A jackscrew (or screw jack) is a type of jack you will have probably seen stowed in your car but they are used extensively in other mechanical devices. They operate by turning a leadscrew to produce linear motion of a ‘load-nut’. An acme thread form is most commonly used, as this thread is very strong and can resist the large loads imposed on most jackscrews. Jackscrews are self-locking, which makes them intrinsically safer than other jack technologies like hydraulic actuators which require continual pressure to remain in a locked position. However if not lubricated they can be dramatically weakened through wear. One of our failure investigations concerns such a failure of a jackscrew device. We are bound by confidentiality agreements from publishing details of the investigation but whilst researching another more prominent and more widely reported incident was discovered. This incident had far more devastating consequences but featured an identical failure mechanism to the one we are studying.
ExcelCalcs Repository Calculation 1 - As is our practise we publish any calculations that might be useful to other engineers on the ExcelCalcs site. Clearly we do not publish any confidential information but in the course of our work we produce generic calculations that are likely to be useful for future problems, these we upload to ExcelCalcs. One such calculation determines the adhesive wear rate of two metals which we can apply to our jackscrew investigation. It employs some data published in “Wear Coefficients - Metals Section IV of 10” by E. Rabinowicz and “Fundamentals of Machine Component Design” by Juvinall. It considers the compatibility of various metal combinations. Materials with increased levels of compatibility have an increased tendency ‘weld’ together and a higher wear rate. It also considers the hardness of the material and the contact pressure between the two materials. Reflecting on the incident it is unlikely that unsuitable materials were selected but it is possible that pressure on minimising weight resulted in selection of a smaller leadscrew operating under high contact pressure and a correspondingly high wear rate.
ExcelCalcs Repository Calculation 2 - A second calculation in the repository considers the strength of jackscrew load-nuts and how the strength of the load-nut reduces as it wears. What is particularly interesting is the difference between the wear rates on a lubricated jackscrew are almost 100 times lower than an un-lubricated one. The calculation is a very useful to set an inspection intervals to ensure that un-lubricated jackscrews will be identified before it leads to a failure. Using the calculation you can identify the most important parameters to decide the best way to enhance the safety of your designs.
The calculations sadly makes it all too clear to see how inadequate maintenance and poor lubrication led to excessive wear and catastrophic failure aboard Alaska Airlines Flight 261. See BBC News Coverage of the incident.
Terminology
Local nominal stress: Nominal stress including macro-geometric effects, concentrated load effects and misalignments, disregarding the stress raising effects of the welded joint itself.
Structural stress: A stress in a component, resolved taking into account the effects of a structural discontinuity, and consisting of membrane and shell bending stress components.
Structural discontinuity: A geometric discontinuity due to the type of welded joint, usually found in tables of classified structural details. The effects of a structural discontinuity are (i) concentration of the membrane stress and (ii) formation of secondary bending stress.
Local notch: A notch such as the local geometry of the weld toe, including the toe radius and the angle between the base plate surface and weld reinforcement. The local notch does not alter the structural stress but generates non-linear stress peaks.
Notch stress: Total stress at the root of a notch taking into account the stress concentration caused by the local notch. Thus the notch stress consists of the sum of structural stress and nonlinear stress peak.
Notch stress concentration factor: The ratio of notch stress to structural stress.
Hot spot: A point in structure where a fatigue crack may initiate due to the combined effect of structural stress fluctuation and the weld geometry or a similar notch.
Hot spot stress: The value of structural stress on the surface at the hot spot (also known as geometric stress or structural stress).
Figure 1 – Hot spot positions a, b and c and explanation of hot spot and notch stresses
Figure 2 – Determine hot spot stress by extrapolation of FE result
Determine Hot Spot Stress for Fatigue Assessment. t is not necessary to model the weld profiles for the following reasons:
• Actual welds may have a variety of profiles and it is not possible to model every variant.
• The SN curve includes the notch effect at the weld toe based on a statistical assessment of a many fatigue tests.
• Inclusion of the weld makes determination of the hot spot stress more difficult to determine. Compare the stress along paths with and without the weld profile modelled.
