0% found this document useful (0 votes)
22 views3 pages

Deflection Measurement Methods Explained

Deflection can be measured absolutely or relatively, each with its own advantages and disadvantages. Absolute measurements focus on the center deflection but require corrections for device-specific factors, while relative measurements eliminate frame deflection using reference supports but may reduce sensitivity. An ideal approach combines both methods to ensure accuracy in deflection measurements during testing.

Uploaded by

3tarebitaraf
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
22 views3 pages

Deflection Measurement Methods Explained

Deflection can be measured absolutely or relatively, each with its own advantages and disadvantages. Absolute measurements focus on the center deflection but require corrections for device-specific factors, while relative measurements eliminate frame deflection using reference supports but may reduce sensitivity. An ideal approach combines both methods to ensure accuracy in deflection measurements during testing.

Uploaded by

3tarebitaraf
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Annex “Deflection measure: Absolute or relative?

The deflection can be measured in two ways: With the reference to the move less ‘ground’
frame of the 4PB device (absolute) or with reference to two symmetrical placed supports
resting on the beam itself (relative). Both measures have their advantages and disadvantages.

Absolute deflection measure

In general the absolute deflection measure is only used for the determination of the deflection
at the centre of the beam where the largest deflection occurs. From a sensitivity point of view
this measure is fine, especially given the low deflection values (< 1 mm). However, the
‘ground’ frame of the 4PB device maybe really move less, but only in theory no
deformation/movement will occur in the outer supports which form the connections between
the bended beam and the move less ‘ground’ frame. The stiffness of the bending frame (in
N/m) is not infinity. The outer supports will deform/bend a little bit when a load is applied.
Therefore a correction has to be applied on the measured deflection before this value is used
in the back calculation procedure. Because no 4 PB device will be exactly the same the
correction factor has to be determined for each device separately. The determination of the
correction factor is based on a simplification of the 4PB device as a system of two springs
(Kframe & Kbeam ; [N/m]) in series. The total deflection Vtotal is given by equation 1.

Vtotal = Vframe + Vbeam = Force.{1/Kbeam + 1/Kframe} (1)

The first spring represents the stiffness of the beam and the second one the stiffness of the
bending frame. If a very, very stiff beam is used it is assumed that the measured deflection is
only due to the deformation of the bending frame. So, for each individual 4PB device using a
very stiff beam a plot can be made of the measured deflection as a function of the applied
load. Ideally this should be a straight line through the origin which will enable an easy
correction procedure in the software for the back calculation procedure. Of course the beam
used for this correction procedure should as stiff as possible e.g. a thick and high beam made
of steel (E = 210 GPa). The geometrical form of this beam is not so important, only the local
form (width and height) of the beam at the location of the supports should be similar to the
form of the (asphalt) beams. This simple correction procedure can be applied if in practice the
frequency range is limited to 10 Hz (pneumatic actuators). The correction procedure can be
carried out by using a pseudo-static loading. However, the influences of mass inertia forces
should be investigated if in practice higher frequencies are used (> 10 Hz; hydraulic
actuators). Assuming a ‘non bending’ stiff iron beam a mass term should be add in the
simplification in order to determine the correction factor. Another option is to model the
complete device in a finite element calculation (e.g. ABAQUS).

An theoretical important issue is the circumstance in which the stiffness of the frame can not
be simplified as an elastic spring but a phase lag has to be introduced. See Annex “Frame
stiffness”.

