Grundmann Report on Options for Use of the Existing Pier with a New Telescope

2. Terms of Reference 2

3. Assumptions made in the Course of Study 3

4. Review of Existing 3.6m Facility 4

5. Review of other Large Telescopes 6

6. Segmented Primary Mirror 7

7. Optical Configurations for a Large Telescope 9

8. Base Line Designs for 15m & 12m Telescopes 11

9. Larger Telescope on CFHT Pier & Foundation 13

10. Telescope Enclosure 15

11. Factors Determining Telescope Size Limit 16

12. Salvage Value of Existing Facility 18

13. Preliminary Cost of a Large Telescope 20

14. Tentative Time Schedule for Implementation of New

Telescope 21

15. Estimate of Shut-down Time to Rework Facility 22

16. Further Study Areas Identified 23

17. Other Telescope Designs & Concepts 25

18. Conclusions 28

19. References 29

20. List of Figures 31

1. Introduction
 
 

This report covers the work undertaken in order to address the four main questions raised by the SAC Working Group on the future of CFHT, 21 September 1996. These questions, as well as the background rationale to them, are reproduced in this report under the following section as Terms of Reference.

In order to be able to answer these four questions, several activities were undertaken. The existing 3.6m CFHT facility was reviewed in order to assess its features and its limitations for supporting a much larger telescope. Physical dimensions and the masses of major optical and mechanical components were recorded, and the pier and foundation's suitability to support a larger instrument were assessed.

KECK and GEMINI telescopes were reviewed in order to obtain comparable physical features as well as information on costs and schedules.

A review of recently published literature on new large telescopes revealed the report of a study of a 25m telescope with a segmented primary mirror. Some very useful and relevant background information was obtained from this paper.

Conceptual base line telescope designs of a 15m CFHT and a 12m CFHT were started in order to obtain some preliminary figures for physical dimensions and masses as well as some preliminary estimates on costs and construction schedules.

2. Terms of Reference

Report of the Working Group on the future of CFHT, 21 September 1996.

It is the opinion of this committee that recent advances in optical and infrared astronomy have been primarily the consequence of improved capabilities in angular resolution, light-gathering power and the ability to conduct observations in non-traditional wavelength regimes. While the CFHT has served its users well by setting the technical standard in these areas, we believe that the "long term" (i.e. starting 2005) facility needs of the three communities would be best served by replacing the existing 3.6m telescope with a segmented mirror instrument of 12-16m aperture on the same site (possibly using the same pier). In comparison with KECK, such a telescope would provide a 50% increase in resolving power in the 3-5m regime and would allow the deepest and most detailed imaging of high-z objects and star-forming regions. It would also provide a doubling of light-gathering area over the largest existing telescopes at all wavelengths. In a subsequent report, we will describe in more detail why we feel that a 6-8m class replacement or instrumentation/focal ratio upgrade would not be competitive or desirable on a 10-year time scale and how our proposed replacement adds value to the joint community assets (which at that time will be at least six 8m facilities) at a fraction of the cost of developing the same telescope from scratch. We request that short-term technical studies be initiated to explore the feasibility of installing a 12-16m segmented mirror telescope on the existing site. Specifically, these studies should:

3. Assumptions made in the Course of this Study

As is to be expected at this early stage in a telescope study, and given the very general terms of reference, some very basic specifications are missing; these should nevertheless be mentioned and recorded:

i. telescope image scale and field size

ii. image quality

iii. telescope focus stations

iv. optical design and focal ratios

v. telescope optical error budget

vi. overall facility error budget

vii. sky coverage

In order to proceed with a telescope base line design, several assumptions needed to be made from the start:

a. the primary mirror is segmented

b. the physical size of the primary mirror segments is

identical to KECK

c. the mirror segment support system is identical to KECK

d. the primary mirror surface is f/1.5 hyperbolic

e. the existing concrete pier is utilized but may be modified

f. the dome track remains in the same place but may be modified

g. the building below the dome track level stays, but the internal quarters may be modified

h. the observing floor level stays at EL 13,772'-0" (4,197.7m)

i. the only telescope focus station is the cassegrain below the primary mirror cell.

