Paddy's Problem Page

Click here for full Image



Solving Problems and Asking Some Questions

This page is devoted to illustrating some of the effects or problems we see when we run our detectors. Anyone who would like to contribute please get in touch at the usual e-mail address- Email Me... apo@ast.cam.ac.uk

OR using the CCD mail exploder- Phooof... ccd-world@cfht.hawaii.edu

As a picture paints a thousand words here we go-

I am prepared to offer the facility to put images in here for discussion amongst the community - in this way I would hope that (any) common problems can be sorted out quickly. My email is given below. If you have any images which you consider worth drawing to the attention of the community - for whatever reason then-

Use of anti-Remenance and
anti-Blooming on EEV CCD42, 14/10/97

The 1st image shows a 1 minute dark frame with an anti-remenance routine running which ensures the chip is clear after a bright, over-exposure etc BUT with NO anti-blooming routine running-

Click here for full image

The second image shows the same 1 minute exposure of the EEV CCD42 with anti-remenance running but now WITH anti-blooming turned on. You will see the large number of 'pocket-pumped' charges, looking for all the world like cosmic rays. There appears to be rather a large number of these. Is this normal - anyone else found this level of pumped charge with any of their detectors...

We have also found similar results with Tek1024 and Loral 2k devices, so the effect isnt just EEV related, in case you are getting worried!!!!!

Click here for full image

Funny effect on Zero exposure Loral image, July 18 1996

Click here for full image
Click here for full image
I have just FTP'd these 2 binned images from our Loral thinned CCD from La Palma at the request of the Island staff to attempt to discover what is happening.

In a zero exposure frame, the first image, there appears to be a gradient in the lower part of the frame with the numbers as shown on the image. The numbers correspond to mean signal followed by the standard deviation as measured using IRAF, imexamine. As you will see in the first image there apeears to be a gradient, decreasing in level as you move up the frame over a region of about 300 pixels. The gradient extends into the horizontal overscan with no sign of a dis-continuity at the image area over-scan boundary. The overscan can be clearly seen in the image.

In the 2nd, 1-second exposure (these images are both taken on our double-beam spectrograph, ISIS) you can see the slit jaw at the top, but no evidence for a gradient in the light - particularly in the horizontal over-scan where on the first image it stretches completely across from one side to another.

We have tried swapping clock cards, controllers, pre-amps, pre-amp boxes which would then include the cabling and still get the same result. A resistive load placed at the heremetic chip connector on the cryostat in lieu of the CCD does not show this effect - just the usual, fairly noisy, flat bias.
This would appear to indicate that there is some subtle effect occuring on the chip but it looks electronic to me for all the world!!!.
Also if we window on the chip, no matter where we put the window the same 2-300 vertical gradient appears - apparently confirming that it must be electronic rather than chip related.....

We are running out of things to suggest for the Islkand staff to look at apart from getting the head back here - anyone any ideas/suggestions...

Mucho appreciado
Paddy

Marks seen on new Loral PPtF CCD, April 1996

Click here for full image

I have just taken this flat field image in red light (using a system devised by Simon Tulloch) from our recent PPtF loral CCD. The QE response for this device can be found at View Latest Loral Data However the flat field image reveals what look to be a multitude of 'scratches' in the horizontal direction. The Horizontal CTE does not appear to be all that good either, though I am in the process of trying to optimise settings to see if I can do anything about this. I wondered what the community think these 'scratches' might be and if they could affect CTE. The UV flooded devices I have looked at so far don't show any of these types of mark or degraded CTE.....

Implentation of Anti-Blooming, Simon Tulloch - Jan 1996

Click here for full image

We have had some problems implementing anti-blooming clock routines with our 1024 square thinned TEK CCD. We have projected pin-hole images onto our CCDs to demonstate that "wobble clocking" does indeed stop blooming in overexposed images. Unfortunately there are some rather unpleasant side effects. This 900 second dark frame shows several hundred bright spots and a region of increased dark current in a strip across the image centre. The bright spots have roughly gaussian cross sections and are approximately round. These are probably due to luminescence within the CCD. The parallel clock electrodes make their connections with the imaging area of the CCD at row 512, this is the row in which the anomalously high dark current is at a maximum. The anti-blooming clocks ran at 700Hz. Charge was shuffled between phases 1 and 2. Temperature was -100C. The brightest of the spots contained 5500 electrons. This charge was proportional to integration time and the anti-blooming clock frequency.

The Spotty CCD Syndrome, 1995

Click here for full Image

The image illustrated here was taken from one of our Tek1024 CCD heads, but the same effect prevalant in the image - the circular darkish disks with a brighter core, have been seen elsewhere using EEV CCDs. This is a false colour image taken with the tek1024 CCD with a few-hundred counts in the image. There appear to be a multitude of these spots, at low level - which flat-field out. Hence, I guess, the lack of complaint about the 'feature'.

What prompted my amazement was no-one had said anything to us before about this PLUS, the FALL (Autumn in english!) copy of CCD Astronomy, Page 15 has exactly the same sort of image used as an example of a Flat-field by Michael NewBerry, taken with an SBIG ST-6 and an 11nch Celestron Schmidt-Cassegrain.

I have already tried to ascertain their origin and I think they may arise from the optical flats used as windows in our heads. I assume there must be the same sort of material used in the SBIG camera in the article mentioned above!

The Poor CTE Problem - 1995

Click here for full Image

It has been reported that on of our Tek CCDs which we operate on La Palma are apparently suffering from poor CTE. These devices have been in operation for a while but inspection has shown that there is some transfer inefficiency in the horizontal direction which we did not see here when the device was set-up. We do not appear to have any information from Tek regarding the light levels used for their CTE measurements, they typically quote figures of 0.99995.

We have been trying to obtain some information from our archived data here, and it looks as if there may be some inefficiency, measured by ratio'ing the signal in lines of pixels in the overscan to that in a block of pixels in the imaging area just before the overscan.

The latest news is that our last camera Tek5 for the AUX focus on the WHT was measured by SITe at -45C NOT -90 and they have offered to replace it even though it was out of the 30 day guarentee period. This was a surprise on both counts as we expected a longer warranty period and that they were willing to replace it. As the CTE problem affects all our Tek devices we are assuming that it's our electronics (i.e. the sequencing) that may be causing the problem. we do not get anywhere near the quoted CTE (a la SITe) at low signal levels. Only at moderately high light levels do we see the quoted CTE being met.

We looked at the possibility that operating the oSW and oLG at the same clock levels as the oH and oV maybe the cause of the problem but this has not proved to be the case. Tek recommend operating these clocks at a volt or so different to the oH and oV.

Has anyone 

From Red Dwarf 3, for a titter...


Rimmer: Erm, I think we're all beginning to lose sight of the
real issue here, which is "What are we going to call ourselves?" erm,
and I think it comes down to a choice between `The League Against
Salivating Monsters' or my own personal preference, which is `The
Committee for the Liberation and Integration of Terrifying Organisms
and their Rehabilitation Into Society'.  Erm, one drawback with that
-- the abbreviation is `CLITORIS'.

Kryten: We could go to Dallas in November, 1963, stand on the
grassy knoll and shout "Duck!" ... I'm sorry; I must have bypassed my
Good Taste Chip.

I am... apo