from Dec. 2000 at the AAT.
If these images are compared, one can see identical features in both images
and the well known, low-level (<10adu) bias structure in the Test system image.
The CCD in question on the AAT image is the one at the top left-hand corner.
2. 1300s dark frame taken at -114C, RSAA, April 2005.
3. 1300s dark frame taken at the AAT, December 2000.
Finally, any doubters?? The fourth image is a 1 second exposure using the Test Box
grid pattern to take a reasonably well focussed test pattern image..
4. 1s exposure of Test Pattern.
Status, April 4th, 2005
1st of WFI CCDs Removed from FPA
I have now removed the 1st CCD from the WFI focal plane without too much
of a hassle and the device has now been installed in the test system.
This is device CCD-04, 10-15-1, top left-hand side looking into the camera
see the image at the bottom of this page.
The dewar is currently being evacuated and a cold test will happen either late
today or tomorrow. I have to say that a room temp read-out looked OK !!
Similar to what I have had from the Gerry Lupino replacements. But the real
test wont happen until the device is cold, later today or early tomorrow.
On the subject of the Luppino CCDs - here's what Barry Burke at MIT Lincoln Labs,
who designed the device, had to say about the 2 supplied to us..
"The MBE devices from lots 17 and 18 are dogs at low temperatures.
This was discovered back in 2002/2003 after they had been fabricated.
It turned out that the MBE layer on the back surface was not properly deposited,
and we did not find out until it was too late....""
Pictures during the Removal of the FPA CCD
and subsequent installation of the device into the Test System
Removal from WFI FPA
1. WFI FPA with the 10-15-1 (CCD-04) removed.
2. CCD 10-15-1 removed.
3. Underside of Focal PLane .
4. Detail of Focal plane with CCD removed.
5. Underside of the 10-15-1 CCD.
6. 10-15-1 surface.
Installation into Test Dewar
1. Installation of CCD 10-15-1 into Test Dewar.
2. Removal of 17-4-6, 2nd of new devices form Test Dewar.
3. Detail of DMB with No CCD mounted.
4. Two Luppino supplied CCDs and WFI CCD on bench.
5. 17-4-6 and WFI 10-15-1 CCD on bench with
Flex cables about to be swapped over.
6. Devices removed from Detector Mount Block and
protective mount respectively.
7. Detector Mount Block and Protective mount
swapped, in preparation for device installation.
8. Both devices now installed, 17-4-6 in Protective mount
10-15-1 mounted in the DMB ready for installation into Dewar.
9. Flex cables being swapped so that the WFI device
gets connected to a good, known to be working flex.
10. 17-4-6 CCD having flex installed.
11. 17-4-6, flex installed.
12. Detector installed in Test Dewar, Flex unconnected.
13. Detector installed in Test Dewar, Flex now connected.
14. Luppino supplied CCDs back in the protective AS box.
15. Top view of 10-15-1 CCD on DMB in Test Dewar.
16. Close up of CCD nanonics connector and bond wires.
Status, March 30th, 2005
No changes to WFI FPA
After further testing of the 2nd Lincoln Labs MBE CCDs, 17-4-6, (which will
continue till the end of this week (1st April!!), it appears extremely
unlikely that these will replace either of the 2 non-working WFI CCDs.
Neither of the test devices has shown adequate scientific performance to
be utilised in the Focal Plane Array.
As stated below in the last Status report - the 2nd device exhibits extremely
poor Spectral response across the whole optical band!
I have contacted Barry Burke at MIT to discuss this matter further.
I propose we attempt to obtain at least two, preferably three further Lincoln
Labs CCDs which can be characterised and then WFI returned once more to RSAA,
and these devices installed into the focal plane.
This will not happen until later in the year.
In the meantime, one of the suspect devices will be removed from the Focal
Plane next week and characterised in the test system - to verify its operational
status. If, as expected, it is now no longer operating, the device will not
be replaced. The Focal Plane will be re-installed, tested and then returned to
SSO for normal observing, commencing on the 1st May.
The 2nd suspected, non operating device will be left in place, pending removal
if and when we obtain 2 new devices which are proved to be scientifically useful.
Status, March 17th, 2005. Device 17-4-6
The 2nd of the Lincoln Labs CCDs,
17-4-6, supplied by Gerry Lupino
is currently being evaluated. First indications are that this is a much better
(looking) device than the 18-1-3. A spectral response (QE) curve will be obtained
today along with a measurement of the read-out noise.
Oh Dear !!
The detector is performing so badly, that it was felt not worth while
obtaining complete optical spectral response data. The data is presented
in the following figure
Spectral Response for the 17-4-6 Lincoln Labs CCD.
The indications at the moment, are that the device has either an absent or defective
AR coating. The is being no Blue or UV response whatsoever..
Further, as the device response to white light is 5 times lower than the 1st
device - it appears that there is inefficiency in either the signal measuring or
the vertical or horizontal charge transfer...
