A New Major Optical Observatory in Canada

By Frank P. Roy, Elektra Observatories and Paul Wiegert, the University of Western Ontario

Thirty years have passed since a major optical observatory was built in Canada. Elektra Observatories, a Canadian not-for-profit corporation is developing a one-metre class, wide-field telescope to be situated in the Madawaska Highlands of Ontario. This area, 100 km north of Kingston as the crow flies, has the darkest skies available in southern Canada. The initiative is breaking new grounds on several fronts. It’s the first time in Canada that a major observatory will be designed, built, managed and operated by a private concern. It will be the largest field-of-view prime focus optical telescope in the world and use the largest monolithic image sensor in the world.

In the March 2003 no. 116 issue of Cassiopeia, Melvin Blake at al. published an article entitled “A Canadian Modest Aperture Telescope”. The article was based on a survey, conducted in December 2002, and clearly indicates that a 1-2 metre class imaging instrument with the widest possible field of view would be highly desirable to Canadian astronomers. According to the article this class of instrument could accomplish much leading edge science.

The One Metre Initiative (OMI) may have a modest aperture but it’s certainly not modest in performance. The telescope has sufficient capability to outperform much larger instruments and certainly all Canadian based imaging optical telescopes in terms of science throughput. State-of-the-art in every respect the Observatory is designed as an autonomous facility requiring no operator to function.

A Canadian based telescope has several advantages. First and foremost is cost, with logistics close behind. The total capital cost of the OMI is approximately $2 million USD. Locating the Observatory in the US south-west would more than double that cost. The logistics would also being much more complex which will raise the cost even more. In all, the same facility in the US south-west would run about $5+ million USD plus much higher operational cost.

Performance Objectives

The telescope, mount, pier, dome and site where all designed and selected for optimum image quality and maintaining the best possible local seeing, which we believe is a more important factor to telescope performance than the more distant atmospheric seeing. Our target is 1-1.2” mean FWHM at the prime focus. Extensive use of carbon fibre in the optical tube assembly, mount and dome will yield an extremely light and stable imaging platform. For maximum efficiency the telescope is configured exclusively as an imaging telescope using the ugriz filter set with the addition of an L filter (400-700 nm) for quick deep surveys (0.5 mag deeper than r’). The utilization of active optics in the form of a hexapod with 6 degrees of freedom in the prime focus assembly together with a corrected wide field of view of 5 degrees² along with a 112 mega-pixel monolithic CCD image sensor with download times under 2 seconds and an exceptionally dark sky with a measured brightness of 21.82 mag/arcsec² will yield a instrument with superb capabilities.