4. Review of Existing 3.6m CFHT Facility

For discussion purpose of some critical areas and for the comparison with other large telescopes, a simple cross-sectional view of the present CFHT facility is shown in Figure 1.

Although the dome is relatively large for a 3.6m telescope, it is not suited for anything larger than a 6m telescope. This is because the main arch girders are only 6.5m apart and all the rib girders attach to these. No modification is possible here and the dome needs to be removed and replaced with a larger enclosure that can accommodate a shutter opening in excess of 12m

The transfer of wind induced dome vibration through the foundation and soil into the telescope structure needs to be addressed if a much larger and non-spherical enclosure is substituted. The 3m separation between building foundation and pier foundation appears adequate for now and comparable to the separation allowed for the Mauna Kea GEMINI telescope for identical soil conditions, however, the subject requires further investigation during a detailed design study, once sufficient details on the enclosure are known.

Figure 2 illustrates some basic and unique characteristics of the pier top. This top end at 0.3m thickness is relatively thin and, although well reinforced, will not be suitable for any direct telescope loads. However, floor regions at the north and south ends increase in thickness to 0.9m; this was done to support the full telescope mass at well defined locations. The equatorially mounted telescope resides on a large steel A-frame which in turn sits on and is anchored to the pier at the north and south edges where the floor is 0.9m thick.

For the existing pier an azimuth track for a large telescope can only be located immediately above the circular structural wall. Such an arrangement, given in Figure 3, shows that the azimuth track diameter is limited to 16m. In order to mount this track, a reinforced concrete ring needs to be added to the existing pier and this additional concrete is required to reinforce the pier top edge as well as to distribute the telescope loads. The reinforcing for the new concrete ring needs to connect and interlock with the existing pier structure since the azimuth track will impose not only telescope gravity load, but also seismic horizontal loads. The existing telescope anchor rods will be of some use but additional anchor rods will have to be placed in the pier top.

The observing room level at EL 13,772'-0" is about 2.5m above the concrete pier top. This space should provide enough room for the additional concrete ring, the azimuth track, the hydrostatic bearing and a new telescope base frame, so that the current location of the observing room floor level can be maintained.

The dome track is 28.8m in diameter and this dimension will to a large degree limit the size of enclosure that can be placed on it, which in turn also limits the size of telescope that can be placed on the existing structures.

5. Review of Other Large Telescopes

Figures 4 and 5 give cross-sectional views of the 8m GEMINI telescope and of the 10m KECK telescope respectively, two recent designs that are of interest to this study. The most apparent difference is that the GEMINI telescope is much farther above ground than the KECK telescope. Because of this and because the sub-soil at Mauna Kea has a low soil modulus, GEMINI had to adopt a large diameter pier base for its telescope.

As the KECK primary mirror consists of relatively thin segments, the overall weight of the 10m telescope is considerably less than that of the 8m telescope. The other reason the 10m telescope is of much lower mass is that the mechanical/structural design has consistently throughout employed light-weight space frame designs, instead of steel plate box sections, for most major components including the primary mirror cell and the yoke frame. Table 1 further illustrates the differences between the two designs in that the 10m is lighter than the 8m for primary mirror, cell, tube and yoke components. Note that for the 8m the base frame weight needs to be added to the yoke weight.

The other important difference between the two designs is the diameter of the azimuth track: 17.2m for KECK as compared to 9m for GEMINI. This large track diameter facilitated a simpler and lighter yoke structure than would otherwise have been possible.

Probably the most interesting observation, however, is that the masses of the 3.6m CFH telescope for most major components are almost identical to those of the 10m telescope. The explanation for that is the use of the segmented primary mirror and the alt-az style of mounting, as compared to the equatorial mounting.

6. Segmented Primary Mirror

For the base line design it is assumed that the mirror segments are identical in size to the KECK segments. These were cut, after grinding and polishing, from a 2m diameter blank and are hexagonal in shape, 1.8m across corners, 75mm thick and weigh 400 kg each. It is not known at this stage whether this is still the most economical and functional segment size today, or will be in the near future, and later studies need to address this.

