Intro

• Page 1:
• top: total tiedown tension vs hour of day
• The cable break occurred at 02:35:55 (red line)
• The telescope stopped moving at 02:56:12 (blue line)
• Before the break, the tiedown tension was about 95kips
• During the break, the tension max,min was:  75 to 150 kips.
• It took about 3 minutes for the oscillations to die out.
• The up/down pattern in the tension was caused by the motion of the dome, azimuth arm (td12 had no tension.. see page 2).
• middle: the telescope azimuth during the break.
• The break occurred at 02:35:55.
  
• The telescope stopped for a fraction of a minute, and then the operator commanded it to start tracking again.
• the telescope finally stopped moving at 02:56:12
• Bottom: dome za during break.
• Page 2: blowup of az,za during break
• Top: azimuth position. Each * is a 1 second sample
• The break occurred at an azimuth of 29.5 degrees.
• 11aug20:
•  the .733 deg azimuth jump turns out to be and encoder jump. It came out of the rack gear, spun a little and then went back in).
  
• the telescope  jumped .733 degrees in the second of the break.
• It had moved  .016 degrees on the previous second and it was slowing down.
• The jump was not part of the programmed motion of the azimuth.
  
• The azimuth position of the auxmain connection is 22.4 deg az.
• It it strange that the cable hit the dome on the south side (since dome az > aux4connection point az) but the dome moved to a larger azimuth (CW).
• When the cable broke, it pivoted on the tower 12 auxmain cable rather than falling vertically down. I need to compute where that  pivot point is in azimuth. It will be at a larger azimuth than 22.4 .
• It could be that the azimuth motion is actually the azimuth encoder jumping a tooth on the azimuth encoder rack gear.
• Since the telescope bounced, the encoder may have lost contact with the rack gear momentarily. one tooth on the azimuth gear is .02 deg, so .7 deg is about 35 teeth (pretty large jump)
• The dome was moving slowly in the +az direction. If the azimuth wheels on the dome side lost contact with the rails momentarily, the ch side of the azimuth may have overdriven the azimuth arm.
• Another possibility is that the dome rotated when hit by the cable causing the dome to move north ,while the azimuth trolleys moved south (probably not).
• Bottom: dome za during the break
• The dome was at 10.72 deg za when the cable broke (this is below the ring girder at za=10)
  
• There is a small glitch in the dome position during the break, but nothing like the azimuth motion.
• The dome motors drive on a rack gear, so there is little possibility of sliding.
  
• The azimuth motors are just sitting on the azimuth rails, relying on gravity for the friction.
• The dome is currently at 12.69 deg za and the cable is hanging partly against the dome.
• When the cable broke the dome was at 10.72. Then the cable fell it may not have hit the dome until it swung past vertical (below the tower12auxmain).
  
• Page 3: tensions in the individual tiedown cables druing the break.
• Top: all 6 td cable tensions for 1 hour around the break.
• The pairs of cables at each tiedown should have the same tension (since the yoke holding both cables can pivot the equalize the tensions).
• Any differences are most likely errors in the dc offset or gain of the loadcell amplifiers.
• middle: blowup 350 seconds around the break
• You can see the oscillation of the platform
• bottom: blowup +/- 10 seconds around the break.
• the tension in td12 went from 2*10kips to 0 kips in about  2 seconds. This is because the td12 corner went way down when the 100kips of load  of the aux4 cable was lost. this cable is about 60 feet from corner 12 along the triangle edge (the triangle edge is 216 feet long).

SUMMARY:

• tower 4 northern auxmain cable (cable #301) broke loose from its tower 4 socket at 02:35:55 on 10aug20.
• At the time of the break:
• azimuth = 29.5 deg
• domeZa=10.68 degrees.
• The telescope continued moving for 30 seconds  before the operator stopped it.
  
• azimuth=29.2 deg
• domeZa=10.72 deg
• At the jump the azimuth jumped by .733 degrees (CW).
• Turned out to be an encoder jump (11aug20)
  
• If the cable was pushing it, you would expect it to move in the other direction.
• 60  seconds after the break, the operator restarted the telescope motion
• It continued to move for 20 minutes
• The final position was:
• az=47 degrees
• domeZa=12.8 deg (it had moved all the way up to 13.6 degrees).
• At the break the total td tension oscillated 75 to 150 kips.
• The aux4North cable connects to the platform at an azimuth of 22.4 deg az.
• The broken cable hung from where it crossed the tower12AuxMainE cable. I did not compute this azimuth position (yet).