OPTICAL
Aperture (clear)	1.016 m (open back cellular), parabolic
Mode	Prime Focus
Focal Length (effective)	2454 mm
f/ratio	2.42
Field of View	5 degrees², 2.22º x 2.22º
Étendue	AW = 6
Spatial sampling	0.76 arcsec/pixel
Field Flattener/corrector	135 mm corrected image circle, 6% Vignetting
Field Flattener/corrector spot energy	0.4/0.7” (4.7/8.3 um) 80% encircle energy in the r’ (centre/edge)
Field Flattener/corrector design	4 lens design, spherical/aspherical, fused silica (3) + CaF2
Areal Central Obstruction	<20%
Mirror/Optical-surface Thickness	146 / 6 mm, Cellular Borofloat
Expected Image Quality	1.25” FWHM mean
Spectral Range	350-1000 nm
Filters	u', g', r', i', z' + L, < 10 s change time
Active Optics (Hexapod)	+/-50 mm in x, y, +/-25 mm in z, +/-1 um Repeatable (maximum range)
Active Optics motion	Lateral motion, Tip/Tilt, Rotate, Focus
Active Optics	All Sky Auto-Collimation and focus. Rapid lateral x,y tracking possible.
Mirror Accuracy	1/16th wave P-V 550nm
Mirror Roughness	~10 nm rms
Coatings	Enhanced Aluminum with SiO2/Ta2O5 over-coatings, >95% 450-650nm, 75% 300-1200nm
MECHANICAL
Mount Type	Equatorial One Tyne Fork
Drive Topology	Friction drive
R.A. and DEC Disc size	1 m
Periodic Error/Period	< 2” P-P in R.A. and DEC with a 1 hour period
Tracking error	< 0.1” (60s no guiding)
Slewing	3 degrees/second maximum
Pointing Accuracy	<10 arc sec rms, zenith to +30º elevation
OTA Material	Carbon Fibre Sandwich Core
Mirror Mass	~68 Kg
OTA Mass	~200 Kg
Mount Mass	~180 Kg
Dome Type	¾ Sphere, Calotte
Dome Ground Elevation	3 m
Dome Diameter and Opening	5m and 1.3m
SITE
Latitude	N 45º 01' 37.4"
Longitude	W 77º 05' 57.4"
Altitude	375 m
Sky Brightness	21.82 mag/arcsec² (v)
Horizon Exposure	<2º
Ground	80 acres cleared land on a hilltop, 12” top soil with granite base
Clear Dark Hours	~1100 hours, (Environment Canada for Kingston)
SCIENCE
Limiting Magnitude	25/25.5 (1800s, s/n = 3, r’/L, Zθ = 0º)
Photometry	1s 0.001 magnitude (1500s, 17th mag, r’, Zθ = 0º )
Surveys	500 deg² (8 hours, mag  = 24, s/n = 3, r’, Zθ = 0º)
Spectral Range	350-1000 nm with the ugriz + L filters
CAMERA
Sensor	DALSA STA-1600A
Array Size	10580 x 10560 pixels
Active Area	95.22 x 95.22 mm,  9067 mm²
Spectral Range	300-1000 nm
Illumination	Thinned back with enhanced UV coatings
Quantum Efficiency	95% in the r’
Pixel Pitch	9.0 x 9.0 um
Total Number of Pixels	111,724,800
Read Noise/Download time	<4e-/14s (full array), 12e-/2s (full array)
Well Size	80,000 e-
Image Size	223/670 MB (mono 16 bits/pixel, tricolor 48 bits/pixel)
Cooling	-100ºC +/- 0.1ºC Cry-Tiger
Dark noise	1e-/hr/pixel
Fill Factor	100%
Read Ports	16
SNR	SNR 78dB @1MHz, 72.3dB@10MHZ, 65dB @25MHz
Non-Linearity	1.0%
Sampling depth	16 bits

Table 1 Performance Specifications

Figure 1 OMI Limiting magnitude and photometric performance calculated for varying s/n for Zθ=0º

Figure 2 Short Exposure Photometric and survey performance

 

 

1. The Site

The site selection was based on several criteria; sky darkness, altitude and local site conditions being the most important.  The sky brightness map published by Cinzano and Thiene, Italy (1998) was used for locating a potential dark site. Our objective was to get a site as far south as possible, yet still have exceptionally dark skies and a large buffer zone. Additionally we wanted the highest possible elevation and a treeless area to minimize micro-climates caused by a forested canopy. The ONLY area that meets all of these criteria is near the small town of Denbigh, which is located about 100 Km north-west of Kingston, Ontario. Road and topographic maps were examined to locate a specific location. The chosen location is identified on topographic maps as Mallory Hill. It has an altitude of 400 meters with 80 acres of cleared land on a hill-top. The area has a good paved road running nearby and no higher elevations in the west for optimum air flow over the local terrain. Sky brightness measurements were made over several months using a Sky Quality Meter by Unihedron with reading was 21.82 mag/arcsec², no direct lights are visible from the site. There is sufficient buffer to allow 100 years of continuing dark skies in the face of expected development. According to Environment Canada we should expect more than 1000 night time cloud free hours per annum.

Figure 3 Panoramic night view at Mallory Hill taken on December 12/13 2007

Figure 4 Sky brightness map of north-east North America showing the location of Mallory Hill with major Canadian Observatories identified. Based on work by Cinzano and Thiene, Italy 1998

2. The Telescope

No effort was spared to design the best possible telescope. The telescope was designed to yield the largest possible FOV with a light rigid frame. The OTA uses a Serrurier truss topology with Carbon Fiber Sandwich Core for a total mass of ~200 Kg. This material is also very rigid and has superb dimensional stability over temperature; the end result is a very light and stable imaging platform. The optical system is unique; according to the optician it has the distinction of being the largest field-of-view prime focus telescope in the world, with a FOV of some 5 degrees². The optical system uses an f/2.34 open back cellular parabolic mirror, with a mass of ~68 Kg, a 4 lens field flattener/corrector is employed yielding a final f/2.4 system with 80% encircle energy of  0.4/0.7 arcsec centre/edge in the r’. The open back structure allows for a thin, ~6mm, optical surface, thereby minimizing the mass and permitting a rapid equilibrium to ambient. This is critical to eliminate the boundary effect, which can seriously degrade the seeing. The mirror has nine 90 mm low vibration fans to accelerate the ambient tracking. Mirror flaps are employed to minimize dust accumulation.