Another point of interest is the location of the sensor tip. In most 4PB devices the head of the
deflection sensor is placed at the top or bottom of the beam. In those cases it is assumed that
during bending the height of the beam will not change. It is from a theoretical point of view
‘better’ to mount a sensor support half way the beam at the (supposed) neutral line.
Relative deflection measure

The main advantage of a relative deflection measurement is the direct elimination of the
frame deflection by the use of reference supports on the beam (deflection ‘bridge’). In general
the two reference supports are placed at x = L/5 and x = 5L/6. Theoretically both the
deflection due to pure bending and the deflection due to shear forces are at these locations
exactly half the values of the deflections in the centre (x = L/2) if the length A between the
inner and outer support equals L/3. Therefore by multiplying the measured deflection value
by two the total deflection in the centre can be calculated. In some devices other distances are
used but in those cases the reduction factors (β = V(X)/V(L/2) in both type of deflections are
not the same. This hampers an exact determination of the total deflection (or the deflection
due to pure bending) in the centre. However, the induced errors are small. For the deflection
due to pure bending the reduction factor β as a function of bridge support location is given by
equation 2. Equation 3 gives the reduction factor β for the deflection due to shear forces.

β Bending =
(
4 3LX − 3 X 2 − A2 ) for A < X < L / 2 and =
( )
4 3LA − 3 A2 − X 2 X
for 0 < X < A (2)
3L2 − 4 A2 (3L2 − 4 A2 ) A
X
β Shear =1 for A < X < L / 2 and = for 0 < X < A (3)
A
For A = L/3 the equations can be reduced to equations A4 and A5 respectively.
 X X2   X2 X
36 27 − 27 2 − 1 36 2 − 3 2 
L L L L
β Bending =   for A < X < L / 2 and =  for 0 < X < A (4)
23 23
X
β Shear =1 for A < X < L / 2 and = 3 for 0 < X < A (5)
L
For X = L/6 both βBending and βShear are equal 0.5 (½).

Because in the back calculation procedure only the deflection due to pure bending (Vb) is of
interest, measuring the relative deflection Vt(L/2) - Vt(A) = Vb(L/2) - Vb(A) = Vb(L/2).(1 -
βBending(x = A)) seems ideal. However, in that case βBending = 4A(3L - 4A)/(3L2 - 4A2) = 20/23
for A = L/3. So, instead of measuring the value Vb(L/2) only (1 – 20/23) = 3/23 = 13 % of this
value is measured. From a measuring point of view this is a large reduction in sensitivity.

Another but minor point of concern are the movements in the locations of the bridge supports
(e.g. at X = L/6 and X = 5L/6). These supports will rest on the top surface of the beam.
However, the locations X = L/6 and X = 5L/6 will move inwards to the centre when the beam
is bend. Suppose the supports are fixed to these locations and will follow any movement. Next
to the expected horizontal movement ∆translation due to the bending of the ‘neutral’ line an extra
horizontal movement will occur due to the rotation of the neutral line and the supports rest on
the top surface of the beam. The induced errors are very small and can be neglected.

Ideal deflection measure

A combination of absolute and relative deflection measurements would be ideal. The relative
deflection measurement should be carried out as follows:
1. Most clamping devices consist out of a rotation frame with an axis halfway the beam
height. The beam is fixed (vertical) in this frame by servo motors or springs.
2. A very stiff rod/beam is placed on the axis’s of the two outer clamping devices in such
a way that if the axis’s move (rotation and horizontal translation) the rod/beam can
move freely (e.g. ball bearings).
3. A deflection sensor is mounted to this rod/beam at X = L/2 with its sensor head placed
on a pin halfway the (asphalt) beam. This requires a special construction.
4. It should be checked that this bridge is stiff enough and will not bend due to its own
weight.
5. A second deflection sensor is used to measure the deflection at X = L/2 with refrence
to the move less ground frame.
6. If the stiffness of the bending frame is high enough (infinity) the two measured
deflections should be the same.
In this way one can check if the correction procedure for the absolute deflection measure is all
right.

Next to the complicated construction this measure is in theory only applicable for constant
deflection fatigue tests. If constant strain cyclic fatigue tests are wanted, only the relative
deflection measure with the bridge supports at X = A and X = L – A will guarantee a constant
strain amplitude in this interval (See Annex “Constant Deflection versus Constant Strain”).

