39 40 42_Layout 1 07/04/2014 16:47 Page 1

PRODUCT FOCUS

OVERHEAD CRANES

CRANE GIRDER CAMBER

An overview of overhead crane girder camber, deflection and bending strength. By Gary J. Davis

ABOUT THE AUTHOR

Gary J. Davis, P.E. is director
of consulting services,
Integrated Machinery
Solutions (IMS), Fort Worth,
Texas. He can be contacted at
GaryDavis@team-ims.com.
Fig. 1: "Free" camber

lant maintenance and

inspection personnel are
occasionally faced with
reports of "lost camber" or
"excessive deflection" of crane
girders. This article provides a
practical overview of crane
girder camber, deflection, and
bending strength to aid plant
personnel in evaluating these
conditions.

GIRDER CAMBER: WHY IS IT

REQUIRED?
Cambering effectively removes
the dead load deflection and
averages the live load deflection
relative to the level profile.
Cambering effectively cuts the
live load deflection in half. This
minimises the trolley rail slope
during live load deflection. The
reduced slope means less power
is required to drive the trolley
uphill toward the end of the
girder and less tendency to coast
downhill toward the center of
the span.
Fig. 1 shows the
recommended camber of 100%
of the dead load deflection plus
half of the live load deflection.
"Free" camber means the girder
is free to assume its unloaded
shape, for example, no dead
load or live load is present.
Camber is usually built into
the top profile of the girder. If
there is limited clearance over
obstructions near the bottom of
the girder, the bottom profile
may also require cambering. Fig.
1 shows a crane girder cambered
on the top and bottom,
unloaded, and laying on its side.
Girders are cambered by
building them with a convex
("crowned") profile, opposite to
the concave rsag produced by
deflection. The camber in Fig. 1
APRIL 2014

a small laser level and a ruler to

measure the elevations. The
trolley should be unloaded and
moved to the end of the bridge
before measuring. A string or a
wire cannot be used as a
reference line; regardless of the
amount of tension in the line, it
will sag too much to produce
meaningful results.
FIELD DATA ADJUSTMENT
If the laser and/or girder are not
level, adjust the field data as
shown in Fig. 2. Make a scale
layout of the data, as shown in
Fig. 2a, and connect the ends
with a straight line. The
adjusted elevations are
obtained relative to the line
connecting the ends of the
curve. Using the adjusted data,
the camber is the difference
between the maximum
elevation and the end point
(Fig. 2b). Typically, the middle
third of the span may appear to
be practically flat.

Fig. 2: Camber measurements

Fig. 3: Girder profiles

is equal to the deflection that

would be produced by the
girders own weight (DL) plus
half of the combined weight of
the trolley and the rated capacity
(LL).

CAMBER FIELD
MEASUREMENTS
The camber profile can be
obtained by measuring trolley
rail elevations at five-foot
intervals as shown in Fig. 2. Use

INTERPRETATION OF CAMBER
MEASUREMENTS
The camber requirement
should be obtained from the
crane manufacturer (or a
qualified person) and
compared to the field
measurements. Fig. 3 shows
ideal camber profiles for no
load, half rated capacity, and
full rated capacity. Fig. 3d
illustrates the difference
between a normal dead load
deflection profile and a
deformed profile from a severe
overload. After a severe
overload, cambered and uncambered girders may have a
slight "kink" near the center of
the span. When the measured

LIFT AND HOIST INTERNATIONAL LIFTANDHOIST.COM 39

39 40 42_Layout 1 07/04/2014 16:47 Page 2

PRODUCT FOCUS

camber is less than the required

amount, check the following:
1. Determine if the original
fabrication included the
required camber.
2. Check for buckled flanges
and webs.
3. Check the trolley rail for
local deformation between
the diaphragm support
points.
4. Check for local buckling at
the top edge of the
diaphragms that support
the trolley rail.
5. Check the welds for the
diaphragms supporting the
trolley rail.
6. A borescope is required for
inspecting internal
components.
7. Determine if the
diaphragm spacing for the
trolley rail support is
adequate. (Engineering
required).
8. Check the condition of the
bottom flange for fractures.
9. If the elevation profile
shows a negative camber
greater than what would be
produced by the dead load
deflection, and/or the
profile shows a "kink" (see

OVERHEAD CRANES

% RATED CAPACITY
Stress
Range
(ksi)
17
13
10

First Onset
of Yielding

First Measurable
Camber Loss

Ultimate Bending
Strength

185
185
305

200
255
330

310
390
510

3.

