Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.mrao.cam.ac.uk/projects/harp/public/HARP-B_Project_Overview_v1.0_20NOV00.pdf Äàòà èçìåíåíèÿ: Mon Jul 2 20:24:44 2001 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 17:44:19 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: âîçäóøíûå ìàññû

HARP-B
Project Overview

Document Id: Version: Date: Authors: Filename: Location:

0018 (formerly IDR01) 1.0 20NOV00 H. Smith, R. Hills HARP-B_Project_Overview_v1.0_20NOV00.doc mraos:/home/segh/HARPDOCS/


Revision History

Ver.
1.0 Original

Notes

Date
20NOV00

Initials
HS/REH


HARP-B
Acknowledgement
Although this document has been put together by Richard Hills and Harry Smith, several sections of the text have been edited from documents supplied by the following members of the HARP-B Project team: HIA: Russell Redman Keith Yeung MRAO: Bob Barker Roger Dace Stephen Riches Stafford Withington Ghassan Yassin UK-ATC: Bill Dent John Harris Peter Hastings Ian Pain Bob Wall

We acknowledge their contributions.

3


HARP-B
Project Overview
1 ­ Introduction
This document gives a brief overview of the management, organisation, and technical aspects of the HARP-B Project.

1.1 ­ Project history
In late 1998 a proposal was submitted to the JCMT board for the funding of a programme to provide the JCMT with two heterodyne focal plane imaging arrays working at `B' and `D' bands ­ the `Heterodyne Array Receivers Programme' (HARP). It was envisaged that each array would occupy its own Dewar (this unit being referred to as a `camera'), and sit on the Nasmyth platform of the JCMT. The cameras would share a common set of fore optics bringing the beam out of the telescope via an image de-rotator, through the RHS bearing and encoder unit. In the original scheme, both cameras were to be developed and built simultaneously under the global project management of the UK Astronomy Technology Centre with MRAO providing the `B' camera and HIA providing the `D' camera. This proposal was subsequently modified when the JCMT Advisory Panel recommended that the B-Band camera proceed immediately, but that the D-Band camera enter a research and prototyping phase. The fore-optics with the `B' camera was termed `HARPB'. A major organisational change also occurred at this point when the project management became the responsibility of MRAO with ATC and HIA now as providers of major components and software. The HARP-B team then produced a conceptual design for the instrument, which was subject to external review at the end of March 1999. The HARP-B project then went through a more minor organisational revision where the project changed little technically, but some workpackages were re-distributed around participating organisations and more `bought-in' components were selected to save UK development effort. Funding was finally secured at the November 1999 JCMT Board meeting In early 2000, a further minor re-shuffle in the distribution of the workpackages occurred due to scheduling constraints at UK-ATC. This re-shuffle was approved by the Director of the JCMT in March 2000. The Project presently proceeds as agreed at that point. The Project management, organisation and technical aspects are described in the rest of this document.

1.2 ­ Global Specification
· A 16-element array where the combination of the receiver noise temperature and beam efficiency, (Trx/eta-beam), weighted optimally across the array is 330 K SSB or better for the central 20 GHz of the tuning range, including the 345.8 GHz CO 3-2 line. 4


HARP-B
· · · · A frequency tuning range of 325 ­ 375 GHz. A calibration system and optics that have load temperatures and efficiencies stable and uniform enough to make 5% inter-pixel calibration possible. (Note that the ACSIS reduction software computes the calibration, using data provided by HARP-B). Beam positions known to accuracy better than 1/10 of the beam width. A tuning system that can tune and make the array ready for operation in 5 minutes.

1.3 ­ Science
HARP-B offers: · · · · · Greatly improved productivity Complementarity with SCUBA. Complementarity with Aperture Synthesis projects. A world-class facility. Opportunity for a wide-ranging scientific programme · Large-scale unbiased surveys of star formation and outflows in molecular clouds · Large-scale studies of hierarchical structure and clumping in molecular clouds · The dynamics and excitation of the molecular gas in the Galactic Centre · The distribution and properties of molecular clouds in face-on and edge-on galactic disks, starburst and Seyfert galaxies, and interacting galaxy systems · Studies of the chemical processing of molecular clouds by shocks and turbulence · The chemical and physical state of comets near perihelion

2 ­ Overview of Management & Organisation
The underpinning management structure adopted is that of a collaborative `HARP-B' consortium of MRAO, UK-ATC, HIA and JAC. Under this arrangement, MRAO will perform the project management, take the scientific lead, will specify the technical details and interfaces for the entire project, and will be responsible for the final delivery and commissioning. UK-ATC and HIA will have responsibility for several key workpackages within the overall technical requirement.

