GNIRS Configuration Design Review
March 15, 1999
 

---

Agenda

TIME	ITEM	PRESENTER
8:30-9:00	Introductory Remarks	Gaughan
9:00-10:00	Requirements	Elias
10:00-11:00	Baseline Configuration	Muller
11:00-12:00	Cold Motors and Mechanisms	Goble
12:00-1:00	Lunch
1:00-2:00	Mechanisms	Goble
2:00-3:30	IFU and OIWFS Integration	Elias
3:30-4:00	Design Issues	Gaughan/Elias
4:00-4:30	Schedule and Status	Gaughan
4:30-5:00	Discussion and Adjourn

 

• Project Tracking
• Drawing Breakdown Structure
• Instrument Weight Budget
• Tracking of Progress

• Primary Support Truss Analysis

• Solid Modeling of Ray Traces

Serrurier Truss Design

•  Translational deflections, no rotation

• Optical Bench, Dewar, Electronics Enclosures
• Put weight into primary support truss instead of separate support structure for Electronics
• Simplifies design – less structure to build

• No damage to ISS

• Manufacture in-house

 Calculations based on “Analysis for redesign of 150-inch Stellar Telescope Serrurier Truss Structure” by M. K. Abdel-Gawad, 1969
 
 

	RESULTS
Total Load on Truss:	2000 Kg  (4400 Lbs.)
Square tube width:	2.5”
Square tube wall thickness:	0.25”
Maximum Deflection:	±0.0027”, 69µm (±0.0007”, 18µm for ¼ g load)
Max loading safety factor:	29.5
Buckling safety factor:	26.0

 

1 Introduction

This document is intended as feedback to the GNIRS team from an informal presentation/review of the instrument held 3/15/99. The order of topics follows roughly the presentation order.

2 Instrument layout (baseline configuration)

The instrument layout presented is mechanically efficient, provides good serviceability and provides the required interfaces to the Gemini facility air pallet. I have no overriding objections to what is proposed and my comments are by way of suggestion, omissions (crane interfaces) or ease of manufacture.

2.1 Truss structure

The truss structure looks good in concept and layout. There is good access to the ISS mounting
fasteners and according to the calculations, the decision to support the electronics off this structure has not compromised rigidity too much.

The interface pintels to the Gemini facility handling cranes have not been added yet. I would not anticipate a problem doing this although this may require some beefing up of the two vertical side members nearest the ISS (perhaps using section instead of solid bar for efficiency?). The handling will also require a lifting 'A' bracket at the middle of the Vertical installation frame. These features allow the instrument to be lifted and tilted from side to upward looking.

The team may want to consider the issue of pinning/bolting the ISS truss ends to the square frame (and at the truss apexes) instead of welding. This is an ease of fabrication/machining issue, not a performance issue.

2.2 Vertical installation frame

It is great to see this feature in at the start. It is an efficient way to provide this interface although the frame restricts access a bit.

The team may want to consider the issue of pinning/bolting the ISS truss ends to the square frame (and at the truss apexes) instead of welding. This is an ease of fabrication/machining issue, but may also allow better access by providing a means to remove one or more truss members.

I have no strenuous objection to the current layout but I think it is worth considering having a square, flat-sided dewar shell that can support the instrument without the need of this frame. The driver for this is ease of manufacture of the Instrument Support Bulkhead, dewar shells, thermally isolating structure and increased room within the instrument.

2.3 Horizontal installation frame

As with the vertical installation frame, it is great to see this feature in at the start.  The frame restricts access a bit but I think this can be easily addressed as the design progresses.

As with the Vertical Installation Frame, squaring off the Instrument Support Bulkhead would simplify this component.

2.4 Instrument support bulkhead

Having this central structural support that supports everything else and provides all the feed-throughs is the way to go.

For fabrication, I believe an Aluminum weldment of flat plates will be most economical.

The thermal isolation/support of the cold mass is a crucial part of the layout of this component. I would strongly advocate G10 truss structures. Being flat, it is difficult to see how these can be inserted between two cylindrical structures, but maybe only a facet is required on the bulkhead and cold structure for each truss (I appreciate the inner cylinder diameter on the ISB is representative).

2.5 Dewar shells

The thin walled stainless concept looks good to me. I have no experience with spun components and so I cannot comment on cost or ease of manufacture but I think it unlikely that the cost would be more than fabricating a 'square' dewar shell from Aluminum plate (that I am suggesting as an alternative for other reasons). The stainless 'cans' will be somewhat lighter too.

2.6 Electronics thermal enclosures

These look to be ideally located for access. If I would have a suggestion on the support brackets it would be to provide a small measure of position adjustment to/away from the ISS for CofG trimming.

3 Mechanisms

I believe the decision to use cold stepper motor mechanisms and datum/step counting open loop control is a good way to go. The main disadvantage appears to be the risks inherent in the fact that NOAO lacks experience in implementing this. The GNIRS team have a good grasp of the design issues but should contact ROE and the IfA (and anyone else they can get with similar experience) who are using cold stepper motors to discuss common problems/solutions and experience. (I might be of some help in contacting ROE).

3.1 Stepper motor selection

Why not use one of the motor types (and controller) in use by ROE or IfA? At least consider these motors if the performance is similar. This has some advantage for serviceability within Gemini too. It will be easy to get basic information on the motors, cost, torque, max speed, steps/rev etc.

3.2 Datum switches

It is a good idea to develop a common datum switch and actuator for all the mechanisms.

3.3 Stepper motor shielding/heatsinking

The measures discussed at the meeting may be overkill for GNIRS. Isolating the motor mounts, shielding and low conductivity drive shafts were used on Michelle, a 10-20 micron instrument with a cold structure about 45K. On CGS4, a 1-5 micron spectrograph with the structure about 75K, there were no problems with the stepper motors bolted directly to the structure (although not directly in the view of the detector).

3.4 Gears/worms

A word in favor of worm gears. Overall, worm gears are hard to beat for these applications. They give a high gear reduction for the minimum number of components. The worms can be made in house on a standard screw-cutting lathe (an important issue when using hi-tech plastic worms).

I think the planetary gear reduction/leadscrew/rack/pinion assembly is too complicated given the alternative.

3.5 Balancing mechanisms

I think the point is accepted about balancing mechanisms but this may be difficult to implement in the case of the translating IfU/slit assembly. I still like the idea of a translating stage but I think the consequence is that Counterbalance weights must be used (as in a sash window).

3.6 Datuming accuracy, point to point accuracy and repeatability

The system design note SDN0002 by Jay is a comprehensive treatment of the accuracy requirements of the mechanisms. However, the sequence of use and correction through calibration can often relax the accuracy requirements significantly to a less demanding repeatability requirement. (An example was identified by Brook at the meeting; I cannot recall what it was).

4 IFU

4.1 Interface detail on IFU Interface Control drawing

I would suggest that the main datum surfaces for this housing be on the outside surfaces and that the datums for the IfU be more localized.

4.2 Translation stage

See comments in the 'balancing mechanisms' section about balancing of this translated assembly.

5 Mass balance

It is good to see the mass balance table in such detail and I think the right approach is to aim for a lower mass than the limit, perhaps getting closer as project develops.

I think the new layout is very efficient in terms of mass and that the mass budget will be met without undue difficulty.

Home | Management | System Design Notes | Requirements | Configuration Images | Subsystems | Archives | Staff | FTP Site | Internal Information |

NOAO Intranet Services

NOAO Copyright  Statement