Re: Ceduna phase stability
- Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]
From: <Brett.Reid_at_email.protected>
Date: Sun, 16 Aug 2009 21:06:03 +1000
Hi All,
I have replied to the whole VLBI group on this one since it is serious.
I was at Ceduna in June prior to the last LBA run with the purpose of
assisting with the maser service. We do believe the maser is OK.
I fixed the coax cables in the azimuth twister (including the 5MHz synch
cable) also.
Sorry I didn't get to the steps 1 and 2, back in June, at 1 GHz,
detailed by Simon below.
I would like to get the Agilent back to Hobart pronto for comparison
with the identical Hobart unit. OK?
I might also get the doubler and 10 MHz amp unit that connects to the
Agilent's ref input.
It gets very hot in the upper equipment room at Ceduna and I know the
operating environment temperature limit of the Agilent has been exceeded
many times. We now need to check this stress has not degraded our
oscillator's ability to perfectly synchronise. Comparison with the
Hobart unit at the higher frequencies will be simplest way to find out.
I am hoping Peter has some gear fast enough to perform these comparisons
in the lab.
For any one interested, the Ceduna Agilent LO freq for different sky
frequencies can be seen at the following link:
http://www-ra.phys.utas.edu.au/wiki/index.php?n=Ceduna30m.ReceiverConfigurationAndSystemTemperatures
It would be nice to know if the amount of phase instability corresponds
precisely to Agilent frequency at the different observing frequencies,
this will confirm that it is worth sending the Agilent over here for checks.
Regards,
Brett
Hayley Bignall wrote:
>Hi all,
>
>It looks like the Ceduna phase stability problem was still there in
>the July session..
>
>So far I have only correlated a small amount of data from v275c (as
>still have most of February to process).
>The attached plot shows much larger phase fluctuations on the Ceduna
>baselines compared with Hobart (although the S/N to Ceduna is also
>lower).
>
>Also Aquib has made a page at
>http://cira.ivec.org/dokuwiki/doku.php/correlator/diagnostics
>with links to plots showing about 15 mins of data from most
>experiments correlated so far.
>These show the clear frequency dependence of the problem - it's not
>significant at 2.3 GHz but starts to show up at higher frequencies and
>is severe at 22 GHz.
>(Also just a note for looking at these plots - ignore phases when the
>amplitude is zero as this is not meaningful data; the off source data
>may not have been flagged)
>
>Cheers,
>Hayley
>
>-------------------------------------------------
>Dr Hayley Bignall
>Curtin Institute of Radio Astronomy
>
>Location: Brodie Hall Building 611, Room 106
>1 Turner Ave, Technology Park, Bentley
>Phone: +61 (0)8 9266 9245
>Fax: +61 (0)8 9266 9246
>Email: H.Bignall_at_curtin.<!--nospam-->edu.au
>
>Post:
>Curtin Institute of Radio Astronomy
>Curtin University of Technology
>GPO Box U1987
>Perth, WA 6845
>Australia
>
>
>
>
>2009/6/3 Simon Ellingsen <Simon.Ellingsen_at_utas.<!--nospam-->edu.au>:
>>Hi Chris et al.
>>
>>>There seems to be a serious phase stability with Ceduna. Attached is a
>>>plot for v277b from Feb09 session. And remember that in Nov08 the data sent
>>>to Bonn seemed to have unstable Ceduna phases.
>>>
>>>Adam feels the phase stability is at such a level that Ceduna data is
>>>useless.
>>John, Jim, Peter McCulloch and I have had some discussions about this and I
>>just want to see what everyone thinks in terms of how best to proceed to
>>solve this issue.
>>
>>To summarize "the facts" as I understand them:
>>- This issue is known to have effected X-band experiments in Nov 08 (v271b)
>>and Feb 09 (v277b), also v257a (Aug08) at 6.7 GHz.
>>- The phase on calibrators varies by 10s of degrees on timescales of minutes
>>on baselines to Ceduna only.
>>- Both polarizations show the same temporal behaviour (this is true for the
>>plots for v257a that Hayley sent on 27 May).
>>- There was a similar, but different problem for vx014a (Mar 08) where there
>>was a rapidly changing phase difference between the two circular
>>polarizations, we believe that this was fixed prior to the Aug08 session by
>>swapping and repairing RF cables.
