|
Previous Page
|
|
|
|
The Famous Bay
Bridge Crack
Page 2
Tony Alfrey (tonyalfrey at earthlink
dot net)
Contents
Repair 3.0
Caltrans
must not have been happy with what they found in their daily
inspections. They certainly couldn't see
inside the actual metal parts to see if there were cracks developing
(although we saw on Page 1 that they can use ultrasound on some
components), or
if there is wear between the Tie Rod, Nut, and Crossbar, but they can
monitor their strain gauges and the hydraulic pressure in their
actuators. Engineers don't like to see things changing;
their measurement might not be exactly the one they'd like to have, but
if they see that the numbers are constant, they feel better. If
the numbers change day-to-day, they begin to wonder.
As of mid-November,
we
heard
that
Caltrans proposed placing an additional Eyebar
alongside the cracked
Eyebar by using a longer Pin (the Pin on which the Saddle rests).
That was impressive.
Pulling out and replacing the existing Pin between the Eyebars is
A Big Deal. If they
were going to go to this much effort, my
opinion
is that all of the Eyebars
in the set
should be replaced so that they all match (have the same length).
The entire Saddle/Crossbar assembly of Repair
2.0 would have
to be removed and a new tooling scheme would have been needed to take
the
tension off of the cracked Eyebar, because Repair 2.0 relied on
exactly the pins that will be lengthened.
And so it begins!!
But Caltrans was actually designing Repair 3.0 in earnest without
revealing much publicly. A
press release appeared on
Dec. 7, and on the evening of Dec. 11, Repair 3.0
began. How does Caltrans go from design speculation in
mid-November, to a full-blown computer-aided design and a pile of parts
ready for installation in about four weeks? They had help.
Here are the graphics from baybridgeinfo.org. The steps
are self-explanatory but I've added some comments and questions.
Summary:
a. Unload the cracked eyebar completely by tightening
the Repair
2.0 tie rods.
b. Cut out the upper section of the cracked
eyebar.
c. Replace the upper section with a four-piece assembly.
d. Install brackets to allow the new upper section and old lower
section to be tensioned.
e. While tensioned, drill and bolt together the old eyebar
section and new eyebar section.
f. Remove old Repair 2.0 parts.
Details:
Step
1.
 |
|
Repair
2.0 (tie-rods-and-saddle assembly) is kept in position and tightened to
take up the
entire load of the cracked eyebar. A section of the cracked
eyebar is
removed. Note that cutting out the broken eyebar where the crack
is at the top is not trivial. The cut must be made through the
cracked eyebar without damaging the adjacent eyebar. Clearly the
eyebar cannot be cut all the way through with a torch, which will
heat-stress the adjacent eyebar. I'm guessing that they will cut
most of
the way through with an abrasive saw and then carefully pry apart the
two
pieces of eyebar.
|
 |
|
Recall that this is the piece of
eyebar that needs to get cut out. Imagine how
difficult this would have been if the center
eyebar was cracked ! They might make the cut toward the upper
part
of the "eye" so that the eyebar will still carry load, and then the tie
rods may be tightened just enough to "unload" the eyebar, and the
eyebar will be slid to one side. This is tricky; if you've
read our Bay Bridge Tutorial, you know that the tie rod assembly is
really not
capable of taking the full load of an eyebar.
|
But someone has to
explain to me
how this cut is going to be made with the saddle assembly still in
place from Repair 2.0. Go back to this
picture and tell me
how they are
going to cut out the eyebar with the saddle/crossbar/tie-rod
stuff on top of, and halfway surrounding, the pin. The eyebar
cannot be cut out
with Repair 2.0 in place and it certainly cannot be cut out
while under load. We'll be waiting for photos on this!
Dec. 15 - Caltrans has managed to
cut the eyebar head out in
pieces; see remarks and photos below in the progress report section. Someone
gets an "A" for this stunt.