The signs are that this device will not be useful as a replcement CCD for one
of the potentially defunct WFI CCDs
Status, March 14th, 2005. Device 18-1-3
The figure below plots the spectral response of the Lincoln Labs
18-1-3 CCD. For comparison, taken at the same time, is data from the E2V
Imager, after the device was cleaned by E2V last month. These numbers are
believed to be good to 10%. As you can see the 18-1-3 demonstrates very
poor response across the whole optical band.
Spectral Response for the 18-1-3 Lincoln Labs CCD.
Other device characteristics, 18-1-3
Good stuff
- Read noise 5-6e
- Cosmetics dont look too bad - see 60 sec pseudo-dark frame, see exposure #1, below
Bad stuff
- Only 1 amplifier appears to be working - see split read-out exposure, '#2', below
- Surface looks 'grotty'
- No data on pixels-to-pixel response variation.
Synopsis
On the basis of the above it is unlikely that this device will be useful
in WFI. Further - as its usefulness is limited, there is not a case for
removing one of the existing dead CCDs to swap for this device.????
Testing the dead CCDs may however be cause in itself to warrant their
removal from the FPA. We need to make a decision about this in the near
future.
The 2nd of the Lincoln Labs CCDs (17-4-6) provided by Gerry Lupino will be
Characterised this week. After this it should be possible to decide how to
proceed with the FPA detector testing, i.e. whether it is worth removing the
potentially dead ddvices at all - with the risk we take of damage to
the other devices for no apparent gain.....
Status, March 7th, 2005. Device 18-1-3
Lincoln Labs CCD is imaging at 163K
The 2 pictures below illustrate the imaging properties of the 18-1-3 CCD.
The image format for this device is -
- 2048 x 4096 imaging pixels
- 10 pre-scan pixels and and 90 post-scan pixels in X
- 104 post-scan pixels in Y
- Total frame size is therefore - 2068 x 4200
The first is a 60 sec dark frame showing-
- Dark current generated signal, mean frame background brighter
than overscan level for any particular row
- Cosmic ray events
- small amout of light leaking into dewar
- bright column on LHS which saturates the electronics and so
signal is seen in Y over-scan
- dead pixel and resultant partial dead column
1. 60 second dark frame on 1st WFI CCD.
The 2nd frame is a 200ms LED Preflash before a Bias exposure read-out showing-
- bright column on LHS which saturates the electronics and so
signal is seen in y overscan
- gradient in signal on LHS due to LED preflash light, as expected
- No LED preflash light on RHS, indicating signal is not being
measured correctly at the output node
2. LED Preflash exposure, split read-out
showing Right hand amplifier is non-operational.
This device will be quantified this week and a decision taken as to its
suitability to replace one of the defunct WFI CCDs
The 2nd of the WFI replacements will then be characterised and its suitability
determind for use in the FPA.
We have now removed the FPA from WFI and are putting together a procedure
and tools to remove the 1st of the defunct CCDs. This will then be operated
in the test system, now that this has been proved for Lincoln Labs CCD operation
CCD operation.
After this - we can decide what to do based on the suitability of the replacements
and the state of the 2 non-operating detectors in the WFI FPA.
The following picture shows the FPA, upside-down on the bench on its supporting legs
with the perspex cover in place protecting the focal surface. One of the heater resistors
has been removed showing the rear of the 1st of the non-operating CCDs. This part
will have to have a tool installed, the 3 cap-head screws released and the detector
lowered into its protective container. It will then be installed in the test system
and characterised....
Rear of the FPA, showing temp. servo components, one heater
removed to gain access to the rear of the CCD.
Status, February 25th, 2005. Device 18-1-3
1st Room Temp images from Lincoln Labs CCD
The first of the Lincoln Labs CCDs supplied by Gerry Lupino is now
in the test dewar. A set of Room Temp. images indicate that this device is
working properly. It will be pumped over the wekend and cooled on Monday
in preparation for Full imaging characterisation in the next week or two.
The DSP code running this device has been cloned form the IMager DSP and
this will need some work to ensure the video processing and analogue sampling
is optimum for the Lincoln Labs (rather than E2V) CCD.
Pictures during the installation of the 18-1-3 CCD
1. Detector mounted, cover plate in place.
2. Detector and detail of flex and temp connections.
3. Similar to above from another angle.
4. From top showing temp and flex component.
5. From front showing reflection of strange being.
6. From side showing connector again and detail of location of CCD.
7. Close up view of Dewar top plate in place.
8. Room Temp R/O showing saturated frame and Split Read
structure..
Status, February 23rd, 2005
The New Detector Mount Block is ready and is being vacuum tested.
Continuity of signals will be checked out today and the CCD installed
either today or tomorrow. The following images show the new DMB and
flex connectors - all ready to go. Temp. sensors work as do the internal
calibration LEDs..
Pictures during the installation of the Modified DMB
1. Detector Mount Block (DMB) showing flex connector.
2. Front of Dewar showing KPT Hermetic connector and Flex.
3. Detail of D-tpye and Flex connector.
4. Another shot of Flex details.
5. Underside of DMB showing cold straps, Temp servo components and
LED calibration sources.