Compare platform position after 2 failures

The plots show how the platform moved before and after the cable failures (.ps) (.pdf)

• The x axis:
• -20 to -10: 10 days prior to the 1st cable failure
• -10 to 0 : 10 days prior to the 2nd cable failure
• 0..x        : days after the 2nd cable failure
• Top: platform avg and corner height positions
• bottom: angular rotation about axis
• + rotation is CCW looking down the axis
• Black: rotation about X(west) axis
• Blue: rotation about Y(North) axis
• Red: rotation about Z(vertical) axis

Note: i will update the plot when more data becomes available.

processing: x10/201120/telrotlr.pro

Monitor platform corners

    After 06nov20 cable failure, distomats 3 and 1 no longer hit the retro reflectors. On 20nov20 they were realigned.

The plots show the corner positions of the platform (.ps) (.pdf)

•  measured by the 6 distomats. updated every 10 minutes.
• Top: corner height above sea level
• Black trace: the average platform height (just averaging the 3 corners)
• Blue trace: corner 4 height
• brown trace: corner 8 height
• red trace : corner 12 height
• dashed purple line: average height of the platform when the optics is in focus.
• Bottom: the rotation angle about the x,y,z axis.
• A positive rotation angle is counter clockwise looking down the axis.
• Black: rotation about the x axis. positive is pointing west
• blue:   rotation about the z axis. positive is pointing vertically up.
• red :   rotation about the y axis. positive is pointing north.

Also take a look at position change during failures;

What was on this page Prior to 06nov20 cable failure :

The telescope is currently positioned at:

• az=257.65, domeza=8.46, ch=8.83
• this is close to the hurricane stow position (but the pins are not yet in)
• The azimuth arm balances when the ch is at 8.83 degrees and the dome is at 7.9 degrees..
  
• So there is a little imbalance pushing corner 4 up and corner 8 down.
• The tiedown cable jacks have been moved up as far as they can go to release tension on the platform.
• the current cable  tensions are:
• Corner 12 .. no tension (since it is so low)
• corner 4:  about 40 kips (since it is high)
• corner 8:  0 to 8 kips
  
• These values will change a little with temperature.
  

Every two minutes the distomats measure the distance to the corners of the platform.
After each reading, a  plot is generated showing the position of the corners (.ps) (.pdf)
( to the following plot, you may have to refresh the page).

• Top: corner height above sea level
• Black trace: the average platform height (just averaging the 3 corners)
• Blue trace: corner 4 height
• brown trace: corner 8 height
• red trace : corner 12 height
• dashed purple line: average height of the platform when the optics is in focus.
• Bottom: the rotation angle about the x,y,z axis.
• A positive rotation angle is counter clockwise looking down the axis.
• Black: rotation about the x axis. positive is pointing west
• blue:   rotation about the z axis. positive is pointing vertically up.
• red :   rotation about the y axis. positive is pointing north.
• The variation in the positions is normally caused by temperature changes.
• If we see sharp change in a single position, it may tell us that something is happening with the cables near it.
  

The auxmain4North cable was attached about 60 feet before the corner 12 vertex. It was carrying about 100Kips of vertical load.
When the cable let loose

•  corner 12 dropped about 2 feet.
• This is seen in the position, and the .6 degree rotation about the east-west axis
• It also moved toward the west. This is seen in the .5 degree rotation about the vertical

Note: if you seen horizontal lines in all the traces.. this will be missing data when the computer failed to read the distomats.

Table of when things occurred that may affect the position

11aug20: moving the telescope to stow position

    On 11aug20 we moved the dome and azimuth arm from the location after the failure to close to the hurricane stow position.

The plots show the telescope motion and the change in the tiedown cable tensions.
Note that positive azimuth angle increases Clockwise (north to east).

Plot 1:overview of telescope positions (.ps) (.pdf)

• Page 1 10aug20 at time of failure. the telescope continued to move for 20 minutes after the failure.
• Page 2: 11aug20  when we moved telescope to hurricane stow position
• Page 3: 20aug20 when we moved the telescope a few degrees for the final hurricane pin insertion.

• Top: azimuth motion
  
• middle: gregorian dome za. this is the optical za. the center of mass za is 1.9 degrees less.
• bottom: the tension in each platform corner from the tiedowns.
  
