NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Projects > KaBandReceiver > KaCommSumm_Baselines_05

Changes | Index | Contents | Search | Statistics | Jump to Topic:

KA Band Spectral Line Characterisation

Summary

Since the Ka band receiver observing runs TKA_01 - TKA_07 (see KaCommSumm05 for summary) the receiver has had improvements made in order to address the problem of gain compression that was being seen. Due to a path length difference in a waveguide the BEAMS HAVE SWAPPED. This should only be significant in reductions of data.

The optimum receiver LO level is now between 4.5 and 5.0 due to changes. We are also now missing a channel coming from the receiver, this is due to the lack of a fourth amplifier that should arrive in a month or so and can hopefully be inserted into the receiver without bringing the receiver down from the telescope.

Data

• 18th Dec 2005

We are seeing low power levels in OD 5 and have high duty cycle levels (low power) due to our missing channel. This channel is L2, though there is still data coming through this channel and, in fact, the antenna temperatures seem unaffected by the low power in this channel. Some thought is needed about how to compensate (if at all) for the low power when reducing/calibrating the data. For now I'm treating the channel as normal.

The LO is struggling at low frequencies (26 GHz), also at 29 GHz where we are seeing high power levels that the attenuators can't cope with. This may be due to the LO1 harmonic, though it is worth noting that the mixer in the mmconverter is at its limit here at 15 GHz and when we move up in frequency the mixer swaps, for whatever reason, the problem goes away at 30 GHz.

Blank Sky Measurements

From scans 89-90 there's a big spike at ~31.7 GHz, not repeated in scans 91-92 though there are other spikes. A few cropped up in the BW tune too. What are these highly transient spikes?

• 23rd Dec 2005

I'm getting a warning that I should 'Check LO power level' but they seem fine, LO1A = 4.75 and LO1B =4.87. As far as I am aware these are within optimum limits? (4.5 - 5.0 and 4.5 - 5.5).

Again I've had poor balancing at low frequencies (26 GHz), also see the high power levels at 29 GHz again.

OD 5 rf power is low again (~1.35) with corresponding high duty cycles.

WEIRDNESS! I did three bandwidth tunings, going from 26 to 40 GHz in chunks of 800 MHz, the three runs using the L, R and LR cals respectively. Due to a configuration mixup the L cal run was only using one beam, this run went fine. However, the R and LR cal runs (using both beams) both encountered a strange problem. At 29GHz the online filler wrote two records to the fits file, this didn't occur at any other frequencies or for the L cal run which only used one beam. Configurations at other frequencies are identical apart from the sky frequency. Have contacted Paul M. about this and will update.

• NH3 observations

Observed Orion hot core in 7 lines of NH3, poor balancing at low edge of band though had a good detection of 1 line at 29.9 GHz.

• Blank sky observations

Again, first time through we had poor balancing but this was cleared by second scan.

Analysis

The data were reduced using ( (sig_a - ref_a) - (sig_b - ref_b) ) / ( (sig_a_cal - ref_a_cal) - (sig_a - ref_a)) and calibrated using the observations of 3C147 made during the bandwidth tune. All blank sky observations were performed using two spectral windows centred at 31.4 and 32.1 GHz.

This waterfall plot is centred at 31.4 GHZ and time runs up the y axis, it is a combination of data taken on the 18th and 23rd of Dec. The first few scans on the 18th have been left off the plot as these were extremely poor. The poor balancing that occured at the beginning of the 23rd does not seem to have affected the scans. The y-scale is in Janskys with an arbitrary offset

It can be seen that the rms scan to scan is roughly stable, becoming worse as time goes on. This is made clearer by plotting the indivdual rms of each scan versus time.

The mean rms of these scans is 0.014 Jy, there being no significant improvement upon the mean rms found during this fall (0.012). However, we appear to be seeing less of the 'structure' that was plaguing our previous observations.

Compare these two scans, taken from this fall and the past run respectively

Last fall...

Recent run...

This is perhaps made clearer by comparing multiple scans....

Last fall...

Recent run...

O.k. so the wiggle seems to be absent or at a lower level than previous runs. BUT!!! If we look at the later runs from the 23rd that show the larger noise we see...

The baselines are looking as bad as ever!!!

We do see an improved drop-off in noise though this may well be due to a fortuitous time-ordering of the bad scans. The x-scale here is in integrations, this is a little ambiguous for the observations from TKA_06 but otherwise corresponds to 3 integrations = 1 minute wall time.

Last fall...

Recent run...

The R.M.S. levels reached after 1440/1776 (fall/recent) integrations are 0.7/1.0 mJy.

So, we are not seeing any improvement compared to Brian's analysis KaCommSumm05. It certainly looks possible to achieve an R.M.S. of 1 mJy in a few hours. However, this looks to be tailing off. Note that I haven't binned this data so a better R.M.S. could be achieved by smoothing at the expense of resolution.

Having spoken to Richard P. it has been decided that a review of the proposals will be made to look for any science proposals that do not require low R.M.S. levels, though this looks to be a maximum of 5 as of 29th Dec. Apart from this, some new investigations will be made to try and determine the cause of the ripple.

Attachment:	Action:	Size:	Date:	Who:	Comment:
old_blank_calibrated_cum_rms_IF0.png	action	10955	29 Dec 2005 - 16:53	LarryMorgan
old_blank_calibrated_cum_rms_IF1.png	action	10374	29 Dec 2005 - 16:53	LarryMorgan
old_blank_calibrated_ind_rms_IF0.png	action	22906	29 Dec 2005 - 16:53	LarryMorgan
old_blank_calibrated_ind_rms_IF1.png	action	25058	29 Dec 2005 - 16:53	LarryMorgan
old_blank_calibrated_waterfall_IF0.png	action	44403	29 Dec 2005 - 16:54	LarryMorgan
old_blank_calibrated_waterfall_IF1.png	action	42812	29 Dec 2005 - 16:54	LarryMorgan
new_blank_calibrated_cum_rms_IF0.png	action	10136	29 Dec 2005 - 16:54	LarryMorgan
new_blank_calibrated_cum_rms_IF1.png	action	8882	29 Dec 2005 - 16:54	LarryMorgan
new_blank_calibrated_ind_rms_IF0.png	action	11943	29 Dec 2005 - 16:54	LarryMorgan
new_blank_calibrated_ind_rms_IF1.png	action	9632	29 Dec 2005 - 16:55	LarryMorgan
new_blank_calibrated_waterfall_IF0.png	action	51496	29 Dec 2005 - 16:55	LarryMorgan
new_blank_calibrated_waterfall_IF1.png	action	53283	29 Dec 2005 - 16:55	LarryMorgan
new_blank_calibrated_scan144.png	action	40795	29 Dec 2005 - 18:22	LarryMorgan
old_blank_calibrated_scan133.png	action	40525	29 Dec 2005 - 18:23	LarryMorgan
old_blank_calibrated_waterfall_IF1_zoom.png	action	22296	29 Dec 2005 - 18:45	LarryMorgan
new_blank_calibrated_waterfall_IF1_zoom.png	action	22797	29 Dec 2005 - 18:45	LarryMorgan
new_blank_calibrated_waterfall_IF1_bad.png	action	25850	29 Dec 2005 - 19:15	LarryMorgan