2.1 - Project Leadership
MRAO provides the overall management and scientific leadership. The HARP-B Project Scientist provides the scientific direction with the HARP-B Project Manager having responsibility for the construction, delivery, and financial management.

· HARP-B Project Scientist: · HARP-B Project Manager:

Richard Hills. Harry Smith.

Each site also has a workpackage scientist who works with the overall HARP-B Project Scientist and/or local/overall HARP-B Project Manager(s) as to scientific performance issues. The local Project Scientists are 5


HARP-B
· · UKATC: HIA: Bill Dent Russell Redman UK-ATC and HIA each have a local project manager having responsibility for all of the workpackages allocated to that site. The local Project Managers have the responsibility of managing the work at their site. The local workpackage managers are · · UKATC: HIA: Ian Pain Lorne Avery

Richard Hills and Harry Smith also fulfil the `local' Project Scientist and Project Manager roles at MRAO.

2.2 - Project Work Breakdown Structure
The HARP-B project is a significant undertaking. For management purposes it has been broken down as far as possible into individual workpackages with clean interfaces that can proceed independently. The work breakdown structure for the project including the integration, testing and I&C phases, together with the allocated responsibilities is detailed in the table below. These areas of technical responsibility for construction are shown diagrammatically in the following figures.

HARP- B - Areas of Technical Responsibility

Mounting Structure (MRAO)

SCUBA

TMU

`B' CAMERA

JCMT Cabin

Calibrator (MRAO) K -Mirror (ATC)

6


HARP-B
HARP-B Camera - Areas of Technical Responsibility
ACSIS CRYOSTAT (ATC)
I. F. (MRAO)

Windows & IR filter (MRAO)

COLD OPTICS & LO Inj. (MRAO)

MIXERS (MRAO) [SIS Junctions (DIMES)]

CONTROL ELECTRONICS (MRAO)

High Level Control S/W (HIA)
CANbus

L.O. (HIA)

K-Mirror (M1-M3) Cryogenics & Support Equipment K-Mirror I & C Nasmyth mount Imaging Array Development Imaging Array production & testing IF Amplifiers Cold Optics Calibrator Warm Optics (M4 & M5) Control Electronics High level Control Software Computing Hardware LO Camera Assembly & test UK System Integration & testing I & C on JCMT

UK-ATC MRAO

HIA MRAO (with HIA and UKATC participation)

2.3 - Timescales
A Gantt chart showing anticipated timescales produced in March 2000 is shown on the next page. Note that the completion date is quoted at 13NOV03. This date, together with the overall status of the Project is presently subject to formal revision in advance of the HARP-B Integrated Design Review in November 2000.

7


HARP-B

8


HARP-B
3 ­ Technical Overview
3.1 ­ K-Mirror (UK-ATC)
The unit consists of three mirrors mounted close to the right-hand stub-axle of the telescope. The mirrors are labelled M1, M2 and M3 in the order that they are encountered by the beam from the sky. The K-mirror performs two functions: a) it forms an image of the secondary mirror at a point near the elevation encoder; b) it acts as an image (and polarization) rotator. The minimum requirement is to provide for the needs of the HARPB receiver, but the K-mirror may be used for other future projects. These may have rather different requirements, especially with regard to the surface finish of the mirrors and the field of view. Movement of the K-mirror Rotation The K-mirror will rotate as a unit about the elevation axis (y-axis) of the telescope. Movements are defined in terms of the angle of rotation of the unit (phi_K) around this axis: note that the effective rotation of the image and polarization is twice the angle phi_K.
Specification a) Range of rotation b) Speed of rotation c) Acceleration in phi d) Accuracy of phi +/- 45 deg Goal *as large as convenient 5 deg/sec 5 deg/sec^2 +/- 0.1 deg

2.5 deg/sec
2.5 deg/sec^2 +/- 0.2 deg

Control The motion will be motor-driven under computer control using an RS-232, with appropriate provisions for information on status, etc. The normal mode of operation will be to drive to a specific position on command and then stay there. The K-mirror is presently visualised thus:

9


HARP-B
3.2 ­ Optics (MRAO)
There are four powered mirrors between the encoder and the array and these are optimised together to take advantage of aberration balancing and deliver the best possible images at the array consistent with other requirements. A two-mirror relay system (M4 & M5) forms a (somewhat distorted) image of the focal plane just in front of the cryostat window and a pupil (an image of the secondary mirror) on the cold stop, which is well inside the cryostat. The cold stop is reflective (i.e. a mirror surrounded by cold absorber) and slightly curved. This curvature and that of the mirrors in the Mach-Zehnder interferometer form an image of the sky with low aberrations at the detector array. The mirror arrangement is as follows:

Cold optics To obtain the best sensitivity it will be necessary for the instrument to operate in a singlesideband mode and for the image sideband to be terminated in a load which is as cold as possible. Because the optics required to achieve this sideband separation inevitably have some losses and because there are also losses involved in going into and out of a cold vacuum space, there is a substantial sensitivity advantage in placing the SSB filter inside the Dewar. The baseline solution for the SSB filter is to use a Polarising Mach-Zehnder Interferometer for the sideband separation. The Mach-Zehnder interferometer is widely used in laser systems. In addition to the SSB filtering other factors which were taken into account in arriving at this scheme were the blocking of IR radiation, the cold "stop" to reduce side-lobes, the injection of the Local Oscillator signal and the provision of low-loss vacuum windows. The large through-put of the instrument means that it is necessary to make these optics quite large, and they will therefore be the dominant factor in determining the size of the Dewar and the layout of the optics. The figure shows the layout that has been adopted as the baseline.

10


HARP-B
3.3 ­ Imaging Array Module (MRAO)
The imaging module comprises the feedhorn/mixer block, and LO injection scheme. This unit will be physically independent of the cold optics, so that repairs and upgrades can be carried out easily. We have designed a single unit that can be plugged into the Dewar with very little disruption to the main body of the system and yet maintain mechanical alignment with the optics. Mixers The basic imaging array will be based on the horn-reflector antenna. That is to say each mixer will comprise a waveguide horn with a reflector mirror at its aperture. This arrangement has already been used for a number of applications and is known to produce highly collimated beams with a high degree of efficiency. In the case of an imaging array, it is necessary to correct the phase error across the aperture of each horn independently, and this is usually achieved by inserting plastic a lens. Plastic lenses are notoriously difficult to design, and introduce standing waves on the LO and signal paths. The hornreflector antenna removes the need for plastic lenses. It is a little awkward to arrange horn-reflector antennas into square arrays. In the figure below we show how the 16-element array will built up from two layers of 8 mixers. The reflectors are designed to produce a focal plane slightly in front of the array.

One of the biggest problems when designing packed configuration so that the sampling on terms, one has to be certain that the apertures the same time ensuring that the mixer blocks

an imaging array is how to achieve a closedthe sky is as tight as possible. In general of the horns lie next to each other, while at can, in reality, be manufactured and

11


HARP-B
mounted. It is particularly difficult to make sufficient space available for the IF connectors. In the case of the horn-reflector array, the IF connectors will be located at the back of the mixers, and the mixers will be mounted on two parallel plates. This arrangement ensures good mechanical stability and thermal performance. LO Injection For HARP-B it was decided to concentrate on an array of beam splitters. The preferred arrangement is shown in the figure above. Here, an array of Mylar beam splitters is mounted in front of the mixers. The arrangement is compact and straight forward to manufacture.

3.4 ­ Cryostat (UK-ATC)
The cryogenic design for HARP-B is intimately connected with the overall instrument layout and will be carried out as part of that design process. The current optical design leads to a simple overall cryostat design. The optics are in one plane in order to allow good aberration performance. This allows a relatively simple cryostat to be constructed along the lines of a cold optical bench with optics on one side and the cryogenics on the other. The cryostat would have a removable side panel and thermal shields to allow easy access to the optics. This is illustrated in the following figure:

12


HARP-B

3.5 ­ I.F. System (MRAO)
The IF system per pixel will consist of four units: an isolator/cold IF amplifier connected to the 4K stage and two warm IF amplifiers separated by a bandpass filter and possibly a fixed, level setting attenuator mounted externally on the wall of the Dewar. A centre frequency of 5GHz has been chosen with a minimum bandwidth of 1.6GHz. The overall mid-band gain will be ~85dB and a noise temperature ~7K.