Common questions

Powered by AI

Integrating absolute and relative deflection measurements provides an ideal solution by combining the ability of relative measurements to eliminate frame deflection with the sensitivity of absolute measurements for center deflection . This method uses the measured deflection at the beam's neutral line and verifies it against absolute deflection under a fixed ground frame, ensuring consistency when the bending frame's stiffness is high enough. However, theoretical limitations include its primary applicability to constant deflection fatigue tests and potential complexity in construction when ensuring the supporting rod remains straight and unaffected by its weight .

The deflection due to pure bending is differentiated from shear force-induced deflection using reduction factors for each type of deflection. For pure bending, the reduction factor β is calculated based on the bridge support location, which then allows the deflection at the center to be deduced from measurements taken at specific span placements. In contrast, the deflection due to shear forces employs a simpler reduction factor allowing distinction between the two deflection causes based on theoretical compliance with structural behavior .

The placement of deflection sensors is crucial for accuracy, as it directly influences the measurement outcomes. Sensors located at the top or bottom of the beam assume no change in beam height during bending, which might not be accurate. It is theoretically better to mount a sensor support halfway along the beam at the neutral line, where deformation effects are minimized and a more accurate measure of bending-induced deflection can be obtained .

When using a stiff beam for correction procedures in a 4PB setup, it is crucial to ensure that the local form (width and height) of the beam at the support locations is similar to the tested beams (e.g., asphalt beams). While the overall geometry of the beam is not critical, its local geometry at the support points must mimic the beam under test to ensure accurate results. This helps in accounting for frame deformation and applying necessary corrections .

Using a "non bending" stiff iron beam in high-frequency applications implies considering mass inertia effects in the deflection corrections. This requires an additional mass term in the deflection model to account for inertia forces, particularly when using hydraulic actuators that surpass the typical frequency range of pneumatic systems. Such considerations are necessary to maintain measurement accuracy and avoid skewed results due to unaccounted dynamic effects .

Maintaining constant strain amplitude in cyclic fatigue tests involves careful selection of deflection measurement techniques. The relative deflection measurement ensures this constraint by placing bridge supports precisely at interval ends (X = A and X = L – A), thus preserving constant strain across the span during the tests. Any deviation in supports’ location or miscalculation of reduction factors challenges the constancy in strain amplitude, risking test integrity unless compensated through precise setup adjustments .

The stiffness of a bending frame affects deflection measurements because it can cause additional deflections that need to be corrected for accurate measurements. In high-frequency applications, where hydraulic actuators are used, mass inertia forces also need consideration. To address these effects, a finite element model (e.g., ABAQUS) can be used to simulate and correct the deflection behavior, or a phase lag might have to be introduced in the model if the frame cannot be simplified as an elastic spring .

Absolute deflection measurement in a 4PB device is advantageous for determining deflection at the center of the beam, where the largest deflection occurs. It offers fine sensitivity, especially for low deflection values. However, complications arise due to the potential movement of the outer supports, which connect the bending beam to the 'move less' ground frame; these supports may deform slightly under load. This requires a correction to be applied to the measured deflection, determined individually for each device by modeling the system as two springs in series .

Relative deflection measurements are preferred for eliminating the deflection of the frame since reference supports are placed directly on the beam itself. This method can accurately determine the total deflection in the center by using pre-determined locations for reference supports where deflection values can be doubled to find the center deflection. However, limitations include potential inaccuracies if the beam supports move or if the assumed reduction factors do not match exactly due to differing bridge lengths, leading to small induced errors .

Reduction factors, β, are critical in translating relative deflection measurements into total center deflection values in a 4PB setup. They serve as the bridge in the back-calculation process by linking the deflection values at specific support positions to the central deflection. These factors account for distribution patterns of bending and shear deflections along the beam, enabling accurate calculation of the total deflection based on measured data, which are inherently reduced by positioning and type of deflection considered .

You might also like