4.

Fig. 4: Estimated percentage overload for camber loss

Fig. 3d), it may indicate a

severe overload.
In addition to this list, the
entire structural and
mechanical load path should be
thoroughly inspected. Refer to
Reference 1: Guidelines for
Inspecting Overhead Crane
Structures for more information.
GIRDER DEFLECTION
Girder deflection is not directly
related to strength. No
conclusions can be made about
girder strength based on a
deflection measurement alone.
A crane may have a large
deflection and ample bending
strength, while a different crane
may be overstressed and show
very little deflection. Design
code deflection limits are based

GIRDER DEFLECTION IS NOT

DIRECTLY RELATED TO STRENGTH. NO
CONCLUSIONS CAN BE MADE ABOUT
GIRDER STRENGTH BASED ON A
DEFLECTION MEASUREMENT ALONE.
40 LIFT AND HOIST INTERNATIONAL LIFTANDHOIST.COM

on the live load only and

determined as a percentage of
the crane span.
MEASURING DEFLECTION
Some purchasing specifications
require deflection measurements
during load testing for design
verification. Deflection
measurements are only
meaningful if the magnitude of
the lifted load and trolley weight
are accurately known.
Deflection can be measured
relative to a point on the floor
using a laser distance meter and
a small level as follows:
1. Move the unloaded trolley
to the end of the bridge.
2. Mark a small target area at
the center of the crane
span, on the bottom flange

5.

6.

7.

8.

9.

of the girder, visible from

the floor.
Place the base of the laser
meter on the floor and use
the level to ensure that the
laser beam is plumb in all
directions while pointed on
the target.
Mark the spot on the floor
where the laser meter is
plumb and pointing on the
target.
After marking the laser
base location, the laser
should already be plumb
for later measurements and
the level should not be
needed.
Measure and record the
distance between the
marked spot on the floor
and the target spot on the
girder.
For this example, lift 50% of
the rated crane capacity
and move the load to the
center of the span.
Measure and record the
distance from the marked
spot on the floor and the
target spot on the girder.
The deflection is the
difference between the two
measurements.

40 ton single failure-proof trolley

built per U.S. Navy specifications.
APRIL 2014

39 40 42_Layout 1 07/04/2014 16:47 Page 3

PRODUCT FOCUS

25/25 ton Monorail Stoplog Hoist built

per U.S. Government Specifications.

If the required camber is

unknown, it can be determined
by measuring the deflection with
50% of the rated crane capacity.
The deflection measured in this
example should be
approximately equal to the
required camber as shown in Fig.
3a.
YIELD STRENGTH AND LOSS OF
CAMBER
If no other girder elements have
failed, loss of camber does not
indicate loss of bending strength.
Steel is elastic below the yield
strength and the girder will
spring back to its original shape
after the load is removed. Design
codes require the stress to be
below the yield strength at rated
capacity. If the load is large
enough to exceed the yield
strength by more than about
10%, the girder will not return to
its original shape and this will
appear as lost camber.
The next time the girders are
loaded (equal to or less than the
rated capacity), they will deflect
the normal incremental amount
and the stress will be at the
normal level. The only difference
is that the starting point for the
deflection increment will be at a
lower elevation due to the lost
camber. Additional camber will
be lost each time the girder is
loaded past the yield strength.
Some ductility will be lost, but
the bending strength of the
girder does not decrease.
INTERNAL RESIDUAL STRESS
Weld shrinkage from girder
fabrication creates zones of
longitudinal residual stress at