Step 2.
 |
|
The "eyebar
head" (the part of the eyebar with the crack in it) is replaced
with a "hairpin". This is simply a flat, U-shaped piece of steel
about the width of one-half of an eyebar
that they will loop over the existing pin that passes through all of
the eyebars. This is a novel feature that eliminates the need to
fuss with the pin itself, which was an earlier suggestion from Caltrans
(see above)
that seemed almost laughable. The hairpin has a
whole series of holes drilled into it to accept a "splice plate"
(next graphic). My original guess was that there was simply not
enough room to slide the hairpin in from the top and simultaneously
avoid the
crossbar, but it does indeed fit. See the Progress Report for Dec. 17
and Dec. 18.
|
A "lower
jacking bracket" (the green
thing) is attached to a set of
holes drilled into the lower section of eyebar.
This
bracket will
allow the end of the replacement eyebar to be pulled together with the
old, lower eyebar section to
make things fit and to apply the appropriate load to the replacement
eyebar.
Notice that the bracket is
not attached to the very end of the lower part of the eyebar.
They leave about two feet of space for connecting the new eyebar
section.
Step 3.
 |
|
Two "splice
plates" (upper green things) are bolted to the "hairpin" like a
sandwich and to the
"new
eyebar body". This four-piece assembly replaces the old piece of
eyebar.. At the bottom end, more splice
plates are bolted to the end of the "new eyebar body", again like a
sandwich. The splice plates go on each side of the new eyebar
body and on each side of the old eyebar section, but they are first
only bolted to the new eyebar body, not the lower part of the original
eyebar. Notice the "arms" sticking
off the sides of the new eyebar body. It turns out that these are
simply for the attachment of a shackle for hoisting and manipulating
the component.
|
Step
4.
 |
|
An "upper jacking
bracket" virtually identical to the device
currently in use at the
bottom of the tie-rods-and-saddle-assembly is fastened over the upper
end of the new eyebar body. Note the orange
hydraulic actuators
that will be used to pull the whole mess together. Tie rods are
added between the jacking brackets and they are tightened up to the
amount of tension needed to equalize the tension on all of the four
eyebars in the set that are attached to the big pin at the top.
Note that the reason that all of this works is that there are no bolts connecting the lower
set
of splice plates to the lower, original piece of eyebar - - the two are
free to slide with respect to each other.
|
Step
5.
 |
|
Finally, when the new eyebar
section
has been
properly "loaded" by pressurizing the orange hydraulic actuators, holes
are drilled through the old eyebar section
using the holes on the splice connection as a "template" or drilling
guide. After all the holes are drilled, bolts are inserted and
tightened. I hope they have enough room for the nuts between
the eyebars. There is a problem with this graphic.
For the splice plates to be as strong as the original eyebar, they must
each be roughly half the thickness of the original eyebar (we've
guessed it is 2" thick). The space between eyebars is also
roughly the same. So by the time that 1" splice plates are added
to each side of the 2" eyebar, there is only 1" of space left for the
nut and a lockwasher. See the results in the Dec. 20 Progress
Report.
|
Step
6.
 |
|
Release the
pressure on the hydraulic cylinders and remove all of the
brackets. At this point, they should be able to remove all of the
parts of Repair 2.0.
Repair 2.0 will
need to be cut off with a torch because it wraps around the eyebars at
the bottom. See the cut-out parts in the Dec. 22 Progress Report and compare with this picture.
Hopefully, they'll fill up or paint all the holes
(that are
not shown in these graphics) that will
remain in the new eyebar section after the upper jacking bracket is
removed and in the lower eyebar section after the lower jacking
bracket is removed so as to prevent corrosion for the remaining four
years.
|
I wonder about the bolt holes. The tensile
strength
of the eyebar is determined by the cross sectional area of the
eyebar. Conservative AISC standards say that the
tensile strength of the eyebar should be de-rated by the reduction in
the cross-sectional area of the eyebar. Let's make a guess.
The eyebars are about 14" wide, and the graphic shows three rows of
bolts. I'm guessing that the bolts are at about 1 1/4" diameter,
so
this reduces the cross sectional area of the eyebar by (3 x 1.25)/14 =
27%. In our Bay
Bridge Tutorial, we decided that the eyebars were being used pretty
close to their minimum yield strength, so the holes in the bar make the
replacement scheme somewhat suspect (from the Dec. 16 photos below, the
bolts may be 1"; get out your ruler and calculator and tell me
what you think!).