Figure 6 is useful in illustrating some of the physical characteristics of a 1.8m hexagonal segment primary. For example a 10m parabolic or hyperbolic mirror has six unique segments in every 60º sector for a total of 36 segments. To have one spare of each type KECK acquired six additional segments. A 10m diameter circle projected on the 36 segment assembly shows that this arrangement does not give a full 10m diameter mirror surface at some angular locations while at some others it actually provides for more. Nevertheless, the surface area of 36 segments roughly equates to the area of a 10m diameter surface.

By adding one outer ring of additional segments a 12m or a 12.5m diameter mirror is achieved. Assuming that the central hole as before results from the missing central segment, the primary mirror is made up of 10 unique segments for a total of 60 segments. If one allows for a spare of each type, a total of 70 segments would need to be made.

By adding a second outer ring of additional segments a 15m diameter mirror is achieved. At this size of primary it is assumed that the central hole may be larger and would consist of not only a missing central segment but also of the missing first ring of six segments. Figure 6 shows that such a mirror would require 13 unique segments, and assuming the same ratio of spares, a total of 91 segments would be required. In order to keep the total number of segments to an absolute minimum, 6 outer segments with less than 30% contribution were excluded from the count. The summed surface area of 78 segments still exceeds the area of a 15m diameter surface.

To support and adjust the position of the mirror segments, the KECK system has 108 actuators and approximately 168 sensors. As shown in Figure 6, the larger segmented mirror would require a larger number of these components roughly proportional to the number of mirror segments.

7. Optical Configurations for a Large Telescope

The KECK telescope optics are of a conventional Ritchey-Chretien design with an f/1.75 hyperbolic primary mirror and an f/15 secondary.¹ The main cassegrain focus occurs 2.5m behind the primary mirror surface but the telescope also has numerous Nasmyth foci. For infrared work there is also an f/25 secondary mirror that can be chopped at frequencies up to 50 Hz and it focuses light at a forward Cassegrain focus near the elevation axis.

For the base line design of a large CFH telescope, it was assumed that the optical design would be similar to KECK except that the Cassegrain focus below the primary would be the only one, and that the primary mirror would be f/1.5 in order to achieve a very compact telescope tube.

Some recent optical designs for very large telescopes employing spherical primary mirrors are of interest. ^{2, 3,4} These designs are very attractive for large segmented primary mirrors since all segments have the same optical surface and are therefore completely identical and interchangeable. Figure 7a shows the 4 mirror system proposed for the 25m NORDIC telescope of which M4 would be a low order adaptive optics system able to correct below 15Hz for such things as wind shake, wavefront tilt, low frequency misalignment, guiding and collimation errors, etc. This 4-mirror design is very compact and allows for the shortest telescope tube possible and also for a very compact enclosure. It should be noted, however, that for this design the field is only 1.5 arc minutes.

Chris Morbey at DAO has examined a 12m telescope optical design of the 4 mirror type for a 0.4 degree field size. For close to diffraction limited performance in the 0.4 to 0.8 micron colour bandpass, the primary needs to be parabolic with the other 3 mirrors being either hyperbolic or ellipsoidal. The design also includes refractive corrective optics and is shown in figures 27a and 27b.

In order to achieve a spherical primary mirror, the field was reduced to 10 arc minutes diameter. The number of elements has stayed the same but the f-ratio has changed from 2 to 2.31. As can be seen in Figures 28a and 28b the spot sizes for this design are very similar and close to diffraction limited over the whole field.

8. Baseline Design for 12m and 15m Telescopes

For the considerations discussed in Section 6, two telescope sizes were investigated. Figure 8 shows a plan view of a 12m telescope arrangement and the base frame for the telescope structure. As mentioned before, existing conditions dictate that the azimuth track diameter be fixed at 16m. In order to obtain a compact and stiff structure, the load paths through the yoke and to the azimuth hydrostatic-static bearing should be as direct as possible. Assuming a 14m mirror cell, a 15m spacing for the two yoke assemblies seems reasonable. The main struts or legs in each yoke would terminate at the base frame with a 10m spacing. This would insure that the main telescope loads are transferred to the base frame very close to the location of the hydrostatic azimuth bearings.

From the above it follows that the base frame is about 12m x 16m and about 1.5m high. It would be essentially of space frame construction with the major loads carried by a plate fabricated X frame but with lateral stiffening provided by much lighter structures. The legs of the X frame, about 10m in length, could be disassembled for shipment and later bolted together on site.