Dewar Wall Warm IF Amplifier

Bias

Isolator

Cold IF Amplifier

Mixer Block

Bias "T"

Load

Gain 26dB

2nd Stage Amp Gain 30dB
Band Pass Filter

4K
20K Heat Shield

Fixed Attenuator 3rd Stage Amp/Detector Gain 30dB

ACSIS

13


HARP-B
3.6 ­ Control Electronics Hardware (MRAO)
The following block diagram provides an overview of the system. The control electronics is built around a central CANbus backbone, which is driven by the HARP-B microcomputer. Bought in items such as motor controllers for the SSB filter and autotuning of the LO are controlled via standard RS232 lines. The following system elements are the hardware nodes that will be developed for the control and the monitoring of parameters within HARP the system. · · · · · Mixer Devices SIS Bias (Interface: CAN bus). Mixer Devices Coil Driver (Interface: CAN bus). Second Stage Warm IF Control / IF Measurement (Interface: CAN bus). Cold IF / First Stage Warm IF Power Supply Monitor (Interface: CAN bus) Calibration Unit (Interface CAN bus).

The above list describes five control nodes, currently the first three have been implemented with the Cold IF / First Stage Warm IF Power Supply Monitor currently under development (OCT00).

14


HARP-B
HARP-B System Block Diagram Detailing Control System Interfaces
Cold Section Cold Interferometer Optics K Mirror Calibration Unit x16 Devices x16 Berkshire IF x16 Atlantic Microwave IF x16 Warm IF

x3 Motor Drive Pressure Temperature

Mixer Device Coil Drive

Mixer Device SIS Bias

Cold / 1st stage warm IF PSU Monitor

Warm IF Control

Local Oscillator

1.1.1.1.1.1.1

CAN Bus HARP-B Micro & instrumentation software (HIA)

RS232 Interface


HARP-B
3.7 ­ High-level Control software
The control system for the HARP-B receiver will be implemented in several pieces running on different CPU's. The highest levels of the HARP-B control system overlap with the rest of the observatory control system (OCS), and in particular with ACSIS, and will run in the central control computer. The mid-level software to tune the receiver will run in a dedicated HARP microcomputer. The HARP microcomputer will not interact directly with the hardware but will instead communicate with a set of dedicated microcontrollers, some commercial motor controllers and specialized sensors, and some custom boards accessible over a CAN bus. The latter can be put into a sleep mode during observing to keep the electrical environment near the detectors quiet. The top layer of the control system consists of a set of TODD scripts that define the standard observing modes. These scripts are being developed at UK-ATC to accommodate the capabilities of HARP-B, but will be general enough to be used with any front end. Most of the HARP-B-specific software consists of a set of EPICS "databases" executing in the HARP-B microcomputer. It has been divided into a "controller" database and a "simulator" database. The controller will provide the functionality normally associated with a control program. The simulator will be used in place of hardware during the HARP-B development phases, and will remain useful for testing and maintenance at the JCMT after commissioning. Interactions between the controller and the rest of the OCS will be defined by the interfaces with the TODD and the RTS (Real Time Sequencer). The TODD interface is a set of DRAMA commands and parameters that that must be implemented for all new JCMT frontends. The central commands in this set are CONFIG, which sets a rest frequency, sideband, etc. in preparation for tuning, and TUNE, which causes the controller to retune the receiver to the configuration specified by the last CONFIG command. During observing, the SEQUENCE command will allow fast-frequency switching to be synchronized with the backend (ACSIS) through the RTS, and will lock out all other attempts to retune the receiver. Detailed specifications for a generic HARP control software have been written. These will evolve into detailed specifications for the HARP-B receiver itself after the HARP IDR, when final decisions will have been made on the capabilities of the HARP-B hardware. Descriptions of the controller and simulator are available through the HARP web site. The EPICS software is being developed primarily by the HIA, with input from the JAC.

3.8 ­ Nasmyth Mount (MRAO)
A preliminary design for supporting the HARP-B cryostat from the upper part of the JCMT `A' frame is shown below. This will also accommodate the warm optics and the calibration unit. It has been sized to provide the high stiffness necessary to minimise any loss of alignment that could result from dynamic or other loading that it might be subjected to in normal service. Note that by providing one support for all of the optical elements, calibtrator and cryostat we have the benefit of doing trial integration in advance of delivery and the possibility of realistic system level testing.