6. Another shot of front of dewar showing LN2 can Temp sensor.
7. DMB D-type connector.
8. Overall view of front showing DMB and perspex cover plate.
9. Same as above with perspex cover plate removed.
10. Details of Temp servo and CCD connectors.
11. As above - overall view.
12. DMB and KPT connector.
13. Wiring to the D-type and Chip nanonics connector.
14. Test Dewar and Test controller on bench. Also shows internal
Dewar LEDs enabled...
Status, February 22nd, 2005
The Detector Mount Block is being modified to permit a safer way of
installing the detector, with the perspex cover plate in-situ. This
has taken a few more days than expected. We should have this new block
today (22/02) and I will then check out the connectivity to the Test
controller which has now been set up - ready to operate a single Lincoln
Labs CCD. If all this checks out, the detector can be installed on
Wednesady or Thursday and testing started...
Status, February 10th, 2005
First Detector in Test Dewar
The set of pictures below show the Test Dewar in preparation for the new
Lincoln Labs CCDs. The first is device 18-1-3, and the one with a very
dirty surface. The 2nd device, 17-4-6 looks much the cleaner of the 2
and this will be characterised later. NOTE - this is a dummy run, as it were
The detector still has its shorting plug in. The dewar and conections
have to be checked out on the Test controller before the CCD is connected
I hope to have this done early next week and the device operating soon
after that. We then have to decide how to clean the surface of this device.
Gerry has recommended pumping in the dewar but I don't want to do this in
the test system. We may have another alternative and an option to bake
the CCD to remove the contamination.
My thanks to Matt Doolan, Col Vest, Helmut, Ross and Rob in the mechanical
section for making the process seemless. To Hans and Andre here for
the electronics effort. And to all for being on 'call' when I needed technical
help with the whole thing.
Now all I have to do is make it all work!!
The last two pictures are the important ones showing the detector installed
but not connected
Pictures during the installation of the 1st of the new WFI CCDs
1. Front of Test dewar after installing DMB (Detector Mount Block).
2. Detail of top of DMB.
3. Detail of DMB from side.
4. Overall view of DMB in test dewar.
5. Directly into front of Test Dewar showing DMB.
6. Inside dewar showing temp. and CCD connectors.
7. Temp. and CCD connectors and LED calibration parts.
8. Detail of CCD connector.
9. Detail of Temp. connector.
10. Another overall view showing LED calibration sources.
11. DMB with installed perspex cover plate and Detector clamps.
12. Another view of Detector cover plate and clamps.
13. Straight into front of dewar showing all components.
14. Straight into dewar with LL CCD (#18-1-3) installed.
15. Close up of LL CCD with perspex cover plate installed...PHEW.
Status, February 7th, 2005
The Wide Field Imager will be brought back to RSAA this Wednesday/Thursday.
We now have all the parts required to run the new devices, supplied by Gerry
Lupino, in our new Test system. We have fabricated parts to allow us to
independently run these devices and characterise them for possible use in WFI.
This work is set to proceed in the period from now until the end of April...
The WFI CCDs which are non-operational will be removed after the initial work
to demonstrate the scientific viability of the new devices. The following picture
shows the new devices in the anti-static container supplied by Gerry
Both of these devices are marked 'Setup Grade' and so it remains to be seen
how useful these devices will be...
The Wide Field Imager for the AAT, 40 inch and 2.3m telescope.
This instrument consists of a suite of 8, Lincoln Labs, 2kx4k thinned
CCDs. These devices form a focal plane array (FPA), a mosaic of 2x4 detectors
- having an 8k x 8k pixel format. The FPA runs on 2, 4 channel ARCII controllers.
WFI suffered a setback earlier this year - a mechanical failure of the fill tube
resulted in a loss of vacuum and WFI was brought to RSAA for remedial action.
During the course of repair it was discovered that 2 of the detectors were
non-operational. Some preliminary investigations failed to discover the cause of
this and WFI was returned to Siding Spring with only 6 out of the 8 devices working.
Click Image for larger view
There is currently a plan to bring WFI back to RSAA at the end of January,
to investigate the cause of the failure of the 2 detectors. This work will
take about 3 months - there is a proposal to install new devices for these and
for a third device, which is showing evidence of delamination, at this time.
Latest News. 21st, December 2004
We have now received 2 new Lincoln Labs detectors from Gerry Lupino along with a new Flex
circuit for testing.
These devices will be tested independantly in the new RSAA test dewar running on a Leach
ARC III CCD controller. Once these have been proven to work satisfactorily, each of the
non-functioning WFI detectors will be removed and installed in the test system for
characterisation work. If these devices do indeed prove to have failed then we will
replace them, one at-a-time with each of the new devices.
It should be noted that these new devices aren't guarenteed
to be any better scientifically, than the original ones, though there may be an option to
obtain such devices from another source for use in WFI
Enquiries about the WFI Repair?
detman@mso.anu.edu.au
Click Image for larger view
Home
About Us
Observing
WFI
-NoVid_Conx.jpg)
_Aftermods.jpg)
_Aftermods.jpg)
.jpg)