With the above dimensions it follows that the rotating floor is about 20m in diameter. The outer boundary of the stationary floor is about 28m in diameter. This leaves an annular stationary floor 4m wide, probably just enough to accommodate an interior elevator shaft. It also provides sufficient space for a hatch opening for transport of mirror segments and secondary mirrors.

Figure 9 shows the 12m telescope on the existing pier. As mentioned earlier, the existing dome does not provide for adequate spacing between the main arch girders, nor is it sufficient in spherical diameter size, and therefore needs to be replaced. Like KECK, all major telescope components would be of space frame construction (instead of plate fabricated large sections) including the tube center section, except for the region around the altitude bearing journals and parts of the base frame.

The preliminary sizing of a 15m telescope followed the same rationale and the results are shown in Figures 10 and 11. The base frame is 14m x 19.2m while the length of the X frame leg is about 12m. The rotating floor is 24m in diameter leaving only a 2m wide annular stationary floor. This dimension is probably insufficient for an elevator shaft or and access hatch for mirror segment transport. Nevertheless, an access hatch could probably be provided in the rotating floor.

Figure 11 shows the 15m telescope on the existing pier. The assembly appears imbalanced since the telescope is larger than the pier. Of primary concern must be the adequacy of the pier foundation (more on this in Section 9) as well as the structural adequacy of the pier itself.

9. Large Telescope on the CFHT Pier and Foundation

High stiffness of an overall mechanical system i.e. telescope tube, azimuth yoke mount, concrete pier and ground supporting the pier is important in order for a telescope control system to correct as quickly as possible for external and internal disturbances to telescope tracking. For a Mauna Kea site where the ground soil modulus is very low because it consists primarily of loose cinders, the low ground stiffness must be taken into account in the overall mechanical system.

Recent large telescope designs have generally aimed at keeping the first fundamental frequency of the system above 4 Hz.^5,6 JNLT (Japan National Large Telescope) has gone so far as to improve the ground conditions, i.e. soil modulus and ground stiffness, by recompacting volcanic cinders through mixing them with cement.⁷

For the Mauna Kea telescope, GEMINI engineers took an alternative approach and decided to make the pier foundation as large in diameter as possible in order to improve the dynamic behaviour of the overall mechanical system as well as to reduce telescope tilt caused by wind loading on the structure. Despite this precaution, the first mode of vibration which is the pier and foundation tilting in the soil (rocking mode) occurs at 2.6 Hz. High wind conditions can produce telescope tilts of up to 1 arc second.⁵

The existing CFHT pier and foundation were designed for a 3.6m telescope but are quite substantial when compared to the GEMINI and KECK facilities. However, a much larger telescope placed on the existing CFHT pier and foundation cannot be much heavier than the existing telescope, its center of gravity cannot be much higher above ground, nor can the telescope top end be much higher above ground.

Table 1 lists various masses and dimensions for some existing telescopes and for 12m and 15m segmented mirror telescopes. The values for the 12m and 15m telescopes are very rough estimates based partially on the 10m telescope and partially on the very preliminary designs shown in Figures 8-11.

Table 2 gives the result of a very simplistic comparison of the dynamic behaviour of these facilities. For this comparison the soil spring factor was assumed to be identical for all, despite the fact the GEMINI has a large diameter pier base. The value for GEMINI was taken from their study report and is comparable to the others because it is based on a much lower soil modulus than assumed for the others which are based on the earlier Dames & Moore studies done in the late 1960's, early 1970's and mid 1980’s.^8,9

These first order calculations indicate that the CFHT pier and foundation can probably accommodate a larger telescope since the rocking mode frequency and telescope tilt are comparable to the existing designs. Note that the GEMINI studies show higher tilt values because they assumed much higher wind loading in their analysis.