Repair
3.0 - Progress Report
Dec. 13.
See http://baybridgeinfo.org/eyebar/mediabar
As of this afternoon, Caltrans says that they are installing rigging;
that is consistent with this photo. See the red woven fiber
straps attached between the top of the bridge and the eyebar. A
cut-out section of the old eyebar below the crack is being lowered to
the roadbed. All of the other components that will be installed
consume the space occupied by this section of old eyebar.
Dec. 15
They've
managed to cut out the eyebar head
in
sections without using magic tools.
In the picture on the left, you will see that only the lower half
of the eyebar is accessable to cutting tools; there is simply no space
between the saddle (the assembly of three larges plates on top of the
pin) and the eyebar to get any cutting tools in between.
On the right, we see the reassembled head of the eyebar after
removal. The engineer on the left has his left hand where the
original crack would be and adjacent to his hand is a section of the
neck of the eyebar
that has been drilled away a bit at a time. But to get the head
of the eyebar out, the top side of the eyebar head has to be cut and it
is inaccessable. But
once the neck of the eyebar is cut straight across (just under the
worker's left knee),
the head of the eyebar can be rotated (counterclockwise only!) around
the pin to expose the
top. This is then cut (the portion immediately in front
of the engineer on the right) almost to the pin and then the
remaining eyebar piece is snapped apart.
|
Here is a view
of the same piece that is lying flat in the photo just above. It
used to be upside down from this orientation and on the opposite side
of the pin. The straight cut at the top of the piece used to
connect to the remainder of the eyebar itself. Note that some
tabs have been welded on to the outside edge of the eyebar to help
manipulate the thing. A worker off frame to the left is pulling
on the tab with a rope attached to a nice shackle and clevis pin.
|
Dec. 16
Parts
arrive from Home Depot, ready for
installation.
The Lower Jacking Bracket arrives. In the front, it is resting on
one of the enclosures for the hydraulic actuators.
|
On the left:
The eyebar is gone.
On the right:
The new hairpin and splice plates arrive. I hope they didn't
forget the bolts (but Home Depot isn't far away). Notice
(roughly) the
same number of bolts total
in the hairpin and in the middle of the splice plate. Step #2
begins.
|
|
Dec. 17
While the still photos are of better quality, a screen shot from one of
the videos gives more information. Step #2 finished, progress
made on Step #3. I'm
impressed; it is all downhill from here.
(Photo source)
The new Eyebar section and attached Upper Jacking Bracket.
(Photo source)
The
original plan was to put the Upper
Jacking Bracket in after
the new eyebar section was installed. You will soon see why this
turned out not to be possible! Read on to the Dec. 19 progress
report.
Dec. 18
|
Step
#3 essentially finished; working on Step
#4 below. Rumor has it that getting the hairpin in was a little
tight (as we expected). Videos of the other night show a little
grinding action before the hairpin was installed. But I'm sure
that any grinding was performed on old Repair 2.0 parts that will be
history soon. Sooner is better for sending the old Repair 2.0
stuff to the recyling yard. But again, the
Upper Jacking Bracket and Actuator Housing are all supposed to be
temporary and used just to tension up the new eyebar, splice plates and
hairpin; nice-looking hardware for a temporary tool.
(Photo source)
|
Here are the new
Tie Rods to be used to tension up the assembly. The hydraulic
actuators are the orange cans in the bed of the pick-up truck.
Look at the photo on the left: the actuator housings are actually
only half-cylinders; the hydraulic fittings exit toward the
eyebar.
(Photo source)
|
Dec 19.
This morning's pictures are less than perfect
because there is a work platform in the way that prevents us from
seeing the Lower Jacking Bracket and the place where the holes will be
drilled through the eyebar for attachment of the splice plates.
