NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Main > TWikiRegistration? > ToneyMinter > RadioAstron

Users | Groups | Offices | Changes | Index | Contents | Notify | Search | Jump to Topic:

RadioAstron Multi-Frequency Synthesis: GBT Compatibility

Introduction

The RadioAstron mission has proposed a Multi-Frequency Synthesis (MFS) scheme for observations in the frequency range of 18--26 GHz. MFS will help provide better images through image synthesis using Space VLBI. A one page information memo on MFS provided to us by Misha Popov (Astro Space Center in Moscow) can be found as a pdf file here.

Overview of Our Understanding of MFS

Basically the requirement is that the VLBI backend should record two different frequency channels at the same time. The first channel is to record with a fixed center frequency of 22.232 GHz. The second frequency channel is to record with a different center frequency that changes periodically - about every 1000 seconds.

There are two separate schemes. In the first scheme the 2nd channel's center frequencies are stepped among the following frequencies: 18.392, 19.352, 20.312, 21.272, 22.232, 23.192, 24.152 and 25.112 GHz. In the second scheme the 2nd channel's center frequencies are stepped among the following frequencies: 22.136, 22.168, 22.200 and 22.232 GHz.

The GBT K-band Receiver

The GBT K-band receiver covers the frequency range from 18 to 26.5 GHz. This receiver is actually has four different feeds. The first two feeds (feeds 1 and 2) are offset in cross-elevation by 178.8 arcseconds and cover the frequency range of 18-22.5 GHz. The next two feeds (feeds 3 and 4) are also offset in cross-elevation by 178.8 arcseconds and cover the frequency range of 22-26.5 GHz. Only feeds 1 and 2 (or 3 and 4) can be used simultaneously. It is not possible to used feeds 1 and 3 (or 2 and 4; 1 and 3; 2 and 3) at the same time.

Furthermore, the system temperature for the upper part of K-band (feeds 3 and 4) is about 10 K higher than the lower part of K-band (feeds 1 and 2) on the water line.

The GBT IF System

The GBT is capable of producing eight different IF signals with different sky frequencies to the backends for observing. The conversion from RF to IF is done by a single LO (LO1) in the receiver. The IF signal is then broken up into two different copies and brought over optical fiber from the telescope to the GBT Equipment Room in the Jansky Lab. Each signal is then divided into four copies (for a total of eight copies) and then each signal is mixed with eight different LOs (LO2) and then sent to the backend of choice. However, only four signals are sent to the VLBA_DAR. Two of these signals will the the two polarizations for a single center frequency and the other two can be the two polarizations at a different center frequency.

Doppler tracking of any astronomical signal is done only with the LO1 signal.

Pointing and Focusing the GBT

The GBT will need to perform a peak and focus observation every 30 minutes from sunrise until roughly 3 hours after sunset. This is because thermal heating/cooling differences in the GBT backup structure affect the telescope due to the GBT's offset axis design. From three hours after sunset until sunrise the time interval between pointing observations can be as much as one hour. A typical pointing and focus check will take about 5-6 minutes.

So pointing observations will disrupt the proposed timing of the MFS observations.

The difference in the pointing between beams 1 and 3 have been measured to be about 4 arc-seconds on the sky. The Gregorian subreflector's focus position differs by about 0.6 mm between beams 1 and 3. These values are from a very good weather day. Although this shows that the GBT can go between observations using beam 1 (18-22.5 GHz) and beam 3 (22-26.5 GHz), it is still recommended that a pointing and focus check is made when switching between these beams.

Feed Resonances

The GBT K-band receiver has several feed resonances. Within the feed resonance there is a large amount of "cross-talk" between the two polarizations of the feed and the feeds response may not be very linear. The resonances are listed in the table below.

As can be seen, one feed resonance for feed 3 covers the water line at 22232 MHz.

Balancing the GBT IF System

The GBT K-band receiver is known to have a large gain slope across the full bandpasses of each feed - both for the 18-22.5 GHz feeds and the 22-26.5 GHz feeds. With two input IFs, one stationary on the water line and the other varying across a large part of the receivers band, the observations will have to be done without any IF filters between the receiver until the signal is split and changed into different IF center frequencies for the MFS scheme. As such the IF Rack cannot be used to change the balance of the IF system when the MFS scheme changes the frequencies. This means that the balancing has to be done using the DAR levels in the VLB backend and changing the Converter Rack attenuation levels. This is currently not automated and must be done by hand. This will require at least one minute of overhead time and maybe more and does not garuantee that the balancing will happen at all.

GBT Compatibility with MFS Scheme 1

GBT Incompatibility with MFS Scheme 1

The GBT is not inherently compatible with MFS scheme 1 because MFS scheme 1 requires frequencies below 22.5 GHz and abover 22.5 GHz for the second channel's center frequency while the first channel's center frequency is held constant at 22.232 GHz. To not change any of the IF path - which seems to be a requirement of the MFS specifications - this would require that the GBT use feeds 1 and 3 simultaneously, which is not possible.

A Way To Possibly Impliment MFS Scheme 1 With The GBT

The fixed frequency of 22.232 GHz can be observed with both feed 1 and feed 3 of the GBT. The same LO synthesizer can be used for the LO1 signal but different mixers and feeds would be involved in the observations. There would have to be a short break in the observations while the GBT is reconfigured to use the different feed of the receiver - about 60 seconds(?). Some components of the IF system would be different when using this scheme but should have a minimal impact on observing. A question for a Russian collegues is, does this satisfy the MFS requirements? This scheme should work but should be tested with real VLBI observations to make sure it is compatible with the RadioAstron observations before the launch of the spacecraft. Such tests will also allow us to determine better estimates for the amount of overhead time in changing frequencies.

GBT Compatibility with MFS Scheme 2 (Water Maser/"fine" scheme)

The GBT may be compatible with MFS scheme 2. All frequencies for the second channel's center frequency would be observable with the same feed (either 1 or 2) at all times. To change frequencies, the GBT would have to stop observing, issue a new configuration which would change the appropriate LO2 reference frequency and then begin a new scan. This would take an estimated 60 seconds of overhead time (dependent on how quickly the IF system could be rebalance which currently would have to happend by hand). Again, test VLBI observations are recommended to help determine what overhead values will be and to confirm the compatibility of the GBT with the RadioAstron MFS scheme.

Summary

The GBT should be compatible with the MFS schemes but likely with larger overheads than allowed for in the RadioAstron MFS specifications. These overheads would include 5-6 minutes every half hour to one hour for pointing observations, 60 seconds per frequency change to rebalance the IF system by hand and a recommended pointing observation when changing between the two K-band receiver feeds. Test observations should be done to confirm the compatibility and to get better estimates of the overhead times.

-- ToneyMinter - 14 Apr 2006

Attachment:	Action:	Size:	Date:	Who:	Comment:
MFS.pdf	action	102384	14 Apr 2006 - 17:38	ToneyMinter