10. Telescope Enclosure

The telescope tube sweep requirements, about 16m for the 12m telescope and 19m for the 15m telescope (see Table 1), combined with the existing dome track of 28.8m diameter, make the use of a spherical enclosure impractical. (See also Figures 9 and 11). One possible way to improve the clearance problem is to locate the dome spring line several meters above the telescope altitude axis while at the same time increasing the dome diameter. (Note that for the present facility the dome spring line is about 0.5m above the telescope altitude/polar axis intersection.) Such an arrangement would also improve the total clearance at the top of the enclosure thereby providing room for an overhead crane. The downside of such an arrangement is that it leads to a very much larger enclosure volume and a large enclosure structural mass, both of which would be difficult to control thermally.

Figure 12 shows a model for an enclosure considered initially for the GEMINI telescopes. (This model and several others of various designs and shapes were constructed to study wind flushing actions by conducting modeling tests in a water tunnel.) Figures 13 and 14 show plan views of this type of enclosure on the 28.8m diameter dome track. Given this limitation, an enclosure for the 12m baseline telescope appears possible since the horizontal profile is almost symmetrical and the shutter side is almost tangent to the track. This is not the case for the enclosure for the 15m baseline telescope; here the profile is far from symmetrical and the shutter side is outboard from the track by at least 5m.

11. Factors Determining Telescope Size Limit

i. Dome track diameter.

While the dome is large enough to house a new altitude over azimuth style telescope in the 6m size range, it would be highly impractical to relocate the dome to another site. It is also highly unlikely that a buyer with a suitable site could be found for the existing telescope. Therefore, except for the mirror blank and cell, it will be assumed in this study that the telescope and dome are only of material salvage value.

In about 1992 the Canadian Consortium WESTAR, inheritor of the assets of the cancelled Canadian national 4m telescope project (QE II, begun circa 1967) were finally able to sell the primary mirror blank. This 4m blank, of Corning fused quartz segments and only rough ground, was sold to a large US defence research company. They required a mirror blank of this size on short notice and paid about US$4,500,000 for it. Using this information as a guide, it is conceivable that CFHT might obtain about US$5,000,000 for its 3.6m mirror and cell, provided a buyer can be found before, during or after the CFHT 3.6m telescope decommissioning stage.

Dave Halliday provided the following information and estimates, should Coast Steel Fabricators Ltd. dismantle the 3.6m facility.¹⁰ Actual procedures and schedules are discussed in Section 15. It is assumed that all material will be removed from the site and trucked to a local depot or warehouse for storage. The enclosure and telescope sections will be left as large as possible and no special protection for extended storage is included. All estimates are in 1997 US$ with neither taxes, duties nor permits. The cost of dismantling and removing all telescope structural and mechanical components would be $523,000 and the material salvage value would be $100,000. The cost of dismantling and removing all of the dome would be $956,000 while the material salvage value would be $200,000.

Besides the above there are other dismantling activities not covered by the CSF estimates. The removal of telescope and dome wiring and controls would be by others, as would the dismantling and removal of the concrete floor at the observing room level EL 13,772'-0". In summary, the cost and benefits for dismantling the facility are:
 

	Cost	Credit
Primary mirror and cell	-	5,000,000
Telescope	523,000	100,000
Enclosure	956,000	200,000
Concrete floor	100,000	-
Wiring and controls	(CFHT staff)	-
Total	1,579,000	5,300,000

Balance on hand US$3,721,000
 
 

13. Preliminary Cost of a Large Telescope

To aid in estimating the cost of a 12m telescope facility, construction costs for GEMINI and KECK telescopes were obtained as well as for the 3.6m CFHT telescope. Estimates for the 25m NORDIC telescope are given in their study report and are included here for comparison.² Table 3 summarizes all of these and also shows the period of construction. Note that for the existing telescopes these costs are historical costs and have not been modified for inflation. Several categories, such as primary mirror and telescope mounting, have changed radically in design and construction since then and it makes little sense to readjust the costs to a common fiscal year base.

Cost information received from the various sources was also difficult to categorize as accounting methods for the various projects were quite different. For instance, some did not separately identify costs for things such as commissioning, coating plants, secondary mirrors, mirror polishing, etc. but instead included these costs in other categories and not necessarily in the same ones from project to project. To obtain more information requires further breakdown of categories and considerably more co-operation from the various institutions. It was felt that any greater detail was beyond the scope of the present study. The costs for the 12m CFHT are extremely preliminary "first pass" estimates and to a large degree are based on KECK. Allowing for existing facility assets, the estimated cost is about US$70 M (1997), excluding any instrumentation, to place a new telescope on the existing structures.