16


HARP-B

3.9 ­ Calibrator (MRAO)
The basic design of the Nasmyth warm optics and calibration system is shown below. To perform calibration or optimisation measurements, the lower relay mirror (M5) is mechanically tilted so that the receiver "looks" into the Cal Unit. A mirror (C1) collects the beams and focuses them onto the calibration "loads" via the flat switching mirror C2. This arrangement is designed in such a way that the loads are located at an image of the cold stop. This means that all the detectors will see the same effective temperature even if there are slight variations in the actual temperature across the load. There are two loads: "cool" and "warm". The cool load will operate at a temperature which is close to that of the absorbing material in the atmosphere (typically ~10K below the ambient temperature at the surface) while the warm load will be about 40K above ambient. For normal observing C2 is set pointing at the cool load and M5 is flipped rapidly to the Cal position where a total power and/or spectrum measurement is made, and then back to the Sky. This "two-load" calibration should take only a second or two and should be sufficient to give the system temperature to an accuracy of a few percent. To obtain a value for the receiver temperature, and also for tasks which require separate knowledge of sky brightness (e.g. sky dips) C2 is also switched to the warm load and a second measurement is made. Such a "three-load" calibration will take a few seconds.

17


HARP-B

It is also planned to inject a low-power narrow-band coherent signal into the receiver from the calibration unit by switching C2 to a third position. This signal will be used for tuning and optimisation procedures and can also serve as a more general test signal. It will be produced by a YIG oscillator and a broad-band multiplier. The frequency will be adjustable (with moderate resolution and high accuracy) to anywhere in the band of the receiver in a few seconds. The optical arrangement is such that the signal should be distributed to all the mixers, although a high degree of uniformity cannot be expected. The amplitude will be controllable but not accurately known. It would normally be set up to produce of order 10% of the total power signal across the full IF band. The important points are: 1) the signal is narrow band and can therefore be used to set up and calibrate the Mach-Zehnder SSB filter; and 2) that it will be possible to modulate it rapidly, by purely electronic means, so that a continuous read-out of the signal-to-noise ratio can be obtained while adjustments are being made. The calibrator is a self-contained unit, mounted directly onto the telescope framework. All of its functions are controlled or monitored electronically via the CANbus, although provision will be also be made for direct (TTL) modulation of the coherent reference signal.

3.10 ­ Microcomputer (HIA)
The baseline design of the HARP-B control computer is a PowerPC-based embedded system running the VxWorks real-time operation system. It receives high-level software commands from the JCMT Observatory Control System (OCS) and other telescope subsystems through the DRAMA messaging facility over a standard ethernet connection. It communicates with the low level control electronics via the Controller Area Network (CAN) bus. The microcomputer is a self-contained enclosure unit consisting of a 5-slot 6U VME-crate, internal fan with filtered intake, with CPU board, a parallel I/O Board, ethernet card and a CAN bus interface board.

3.11 ­ Local Oscillator (HIA)
The figure on the following page shows a block diagram of the proposed Local Oscillator (LO) System of the HARP-B array receiver. 18


HARP-B
RS-232 Motor Controller RS232 from HARP-B Microcomputer Zimmermann Gunn Osc.
Backshort Tuning

Zimmermann Multiplier

Input Backshort

Output Backshort

f Bias Tuning +/-100MHz 450MHz PLL Ref. Ref. Monitor Mini-Circuit Splitter Mini-Circuit Splitter PTS-500 Frequency Synthesizer f1 30 wire TTL `bus' from HARP-B Micro

Gunn

-20dB Crossguide Coupler Modulator Control Voltage (option) PLL IF Mon

Feedhorn ISOLATOR or Hughes Ferrite Modulator RF o/p

Modulator
Multiplier Bias Voltage

MRAO PLL System

f Ref. Mon

Gunn

CAN-Bus LO Controller

CAN Bus From HARP-B Microcomputer

Micro Lambda MLSL YIG Osc. f2 10 MHz Observatory Ref.

Pacific MM MD2 Diplexer

Pacific MM Model FM Harmonic Mixer

MicroLambda Frequency Control Signals

19


HARP-B
3.12 ­ Assembly, Test and I&C
These workpackages are presently inactive. The current work plan leads to the assembly, integration and test of all of the HARP-B receiver components over a six month period at MRAO before shipment to Hilo. Definition of the specifics of the test regime are yet to be made, however, the basic tests will be using a coherent source on a 2-D scanner to test the imaging properties, followed by a relatively simple radiometric test scheme to measure coupling efficiencies.

20