>>
>>Since the problem is common to both polarizations and seen for different
>>receivers, the only likely culprit I can think of is the clock/LO system.
>>Either problems on the cable/signal path which sends the 10 MHz signal from
>>the maser to the Agilent used for the first LO, or problems with the Agilent
>>phase stability. Does anyone else have any other ideas?
>>
>>Brett will be going to Ceduna in late June when the maser repairs are done
>>and so that is our best opportunity to find and fix the problem, however, it
>>is unlikely that we will have time between his visit and the July LBA
>>session to observe and correlate a VLBI test. This is how we propose to
>>look for the problem:
>>
>>1. Take a 1 GHz signal from an oscillator in the control room, split it,
>>send one part up the 10 MHz cable from the maser to the agilent, then back
>>through another RF cable. The put that signal and the other from the
>>original splitter into a vector voltmeter, put the output on a chart and
>>then run a fake LBA schedule for an hour or two and see what happens.
>>
>>2. Take a 1 GHz signal from the Agilent and a 1 GHz signal generated by an
>>oscillator in the control room and compare those on a vector voltmeter over
>>time.
>>
>>Any other comments or suggestions welcome. We are keen to fix the problem,
>>but it seems to be quite difficult to isolate and test as it is only
>>apparent in the processed data and isn't effecting fringes or anything
>>simple like that.
>>
>>Regards
>>
>>Simon
>>--
>>Simon Ellingsen : Senior Lecturer Physics & Astronomy, University of
>>Tasmania
>>email : Simon.Ellingsen_at_utas.<!--nospam-->edu.au
>>WWW :
>>http://fcms.its.utas.edu.au/scieng/physics/pagedetails.asp?lpersonId=681
>>Phone : 6226 7588 ; Area Code : +61 3 (International)
>>6278 8636 (Home), 6226 2410 (Fax) 03 (Australia)
>>
>>
>>
>>
>>
Date: Sun, 16 Aug 2009 21:06:03 +1000
Hi All,
I have replied to the whole VLBI group on this one since it is serious.
I was at Ceduna in June prior to the last LBA run with the purpose of
assisting with the maser service. We do believe the maser is OK.
I fixed the coax cables in the azimuth twister (including the 5MHz synch
cable) also.
Sorry I didn't get to the steps 1 and 2, back in June, at 1 GHz,
detailed by Simon below.
I would like to get the Agilent back to Hobart pronto for comparison
with the identical Hobart unit. OK?
I might also get the doubler and 10 MHz amp unit that connects to the
Agilent's ref input.
It gets very hot in the upper equipment room at Ceduna and I know the
operating environment temperature limit of the Agilent has been exceeded
many times. We now need to check this stress has not degraded our
oscillator's ability to perfectly synchronise. Comparison with the
Hobart unit at the higher frequencies will be simplest way to find out.
I am hoping Peter has some gear fast enough to perform these comparisons
in the lab.
For any one interested, the Ceduna Agilent LO freq for different sky
frequencies can be seen at the following link:
http://www-ra.phys.utas.edu.au/wiki/index.php?n=Ceduna30m.ReceiverConfigurationAndSystemTemperatures
It would be nice to know if the amount of phase instability corresponds
precisely to Agilent frequency at the different observing frequencies,
this will confirm that it is worth sending the Agilent over here for checks.
Regards,
Brett
Hayley Bignall wrote:
>Hi all,
>
>It looks like the Ceduna phase stability problem was still there in
>the July session..
>
>So far I have only correlated a small amount of data from v275c (as
>still have most of February to process).
>The attached plot shows much larger phase fluctuations on the Ceduna
>baselines compared with Hobart (although the S/N to Ceduna is also
>lower).
>
>Also Aquib has made a page at
>http://cira.ivec.org/dokuwiki/doku.php/correlator/diagnostics
>with links to plots showing about 15 mins of data from most
>experiments correlated so far.
>These show the clear frequency dependence of the problem - it's not
>significant at 2.3 GHz but starts to show up at higher frequencies and
>is severe at 22 GHz.