• This is the sum of the 2 cables in each corner.

processing: x101/200917/telescopePos.pro, 200811_poskips.pro

13aug20: measuring the distance between the auxmain connection points

    On 13aug20 felix and I used the tm50 theodolite to measure the distance between the connection points of some of the auxmain cables to verify how long any replacement cable needs to be. This will be an approximate value since the platform is currently not in the standard position  (see monitor positions above)... but the cables are currently spanning the distances we measured.

    The technique was:

• Place the theodolite where we could see both connection points
• make 3 separate measurements of the tower pin, then 3 measurements of the platform pin.
• We then moved to a different position (usually within 50 feet of the previous location) and then repeated the measurements.
• the xyz locations for the 3 measurements at each spot were averaged and then the difference was used to compute the distance between the points.

The table below contains the results:

• The xyz  measurements are the distances from the theodolite positions to the point being measured.
• the z direction is vertical up.
• the x,y direction form a plane perpendicular to z. The origin is at the theodolite. We didn't bother to define a 0 for the azimuth so the orientation is a bit arbitrary  (but that doesn't matter since we're only interested in the differences of the two points).
• The theodolite recorded 2 images for each point. An overview and a high resolution image of the point.
• The measured location will be the exact center of each image.
• The longest theodolite measurement was 1280 feet to tower 8  from in front of the control room.
  

• AuxMain connection point Measured lengths (all measurements in Feet)

  cable/Position #	Platform Point			Tower Point					platform		Tower		Notes
  	x	y	z	x	y	z	distance	avgDist	ovrView	hiRes	ovrView	hiRes
  aux8North Pos1	-521.406	-571.829	225.813	-1159.633	-280.938	351.733	712.606		jpeg	jpeg	jpeg	jpeg	1
  aux8North Pos2	-651.977	-418.762	225.909	-1199.302	19.896	351.820	712.628		jpeg	jpeg	jpeg	jpeg	2
  								712.617
  aux4North Pos1	-358.960	-468.738	129.520	-142.594	-1136.603	254.269	713.036		jpeg	jpeg	jpeg	jpeg	3
  aux4North Pos2	520.059	-232.064	129.861	1146.807	84.310	254.593	713.066		jpeg	jpeg	jpeg	jpeg	3
  								713.051
  aux4South Pos1	-122.990	532.102	228.354	577.866	573.964	352.275	712.958		jpeg	jpeg	jpeg	jpeg	4
  								712.958

  Notes:

• 0: the rsi drawings has the auxmain 301 cable lengths  at 713.2 ft +/- 1 inch at 602Kips. 
  
• 1:
•  platform side: the hanging cable did not bother the measurement.
• tower side: we shot under the horizontal handrail.
  
• 2:
• platform side: we are way clear of the hanging cable, and we got the same value as position 1. so position 1 was ok.
• 
  
• 3: aux4north is the cable that came out of the socket.
  
• tower side: we were not able to shoot the pin.
  
• The measurement was made to the top front of the socket face above the socket hole.
• looking at the high res images, it almost looks like the pin is bent.
  
• 4: We could only find a single position to shoot aux4south .
• 5: There was no accessible spot where we could should the pins for aux8south.

processing: x101/survey/200813/

12nov20 - distance top of corner 4 td mast to broken cable

Notes:

• The measured distance from the top does include the 1 or 2 feet that the lighting arrestor wires stick up (we couldn't get a reflection from them).
  

processing: x101/survey/201112/t4mast/cmpdist.pro

12nov20 - distance from winch base to top of T4

    Winches will be used to raise the failed T4auxNorth cable. On 12nov20 we measured the distance from the center of the
cement block used to mount the winch to the top of tower 4. We measured two blocks:

•  Cement block at SouthWest corner of the helipad
• Cement block close to tower 4 by the road.
  

The table below shows the distances. It includes pictures of the site line from the block to the tower top.

Notes:

• I forgot to turn on the cross hair in the images. The center of the image is the measured point
• The Z value above includes 5.60 feet (the height of the theodolite measurement point above the cement block).
  
• 1. The measured point is just to the right of hole for the missing pin
• 2. The measured point is under the flooring below the hole for the missing pin.
• 3. The measured point is on the upper most cable (t4Main north) where it crosses the handrail.