Antenna Working Group Update, Sep96

Peter Napier, Chair

Minutes of MMA Antenna Working Group Meeting, 21 Aug 1996.

This phone meeting was attended by J.Cheng, D.Emerson, J.Kingsley, J.Lugten and P.Napier. The agenda was:

(1) Current work on new conventional design.
(2) Modeling of BIMA antenna to test resonant frequency prediction.
(3) Status of mount comparison report.

Main points of discussion: (1) Jingquan reported that by using an asymmetric yoke design he is able to reduce the yoke contribution to wind pointing by about 40%. The increase in weight is about 2 tons. For improved close packing and wind pointing it is desirable to decrease the depth of the reflector backup structure. Decreasing the depth from 2m to 1.5m causes the horizon surface gravity rms to increase from 18 micron to 25 micron. Zenith rms is unchanged at 6 micron.

(2) After discussion it was concluded that it is not worth the investment in Jingquan's time to reanalyse the BIMA structure to predict resonant frequencies. This time would be better spent doing a detailed analysis of the dynamic performance of the new MMA design.

(3) The mount comparison report is waiting for attention from Peter.

Notes of MMA Antenna Technical Meeting, 12-13 Sept 1996.

This meeting was attended by J. Lugten, J. Kingsley and J. Cheng. P. Napier joined the meeting on Sept 13, 1996. The goal of the meeting was to make some decisions on details of the new MMA Conventional Design so that this design could proceed to the next level of detailing and analysis. An additional goal was to apportion work between Cheng, Kingsley and Lugten.

A wide range of topics were discussed in this technical meeting. Some of these are fundamental and some are design details. The fundamental ones are antenna optics, close packing and wind pointing errors. The detailed ones are bearing specification, base design, cabin design, dish depth, transporter width, foundation, and others.

1. Optics and close packing.

The choice of optics is related to the close packing issue. The present design (Design 1) uses the following numbers: Primary f/ratio F1=0.42, Secondary f/ratio F=12, The distance between vertex to elevation axis L=86 inches. The feedleg support radius Rf> 4 m. The close packing is poor.

If L=72 inches(4 inch panel adjuster and 8 inches for distance between elevation and dish bottom), F1=.4, the secondary mechanism is 21 inches beyond the primary focus, the feedleg is within aperture radius. then distance to secondary, r1= 222 in, safe radius 1.388D distance to dish edge r2=204 in, 1.275D(>26 deg ele)

Other cases considered are:

L= 60 inches,
F1=0.4
r1=210 inches, 1.313D
r2=197 in      1.231D(>23 deg.)

L=60 in
F1=0.35
r1=206 in 1.288D
r2=208.4 in 1.303D(>?deg)

The geometry finally chosen is optimized for minimum blockage so that the diameter of the subreflector is same as the diameter of the hole through the surface required for 30 GHz operation or a 3x3 focal plane array at 100 GHz.

F1=0.38 
r1= 215.6 in  1.348D
r2= 205.6 in 1.285 D(>19.5 deg)
F=9.3, subreflector diameter 18.9 in, central blockage .35%.

dish depth and surface rms error
80 in depth, 66 in cone  zenith 6(resulting 14) um horizon 18 um
60 in        66                 6                          25
60           56                 8                          29 um

conclusion:
f/0.38 may be the best, providing L=72 inches.

2. Wind pointing error:

1/10th of the beamwidth at 800GHz: 1.4 arcsec

wind pointing error

       9m/s    6m/s    +unsymmetric yoke  +independent encoder support
dish   1.8     0.9      0.9                0.9
yoke   2.9     1.4      0.8                0.55
total  4.7              1.7                1.45
weighted 3.5 **         1.3**              1.1**

** weighted point error is error averaged over all pointing
directions(Mark's memo)

Lugten's breakdown of wind pointing error

yoke with independent encoder mounting  0.3
az bearing                               0.05(?)
base                                     0.2
foundation                               0.15
Total   0.52 arc sec

3. Az bearing specification:

Diameter  2.1 m

axial repeatable TIR  30 um
axial non-repeatable TIR 10 um
radial repeatable TIR 30 um
radial non-repeatable TIR 10 um
maxi repeatable wobble  5 arc sec
maxi nonrepeatable wobble 0.4 arcsec

maxi starting torque with normal loading
                     10,000 N-m
maxi running torque with normal loading
                     10,000 N-m
maxi running torque variation with normal
loading             +- 20 %

nominal loads
  radial 340,000 N
  axial  340,000 N
  moment 7,500 N-m

survival loads
  radial  500,000 N
  axial  1,000,000 N
 moment  250,000 N-m

minimum stiffness under nominal loading:
  radial  2,500 N/um
  axial 2,500 N/um
  moment  110,000 N-m/arcsec

Bearing type choice

4 pointing ball: BIMA SMA 12 m, slippage, poor
2 row ball: not recommanded by manufacturer
crossroller: VLA, cost more
3 rollers: SMT most expensive


contacting manufacture Rotex(Lugten), Aven, Kaydon(Kingsley)

4. Foundation:

SMA cost $40-50K/each
It will be significant cost for MMA.

( Chile has a seismic design code of 0.4 g but prediction for MMA site is
0.25 g)

5. Base:

cylinder on top of a triangle. 12 points equal softness. three screws to fix at the foundation.

Foundation nuts above ground to keep surface clean and make it easy for alignment.

Do we need to truncate the angle to minimize the spacing between transporter wheels( Lugten will do deformation analysis later)?

6. Transporter:

4 wheels, width: 20 ft.
wheel capacity: 28,000 Ib/tire,  OD 84 in.

Transporter picking points are above the azimuth bearing for
easily aligning with the foundation(rotating the AZ drive system).

all wheels have steering ability.

7. Yoke design

unsymmetrical design, allowing for horizon pointing or 5 degrees above horizen.

At one side, design independent encoder support. If the support uses CFRP, the 50 degrees thermal change produces 1.8 mm difference due to 3 m steel expansion. Is it possible to make CFRP that has the same thermal property as steel? Can the coupling take this displacement?

Kingsley will prepare a sketch.

8. Cabin design

Lugten suggested that there is no ring under the dish structure. Lugten will try to optimize the 32(16) support points for the dish bottom plate. Kingsley will check the structure using Nastran program.

cabin size: 3 m x 3 m x 2 m

Cabin shape stability(?)

9. Dish design

The new design( DESIGN 2) should has a depth of 60 inches, a f/ratio of 0.38, and the feedleg support points just on dish radius.

Using CFRP as the main material.
The bottom support ring  58 inches.
Using a cylinder to replace the cone.

panel weight 25 Kg/m^2 or less(SMA 28 Kg/m^2),
panel style: BIMA type

panel adjuster(check the thermal expansion force of 10 deg
on the adjuster bending).
weight  1 kg/ supporter.

for providing the possible scaling up to 10 m. using 5 ring
and 32 sections.

Joint weight: ~ 12 Kg(SMA 1300 kg/240 nodes)

weight of secondary(SMA 40 Kg, tetrapod SMA 144 Kg) ? Ib

weight of sun shielder(5 kg/m^2?)

dish center hole 22in dia.

dish bottom coordinate in z direction +6 in(?).

Last modified 4 Oct 96

mholdawa@nrao.edu