The costs for the NORDIC telescope are also highly preliminary since little detail design work has been done at this stage; they must be considered to be on the optimistic side.

14. Tentative Time Schedule for Implementation of New Telescope

Construction schedules for the 3.6m CFHT, the 8m GEMINI MK, the KECK I telescope and the 25m NORDIC telescope, shown in Figure15 to 18, are reproduced here and were used to arrive at a realistic estimate for the construction time of the 12m CFHT. For a quick overview and for ease of comparison, the number of activities was reduced to a bare minimum and all activities incorporate numerous sub-activities not shown here.

For a large telescope the overall construction schedule appears to be about 8-9 years from start to the stage of completion for at least part time science. This was also true for the 3.6m CFHT, as a large percentage of the detail design of the telescope was done prior to the Canada-France-Hawaii 1973 MOU, when it had been a French national project for several years.

The time schedule estimate for the 12m CFHT is shown in Figure 19. The overall schedule is about 7 years from start to the stage of completion for part time science. It is shorter than the above schedule because of the savings in site development, foundations, main building, etc. It is assumed, however, that some of the dismantling and modification activities can be con-current with the new dome erection activity.

15. Estimate of Shut-down Time to Rework Facility

Coast Steel Fabricators Ltd. provided estimates for the dismantling and removal of the dome and telescope.¹⁰ They would proceed with the dismantling of the dome enclosure first. This would take about 5½ months and would be followed by the dismantling of the telescope, which would take about 3 months. Since CSF were responsible for the assembly of the telescope at site, these estimates should be very realistic. The telescope was erected after completion of the dome by them. They designed and provided special equipment to allow telescope components to be brought from a truck bed at ground level, lifted up through the large hatch and onto the observation floor. The overhead crane was then used to accurately place the telescope components.

CSF suggest that, for dismantling, the dome should be removed first and then the telescope. This will reduce the requirement for special equipment to move telescope components; an externally situated and conventional 120 to 150 ton crane can then handle all telescope and dome components.

The above activities, as well as others, are shown in Figure 19. Some modification to the building might start even earlier than shown, but this activity is not critical to the overall schedule. Activities such as removal of the concrete floor at EL 13,772'-0" and modifications to the pier top are assumed to take place simultaneously. It is estimated that a shut-down time of between 3½ to 4 years is required in order to implement the conversion of a 3.6m facility to a 12m facility.

16. Further Study Areas Identified:.

i.. Modification to pier

• azimuth track concrete thickness
• method of attachment of new concrete to existing concrete
• suitability of pier top edge for azimuth track
• determine foundation characteristics with new telescope for tilt due to wind loading, natural frequency of complete structure, vibration feed back from enclosure to telescope.

v. Error budget

Although not in the original terms of reference, CFHT subsequently requested that the study be extended to include an 8m telescope. In order to complete the range it was decided to also have a brief look at a 10m and an 8.5m telescope. The primary mirror for all these telescopes will be of the segmented type because the individual segments can be handled in the existing building and enclosure confines. Segmented mirrors also result in much lower overall weight for primaries as compared to single dish meniscus mirrors, with the overall benefit of a much lower telescope mass. Table 4 gives a summary of estimated approximate costs for these options and are based on the 12m base line design.

i. 10m CFH Telescope

Although further detail work is called for in several key areas, it appears highly probable that a 12m segmented mirror telescope is the maximum size that can be accommodated on the existing CFHT pier. The main limitations for a large telescope appear to be the existing dome track diameter, the structural adequacy of the pier top, the size of the rotating floor and the adequacy of the existing pier foundation. The existing dome would have to be replaced with one of non-spherical shape and inferior aerodynamic characteristics.

Alternatively a 10m segmented mirror telescope replacement is a much better fit for the existing pier and building structure. It also would necessitate dome replacement for one with a 12m shutter opening but otherwise the replacement would be almost identical in size, shape and profile and might even utilize some of the existing dome components.

The cost for a 12m and a 10m telescope conversion would be about US$70,000,000 and US$58,000,000 respectively, excluding instrumentation. The estimated shut-down time of the CFHT facility would be about 4 years.

