The ALMA Cycle 2 Technical Handbook discusses the bands and their receivers (Single Sideband; SSB, Dual Sideband; 2SB, and Double Sideband; DSB). This knowledgebase article assumes this has already been read, and is intended to answer additional questions one might have about the receivers associated with ALMA bands.
1. Why will Bands 1 and 2 be SSB?
Answer: The issue of why different types of receivers are built for different bands is often a question of engineering rather than solely a science-based decision. The frequency range of bands 1 and 2 is such that the range of each band can be fully covered efficiently by an SSB receiver, thus removing the additional complexities that come with a multi-SB receiver.
2. The Handbook says that for the 2SB receivers "both the upper and lower sidebands are provided separately and simultaneously". Later in the paragraph, however, it talks about an "unwanted sideband rejection". Are both sidebands kept or is one rejected?
Answer: For the 2SB receivers in bands 3 through 8, the receiver provides separate outputs for each sideband in which the contribution from the opposite sideband has already been reduced by a factor of 10-100. Further rejection by a factor of >100 is achieved by a clever application and removal of frequency offsets in the LO system. By default, for each spectral window, the signal from one sideband will be kept while the signal from the other sideband will be rejected. However, spectral windows can be independently tuned in the two sidebands: a spectral window in the lower sideband can be tuned to pick up a frequency that was rejected from a different spectral window in the upper sideband. This is different from an SSB system, where the signal from only one sideband is retained by the receiver hardware, and from a DSB system, where the lower and upper sidebands are super-imposed approximately equally. (In one sense, a 2SB system is similar to two SSBs tuned separately by a fixed amount.)
3. When discussing the band 9 receivers, the Handbook states: "In the future, suitable phase switching will be introduced in the correlator, and both sidebands can be correlated and processed independently, thus doubling the effective system bandwidth." How far away is that future? Will both sidebands be available for Cycle 2 observations? When this future arrives, how will this system be different, from an observer's point of view, from a 2SB system? Will both sidebands will be available separately for both systems?
Answer: 90-degree phase-switching will NOT be offered in Cycle 2. The full Cycle 2 capabilities are available in the Cycle 2 Technical Handbook. However, both sidebands will be available for Cycle 2 in a similar manner as described above for the 2SB receivers. Like those bands, band 9 will have independently tunable spectral windows. What is different between DSB and 2SB observing with ALMA is that: (1) the noise from the "unwanted" sideband is not suppressed at all by a DSB receiver, and (2) for single-dish observing the signal from the opposite sideband is not suppressed either. The latter will always be true for single-dish mode. However, when 90-degree phase-switching becomes available, it will be possible to separate the signals from the two sidebands of a DSB receiver. In other words, the signal from both sidebands associated with a baseband can be accessed at the same time as long as two spectral windows are employed, which is where the effective bandwidth doubling arises. The system with 90-degree phase-switching will thus provide the option for twice as many spectral windows, but the spectral windows will no longer be completely independently tunable. (e.g., 4 independent tunings of 8 spectral windows in DSB, instead of 4 independent tunings of 4 spectral windows in 2SB.)
1. Why will Bands 1 and 2 be SSB?
Answer: The issue of why different types of receivers are built for different bands is often a question of engineering rather than solely a science-based decision. The frequency range of bands 1 and 2 is such that the range of each band can be fully covered efficiently by an SSB receiver, thus removing the additional complexities that come with a multi-SB receiver.
2. The Handbook says that for the 2SB receivers "both the upper and lower sidebands are provided separately and simultaneously". Later in the paragraph, however, it talks about an "unwanted sideband rejection". Are both sidebands kept or is one rejected?
Answer: For the 2SB receivers in bands 3 through 8, the receiver provides separate outputs for each sideband in which the contribution from the opposite sideband has already been reduced by a factor of 10-100. Further rejection by a factor of >100 is achieved by a clever application and removal of frequency offsets in the LO system. By default, for each spectral window, the signal from one sideband will be kept while the signal from the other sideband will be rejected. However, spectral windows can be independently tuned in the two sidebands: a spectral window in the lower sideband can be tuned to pick up a frequency that was rejected from a different spectral window in the upper sideband. This is different from an SSB system, where the signal from only one sideband is retained by the receiver hardware, and from a DSB system, where the lower and upper sidebands are super-imposed approximately equally. (In one sense, a 2SB system is similar to two SSBs tuned separately by a fixed amount.)
3. When discussing the band 9 receivers, the Handbook states: "In the future, suitable phase switching will be introduced in the correlator, and both sidebands can be correlated and processed independently, thus doubling the effective system bandwidth." How far away is that future? Will both sidebands be available for Cycle 2 observations? When this future arrives, how will this system be different, from an observer's point of view, from a 2SB system? Will both sidebands will be available separately for both systems?
Answer: 90-degree phase-switching will NOT be offered in Cycle 2. The full Cycle 2 capabilities are available in the Cycle 2 Technical Handbook. However, both sidebands will be available for Cycle 2 in a similar manner as described above for the 2SB receivers. Like those bands, band 9 will have independently tunable spectral windows. What is different between DSB and 2SB observing with ALMA is that: (1) the noise from the "unwanted" sideband is not suppressed at all by a DSB receiver, and (2) for single-dish observing the signal from the opposite sideband is not suppressed either. The latter will always be true for single-dish mode. However, when 90-degree phase-switching becomes available, it will be possible to separate the signals from the two sidebands of a DSB receiver. In other words, the signal from both sidebands associated with a baseband can be accessed at the same time as long as two spectral windows are employed, which is where the effective bandwidth doubling arises. The system with 90-degree phase-switching will thus provide the option for twice as many spectral windows, but the spectral windows will no longer be completely independently tunable. (e.g., 4 independent tunings of 8 spectral windows in DSB, instead of 4 independent tunings of 4 spectral windows in 2SB.)