next up previous contents
Next: RF performance Up: Installation of the #1 Previous: Installation of the #1

Main technical features

The new receiver uses basically the same general layout as the current Plateau de Bure receivers. It is built inside a hybrid Infrared Labs cryostat. Such a cryostat is already used at the 30-m for the 3mm2 receiver, which is single-channel. The local oscillators are fed by waveguides to couplers for each channel.

The nominal operating frequency ranges of the two channels, defined essentially by the LO subsytems, are : 83.5-115.5, and 200-255 GHz, respectively. The mixers operate in SSB mode, with a rejection of the USB of, respectively, 12dB and 20dB (typical).

The IF bandwidth is 0.5 GHz for receiver A100 and 1GHz for receiver A230 (recommended nomenclature for the future).

The cold optics comprises two horn-lens assemblies back-to-back, and the two beams are coming out of the cryostat on opposite sides, refocussed by elliptical mirrors, and recombined by a polarization grid (G1). What is new, compared with the Bure receivers, is that grid G1 is followed by a Martin-Puplett diplexer, which allows a single linear polarization coming from the sky to be transformed into two orthogonal polarizations at the respective operating frequencies of the two channels.

The general principle of frequency-dependent polarization transformation is similar to the one currently implemented at the 30-m telescope to allow 4-frequency observing, but, compared with the dichroic grids, the MP diplexer allows to reach relatively large (up to 50mm) optical path differences, and generally to find a setting with low losses for both frequencies; the dichroic grids allowed only small path differences, and one had often to accept a sacrifice on the 3mm optical losses to preserve the 1.3mm receiver temperature.

Do note, however, that for certain frequency ratios, no good compromise can be found, no matter how large the allowed path difference. The worst such ratio is tex2html_wrap_inline1260 , for which the theoretical diplexing loss is 5%.

To allow the best optimization of the diplexer, the astronomer should indicate to the operator whether he/she intends to observe in narrow-band mode (up to approx 100 MHz) or in broad-band mode. Please note that this narrow/wide choice is completely disconnected from the narrow/wide choice attached to the use of 230G2 (and now A230), depending on whether one is using or not the 1-GHz wide filterbank.


next up previous contents
Next: RF performance Up: Installation of the #1 Previous: Installation of the #1

guelin@iram.fr