OVERHEAD CRANES

(or near) the yield point. This

locked-in stress will cause a
slight camber loss after load
testing. This type of camber
loss is a one-time event and is
too small to measure. Initial
loading with the test load
provides mechanical stressrelief for the welds by
removing most of the internal
stress in the working stress
range.
ULTIMATE GIRDER STRENGTH
At the first onset of yielding,
only the outermost surface of
the girder starts to yield and
the remainder of the section
remains elastic. The yielded
portion can only support the
load that caused the first
yielding, and nothing more. As
the load increases, more of the
section yields. When the entire
cross section has yielded, the
girder cannot support any
increase in load. If additional
load is applied, the girder will
not offer any resistance and
only more deformation will
occur. This is the ultimate
strength for bending of the
cross section. The girder can
fail by other limit states
(buckling, shear, welds, etc.)
before ever reaching the
ultimate strength.
OVERLOADING SCENARIOS
A computer model for bending
of a typical box girder was
used to estimate the
magnitude of overload
required to cause a measurable
camber loss. Three different
stress ranges are listed to
represent girders designed for
different utilization classes.
The results are shown in Fig. 4.

The threshold for a

measureable camber loss was
assumed as 1/8". At the first
onset of yielding, the girder
still returns to its original
shape when the load is
removed. The ultimate bending
strength is included as an
upper limit reference. The
results shown in Fig. 4 are
based on specific assumptions
and are only intended to
provide an "order of
magnitude" for the various
loadings required for camber
loss.
The analysis shows that
camber loss from an overload
is certainly possible, but the
likelihood depends on the
girder design classification. It
also depends on the type of
hoist drive. A.C. magnetic
wound rotor hoist drives can
deliver up to a 300% overload
based on the motor "pull-out
torque".
D.C. series magnetic hoist
drives can produce up to 600%
of the rated motor torque for
short periods. Present-day VFD
drives cannot produce an
overload large enough for
camber loss. Typical VFD
drives monitor the motor
current and shut down at

about 125% of the rated

capacity. When considering
overload multiples of two or
three times the rated capacity,
a "load snag" scenario would
be a more likely operator error
versus hoisting a freely
suspended load. The ultimate
bending strength for wide
flange girders is about 40%
less than box girders.
CONCLUSIONS AND
RECOMMENDATIONS
Loss of camber (for box
girders) due to an overcapacity
lift requires lifting 185 to 200%
the rated crane capacity. Other
failure modes can occur before
reaching the load that causes
camber loss. If overload is
suspected as the root cause for
camber loss, the entire
structural and mechanical
load path should be
thoroughly inspected. Consult
with an expert when making
this assessment and deciding
on a course of action. If inhouse expertise is not
available, consult with an
engineering firm that is
regularly engaged in structural
design and inspection of
overhead cranes and hoisting
equipment.

REFERENCES
1. Gary J. Davis, Guidelines for Inspecting Overhead Crane
Structures, Integrated Machinery Solutions, Fort Worth,
Texas, July 2011. Full text available at:
http://www.scribd.com/doc/83331227/Guidelines-forInspecting-Overhead-Crane-Structures-Full-version
2. Charles G. Salmon, John E. Johnson, Steel Structures,
Design and Behavior, Intext Educational Publishers, New
York, NY, 1971.
3. R. L. Brockenbrough, B. G. Johnston, Steel Design Manual,
United States Steel Corporation, Pittsburgh, PA, 1981.

IF IN-HOUSE EXPERTISE IS NOT AVAILABLE, CONSULT WITH AN ENGINEERING FIRM THAT IS REGULARLY
ENGAGED IN STRUCTURAL DESIGN AND INSPECTION OF OVERHEAD CRANES AND HOISTING EQUIPMENT.
42 LIFT AND HOIST INTERNATIONAL LIFTANDHOIST.COM

APRIL 2014