The hydraulic lines are in place (for pressurizing the actuators) and
several strain gauges are installed; it appears that there are about 3
or 4 per side (the bundle of wires that can be seen
tied to the eyebar to the left). Tie Rods are in
place and other movie files shows the use of the strain gauge
monitoring software to determine the correct hydraulic pressure for the
proper tensioning of the eyebar. After the eyebar is drilled and
bolted, they can't go back and change things easily at all. In
other words, they only get one shot to drill the stub of the remaining
lower part of the eyebar, so I'm sure that engineers will spend some
time on this, making sure that all of the strain gauge readings are
consistent and at the values needed to provide a balanced load on all
the eyebars. Step #4 is nearing completion.
(Photo source)
Just
a piece of observational fluff: The original plan in Step
#4 (see the graphical plan way above near the top of the page) was to
put in the Upper Jacking Bracket after the new eyebar section was
attached to the splice plates, and in Step #6, to pull out the Upper
Jacking Bracket after the new and old eyebar sections were
connected. I saw that the Upper Jacking Bracket and new eyebar
section went in together as an assembly but did not think too much
about it. Now I see why. Note in the photo above that the
old, uncracked eyebar adjacent to the new eyebar passes through the
reinforcing gussets on one half of the Upper Jacking Bracket.
This means that the only way to get the Upper Jacking Bracket and the
Lower Jacking Bracket out after
everything is assembled is to cut them out.
Dec. 20
|
In this photo, we
see the new tie rods used in the new tensioning
system (in the engineer's left hand in the right side of the frame) and
we see the old, cheesy tie rods (in the same engineer's right hand)
used in Repair 1.0/2.0. The sizes are a litle distorted by the
perspective of the photograph, but the new tie rods are substantially
larger. This makes sense because two new tie rods must do the job
of four of the original tie rods. Also, the threads are of a much
finer pitch and appear to be machined threads instead of cast or forged
threads (a geeky difference only of importance to geeky engineer types).
(Photo source)
|
|
The crew made very
rapid progress. This photo shows that the eyebar
has been loaded and drilled, and the lower jacking bracket has already
been cut off. The hydraulic actuators and tie rods have been
removed.
Step #5 is complete, Step #6 is well under way.
(Photo source)
|
|
As I expected,
this half of the lower jacking bracket (which used to be below the
splice plates in the photo to the upper right) had to be cut out.
(Photo source)
|
|
Earlier
I was worried that there was not enough space for splice plates,
nuts and lockwashers between the eyebars. It turns out that the
eyebar spacing seems to be a little larger than 2", and only 1" bolts
are used, with no lockwashers, so there seems to be enough space.
|
Dec. 22
Early on
Dec. 21
when no new photos
were available,
Caltrans essentially finished the
job! We still have no photos of the completed installation, but
some Repair 2.0 parts were removed early on the morning of Dec.
22. A drive across the bridge on Dec. 24 showed the remains of a
few vibration damper parts still bolted on to the eyebars.
Here, the Tie Rods are gone, and the Lower Saddle Assembly has been cut
out and is being
carried away and off to the scrap yard somewhere. Compare the
image below to this
image. Can you see what was cut to remove the assembly?
Maybe it will
turn into a nice plant stand at the Caltrans office. I vote for
sending it over to Gordon Biersch on Embarcadero; they can put a
big beer keg into the saddle.
For the
obsessive among us, here is a
picture of the upper
saddle. The inside crossbar (extreme left) was cut into pieces so
that the entire assembly could be removed. It was originally
installed in several pieces and welded together up in the air.
Refer to here to
see what the parts looked like as they were going
in. The "Stinger Welding" label is attached to the saddle.
This part
used to be sitting on top of the pin closest to the
camera in
this picture. The eyebar repair consists of (from top to bottom)
the hairpin, the first set of splice plates, the new eyebar and the
second set of splice plates. The old eyebar is not visible, out
of frame to the bottom. Wires from strain gauges are seen
everywhere. The lights of Oakland, Berkeley and Emeryville are in
the background.
_________________________________________________________________________________________________________________________________________
Footnotes
****Eyebar
Picture. The piece of eyebar toward the bottom of
the frame must be upside down with respect to the piece of eyebar
higher in the frame. Paint drips down, therefore his left hand is
on the top of the eyebar
head. Also, we're looking at the
unpainted side that faced the other eyebar. Finally, compare the
width of the plates at either end of the section.