The 135mm diameter fully corrected field has 6% of vignetting. Enhanced aluminum is used with SiO₂/Ta₂O₅ over coatings with 95% reflectivity at 450-650nm and 75% from 300-1200 nm.

The u’, g’, r’, i’, z’ filter set will be used with <10s change time between the filters. In addition an L filter (400-700nm) will be available. A Bonn 125 mm square shutter will be used.

Figure 5 3D model of the One-Metre Initiative, provided by Dream Telescopes & Acc. Inc.

Figure 6 Spot diagrams based on 80% encircled diameter, boxes are 25 um

Figure 7 Spot curve with 80% encircle energy in arcsec, based on 9u pixel pitch and 0.76”/pixel

3. Active Optics

Integrated into the prime focus assembly is a hexapod with 6 degrees of freedom. This allows motion in lateral, rotation, z axis, and tip/tilt with +/-1 mm repeatability. The hexapod has the ability to respond quickly and may be used in the future as a form of active optics. The hexapod allows a collimation map to be applied as the telescope changes position on a continuous basis, thereby compensating for gravitational distortions and assuring sharp edge-to-edge images across the sky. In addition the hexapod will be used for precision focus.

4. The Camera and Guider

The Camera will be made by Spectral Instruments and employs the largest chip ever made, a 4” x 4” CCD image sensor with 112 million pixels. The sensor, made by Waterloo’s Dalsa, is produced at Canada’s only chip factory, in Bromont Québec. The thinned back illuminated 10,580 x 10,560 pixel sensor has pixel pitch of 9.0 x 9.0 microns and a peak quantum efficiency of 94% in the r’. The entire 112 Mpixel array can be downloaded in 2 seconds. This is accomplished by the use of 16 read ports. Cooled to -100ºC the camera has effectively no dark noise (1e-/pix/hr). 16 bit quantization is used with an 80,000e- well size. The use of a monolithic sensor will improve overall efficiency of the camera, by eliminating the gaps found in more traditional large format mosaic cameras and will simplify the calibration, image processing and correction.

Figure 6 STA-1600A 112 MPixel CCD

 

           

Figure 8 QE for a STA1600A with backside illumination

Four Dalsa FT-50, 1024 x 1024 pixels, 5.6 mm pitch frame transfer CCD’s will be used for guiding. These will be situated at the same corrected focal plane as the main image sensor. Tracking is expected to be better than 0.1 arcsecs for a s/n = 5 guide star (r’ mag. 14, 0.1 s exp.). Each sensor covers an area of 0.13º x 0.13º or about 0.07 deg² total for the 4 sensors. This will allow a sufficiently bright star to be found even at the galactic poles.

.

Figure 9 The focal plane of the OMI camera showing guiders, corrected field and shutter. The guider chips are 1.5 mm from the main sensor chip

5. The Mount

Due to the low mass of the OTA, a one tyne equatorial mount will be employed. Since the mount contains a great deal of mass every effort is being used to minimize its weight, thus thermal footprint and yet still produce a precise, accurate and stiff mount. Thus a combination of steel, aluminum and carbon fibre will be used were appropriate, yielding a total mass of less than 200 kg. The drive will be of the friction type, thus minimizing backlash and improving tracking and slewing. The maximum slew rate is 3º/sec with the tracking accuracy better than 0.1 arcsec over 60s maximum error (no guiding). The mount is software programmable for added flexibility, motor currents will be monitored and limit switches will be used for a fail safe operation.