An interesting alternative and a variation to the above is a design with a 10m mirror reduced to an effective 8.5m mirror by removing the two outer rows of segments. Such a mirror would be approximately rectangular in surface area and its width would correspond to the existing shutter opening, necessitating a close coupling of dome rotation to telescope azimuth rotation. The existing dome would be retained as is, resulting in considerable overall savings and a one year reduction in conversion shut-down.
 
 

19. References

1. "Overview of the performance of the W.M. Keck Observatory," J.E. Nelson, P.R. Gillingham, S.P.I.E. Proceedings, Advanced Technology Optical Telescope V, Vol 2199, March 1994.

2. "Optical Design of a 25m Telescope for Optical Wavelenths," M. Owner-Petersen, T. Andersen, A. Ardeberg, S.P.I.E. Proceedings, Advanced Technology Optical Telescopes V, Vol 2199, March 1994.

3. "A new 4-mirror optical concept for very large telescopes with spherical primary and secondary mirrors, giving excellent field and obstruction characteristics," R.N.Wilson, B. Delabre, F. Franza, S.P.I.E. Proceedings, Advanced Technology Optical Telescopes V, Vol 2199, March 1994.

4. "Breaking the 8m barrier; one approach for a 25m class optical telescope," A. Ardeberg, T. Andersen, T. Korhonen, P. Søndergard, ESO Conference on Progress in Telescopes and Instrumentation Technologies, Garching, 27-30, April 1992. Pub. European Southern Observatory, ed. M.-H. Ulrich.

5. "GEMINI 8m telescope critical design review," GEMINI Project Office, RPT-TE-G0018, March 1994.

6. "Preliminary study of piers and foundations for an 8m Telescope on Mauna Kea, Hawaii," KPA Engineering Ltd., April 1991.

7. "Stiffness of the ground to support the pier of JNLT atop Mauna Kea," F. Tatsuoka, Y. Kohata, H. Karoji, A. Miyashita, S.P.I.E. Proceedings, Advanced Technology Optical TelescopesV, Vol 2199, March 1994.

8. "Preliminary Geotechnical Services U.C. Ten-Meter Telescope, Mauna Kea, Hawaii," Harding Lawson Asociates, August, 1984.

9. "Soil Investigation, KECK Observatory 10-Meter Telescope, Mauna Kea, Hawaii," Harding Lawson Associates, November, 1985.

10. Personal communication from D. Halliday, Coast Steel Fabricators Ltd., January 10, 1997.

11. "The Hobby-Eberly Telescope: A Progress Report," T. Sebring, L. Ramsey et al, S.P.I.E. Proceedings, Optical Telescopes of Today and Tomorrow, Vol 2871, May-June 1996.

20. List of Figures
1. CFHT 3.6m Telescope Facility
2. Plan & Sectional View of Pier Top End
3. Cross-section of Azimuth Track
4. GEMINI MK 8m Telescope Facility
5. KECK 10m Telescope Facility
6. Segmented Hyperbolic Primary Mirror
7a. 25m NORDIC Telescope Optics
7b. 25m NORDIC Telescope & Enclosure
8. 12m Telescope Plan View & Base Frame
9. 12m Telescope on CFHT Pier
10. 15m Telescope Plan View & Base Frame
11. 15m Telescope on CFHT Pier
12. Enclosure Model for Large Telescope
13. 12m Telescope Enclosure Plan View
14. 15m Telescope Enclosure Plan View
15. 3.6m CFHT Construction Schedule
16. 8m GEMINI MK Construction Schedule
17. 10m KECK I Construction Schedule
18. 25m NORDIC Construction Schedule
19. 12m CFHT Construction Schedule
20. 10m Telescope on CFHT Pier
21. 8.5m Segmented Primary Mirror
22. 8.5m Telescope on CFHT Pier
23. 8m Segmented Primary Mirror
24. 8m Telescope on CFHT Pier
25. Variation of the SST Pupil
26. Cutaway View of UT/Penn State SST
27a. 12m Parabolic Primary Mirror for 0.4\072 Field
27b. Spot Diagram
28a. 12m Spherical Primary Mirror for 0.16\072 Field
28b. Spot Diagram