>(Also just a note for looking at these plots - ignore phases when the
>amplitude is zero as this is not meaningful data; the off source data
>may not have been flagged)
>
>Cheers,
>Hayley
>
>-------------------------------------------------
>Dr Hayley Bignall
>Curtin Institute of Radio Astronomy
>
>Location: Brodie Hall Building 611, Room 106
>1 Turner Ave, Technology Park, Bentley
>Phone: +61 (0)8 9266 9245
>Fax: +61 (0)8 9266 9246
>Email: H.Bignall_at_curtin.<!--nospam-->edu.au
>
>Post:
>Curtin Institute of Radio Astronomy
>Curtin University of Technology
>GPO Box U1987
>Perth, WA 6845
>Australia
>
>
>
>
>2009/6/3 Simon Ellingsen <Simon.Ellingsen_at_utas.<!--nospam-->edu.au>:
>>Hi Chris et al.
>>
>>>There seems to be a serious phase stability with Ceduna. Attached is a
>>>plot for v277b from Feb09 session. And remember that in Nov08 the data sent
>>>to Bonn seemed to have unstable Ceduna phases.
>>>
>>>Adam feels the phase stability is at such a level that Ceduna data is
>>>useless.
>>John, Jim, Peter McCulloch and I have had some discussions about this and I
>>just want to see what everyone thinks in terms of how best to proceed to
>>solve this issue.
>>
>>To summarize "the facts" as I understand them:
>>- This issue is known to have effected X-band experiments in Nov 08 (v271b)
>>and Feb 09 (v277b), also v257a (Aug08) at 6.7 GHz.
>>- The phase on calibrators varies by 10s of degrees on timescales of minutes
>>on baselines to Ceduna only.
>>- Both polarizations show the same temporal behaviour (this is true for the
>>plots for v257a that Hayley sent on 27 May).
>>- There was a similar, but different problem for vx014a (Mar 08) where there
>>was a rapidly changing phase difference between the two circular
>>polarizations, we believe that this was fixed prior to the Aug08 session by
>>swapping and repairing RF cables.
>>
>>Since the problem is common to both polarizations and seen for different
>>receivers, the only likely culprit I can think of is the clock/LO system.
>>Either problems on the cable/signal path which sends the 10 MHz signal from
>>the maser to the Agilent used for the first LO, or problems with the Agilent
>>phase stability. Does anyone else have any other ideas?
>>
>>Brett will be going to Ceduna in late June when the maser repairs are done
>>and so that is our best opportunity to find and fix the problem, however, it
>>is unlikely that we will have time between his visit and the July LBA
>>session to observe and correlate a VLBI test. This is how we propose to
>>look for the problem:
>>
>>1. Take a 1 GHz signal from an oscillator in the control room, split it,
>>send one part up the 10 MHz cable from the maser to the agilent, then back
>>through another RF cable. The put that signal and the other from the
>>original splitter into a vector voltmeter, put the output on a chart and
>>then run a fake LBA schedule for an hour or two and see what happens.
>>
>>2. Take a 1 GHz signal from the Agilent and a 1 GHz signal generated by an
>>oscillator in the control room and compare those on a vector voltmeter over
>>time.
>>
>>Any other comments or suggestions welcome. We are keen to fix the problem,
>>but it seems to be quite difficult to isolate and test as it is only
>>apparent in the processed data and isn't effecting fringes or anything
>>simple like that.
>>
>>Regards
>>
>>Simon
>>--
>>Simon Ellingsen : Senior Lecturer Physics & Astronomy, University of
>>Tasmania
>>email : Simon.Ellingsen_at_utas.<!--nospam-->edu.au
>>WWW :
>>http://fcms.its.utas.edu.au/scieng/physics/pagedetails.asp?lpersonId=681
>>Phone : 6226 7588 ; Area Code : +61 3 (International)
>>6278 8636 (Home), 6226 2410 (Fax) 03 (Australia)
>>
>>
>>
>>
>>
-- Brett Reid Observatory Manager, Radio Astronomy Group University of Tasmania, School of Mathematics and Physics Phone +61 3 6248 5285 Mobile 0407 955 283 email brett.reid_at_utas.<!--nospam-->edu.au web http://www-ra.phys.utas.edu.au/~breidReceived on 2009-08-16 21:06:40