6. The Dome and Pier

The dome plays a critical part of the telescope/mount seeing equation. A ¾ sphere Calotte configuration was chosen. The dome will be built of a carbon fibre sandwich core; this material is light, stiff and offers a very low temperature coefficient and addition to being tough. A Calotte type dome offers many benefits. First off it offers an improved air flow profile thus reducing air turbulence around the dome, second the circular aperture will reduce wind buffeting and restrict bright off-axis star light, thereby permitting smaller light baffles in the telescope which reduces the areal obstruction, thirdly its less mechanically complex thus improving reliability and fourth it is much more immune to snow accumulation. The dome will feature a force filtered side input air vent to create a positive air pressure within the dome. This will minimize dust accumulation on the optics and accelerate ambient temperature tracking. All heat generating equipment will either be located in the control building or isolated and vented to the exterior. The pier will be a concrete cylinder with a 48” OD and a 36” hollow core to minimize its thermal load. The dome will be elevated by some 3 metres from the ground to improve air flow, isolation from ground currents and rapid ambient temperature tracking. The ground has 12” of topsoil with a granite base. This combination of topologies, materials and designs will produce rapid ambient temperature tracking together with excellent air flow and minimal ground turbulence and will improve the dome seeing substantially.

7. Control system and computational resources

All hardware control is via a PLC with fault detection and shutdown built-in with redundant control computers used to minimize down time. Obeservations will be done in the queing mode. The dome aperture and rotation motors will have current monitors and the site will feature a weather station. A small super-computer will be available with 2-5 TFLOPS for on-site data analysis. The observatory is solar and wind powered.

8. Science

The One Metre Initiative will have sufficient performance to benefit several key areas of research. There are many other areas where the OMI could be very productive beyond the ones mentioned here.

• Search for Earth-sized planets: With its large 5 degrees² FOV and deep limiting magnitude the OMI is well suited for this task. Using the transit method a change of 0.001 magnitude can be detected for a 17 magnitude stars (r’) for a 1500 second exposure. This is the sensitivity level required to detect Earth-sized planets orbiting < M5 red-dwarf stars, which are 8-10% the diameter of the sun and form some 80% of the stellar population.
• NEO search: Capable of searching thousands of square degrees per night the OMI can greatly contribute to our knowledge of Near Earth Objects. The large FOV, quick download time (2s) and the ability to see very faint, magnitude 23 in under a minute, should enable the OMI to detect several hundred of NEO’s per year.
• Supernovae search: The OMI will have the ability to discover thousands of SNe per year depending on how much time is devoted towards this task.
• Galaxy clusters and Super-Clusters: Able to reach magnitude 25 (r’, s/n = 3) in less than 30 minutes with a 5 deg² FOV will allow study of the biggest galaxy clusters and super-clusters.
• Gravitational Sheer and SNe light echoes: With an AW of 6, the OMI is a good instrument to find and study gravitational sheers and SNe light echoes’.
• Rapid Variation Astronomy: The camera is capable of full-array downloads in < 2 seconds. This opens up the possibility to search for transient phenomenon over a large FOV. GRB’s, red dwarf flares, KBO’s, slowly varying pulsars, unusual objects are some of the possibilities.
• Surveys: The OMI is a powerful survey tool. For example a survey to magnitude 24 in 3 colors down to -25º DEC. could be accomplished in 6 years with 225 hours per year devoted towards the task.

9. OMI and CFHT comparison

 The OMI with its large AW compares favorably to the CFHT in terms of survey work. For magnitude 23.5 (r’) and brighter, the OMI can actually outperform the CFHT. Due to the CFHT’s very long download times, short exposures are very inefficient.

Figure 10 OMI and the CFHT exposure ratio with varying seeing. The CFHT is assumed to have 0.8" seeing.

Conclusion

The OMI was designed to take advantage of the latest technological and design innovations together with new high performance materials and an understanding of the seeing equation to design a modest cost (~$2 million) facility with performance unmatched by any one-metre class telescope in the world. Situating the Observatory in Southern Ontario has drawbacks in terms of climate, the low capital and operational cost together with a high yield imaging throughput however more than compensate. In addition, the proximity to several universities with astronomy departments in Ontario and Québec is a benefit. The establishment of the One Metre Initiative will become a valuable asset to Canadian astronomers. The Mallory Hill site has much potential for optical astronomy; for it is has darker skies than the major observatories in Arizona and California.

Frank Roy frank.roy@elektraobservatories.org
Paul Wiegert pwiegert@uwo.ca
http://elektraobservatories.org