This crashworhiness assessment considered re-engineering the front end of a leading vehicle in a train to improve vehicle crash performance. The study provides an interesting comparison between a vehicle with no provision for crashworthiness (unmodified) and one with a 2MJ collapse structure on the leading vehicle (modified). Which train would you rather be in? The top one or the bottom one? Now play the video and see for yourself the benefits of the designing for crashworthiness. The article outlines the calculations for the train end crash structure. It considers kinetic energy and its conversion to strain energy and compares resulting forces and deceleration rates of the two trains.
These videos are best viewed in full screen mode. Press play and double click the screen to maximise.
Figure 1 - Parametric Model
Figure 2 - Basic Energy Calculations
Figure 3 - Velcocty vs Time
Figure 4 - Impact Force vs Time
Figure 5 - Acceleration vs. Time
Figure 6 - Summary
The detail
The 1D Analysis Model - This analysis considers GM/RT 2100 Issue 4 crash scenario 1 (similar trains impacting with a closing speed of 60kph). A dynamin model has been developed for this crash assessment. To maintain client confidentiality the vehicle is referred to as the EUKV. The loadcase starts when the train is travelling at 5m/s and is 2m away from the wall. Initially a time step of 10ms was considered over a period of 3 seconds. Finally the analysis was re-run with a time step of 1ms over the first second (as the impact is pretty much over within this time).
Determine the Model Parameters - The calculation above determines the material properties that are assigned to the LSDYNA elements to model the mass and crush behaviour of the EUKV, coaches and loco (driving the train forward from the rear - see Figure 1).
Basic Energy Calculations - The model parameters and verification calculations are set up using Excel (see Figure 1). The script for model generation and initial conditions is generated from the Excel worksheet too so that a number of scenarios could be quickly assessed.
The natural frequency of the train vibrating as a rod is calculated because this can be observed in some of the time history plots in the results section (see Figure 1). The train travels 0.2 s before impact with the wall, the structure deforms during the impact and then recoils backwards vibrating at this frequency. The combined load defection characteristic of the complete train is easily constructed from its constituent parts. The plot above shows the load defection characteristic for the modified train which includes the crash structure fitted to the front end of the EUKV. It also calculates the area under the curve for each straight line segment (i.e. the energy absorbed). It is interesting to note that only small amounts of energy are absorbed by the elastic deflections and the majority of the absorption occurs when the structure plastically deforms. Once the crash structure is fully collapsed the main structure begins to be plastically deform with the resulting loss in survival space for passengers and increase in compression loads.
Two simple calculations are presented below to estimate the loss of survival space for the unmodified and modified trains. The calculation serves to present the basic mechanics to aid understanding of the problem (see Figure 2). Results - The results are presented first for the unmodified EUKV and then the modified EUKV. Velocity - Figure 3 compares the velocity (m/s) vs. time. The train travels at 5m/s for 0.2 seconds before the front of the EUKV strikes the wall. You will observe the short time interval as the shock wave travels down the train and each of the cars in turn begins to decelerate. The train recoils from the wall at about 1.5m/s vibrating as it does so (note the resonant frequency can be observed - see calculation in section 0). The impact is more or less complete for both the unmodified and modified trains in less than 1s. Impact Force - Figure 4 compares compression force levels (N) vs. time. In the unmodified vehicle the impact force quickly rises to the 4MN level when the cars begin to collapse. As the shock wave travels down the train and each car sees the compression load a fraction of a second after the car in front. When the modified vehicle strikes the wall compression loads are initially limited to 2MN until the crash structure completely crushed after about 0.4s and then the load rises to the 4MN level as the cars then begin to collapse. Acceleration - The net force on each car is calculated from the force information and the mean acceleration obtained by dividing by the mass of the car. This information is presented in Figure 5. prEN 15227 requires that accelerations are limited as far as practical to 5g and should not exceed 7.5g. It concedes that it will usually be necessary to accept higher levels of acceleration in the cab but only as transients lasting less than 5ms.
In the unmodified vehicle acceleration levels peaks at 9.1g and is consistently greater than 7.5g for time periods longer than 5ms. Thus it does not comply with the requirements of prEN 15227.
The modified vehicle shows significantly reduced acceleration levels rising to a peak of 7.6g but generally less than 5g. It is interesting to note that the peak 7.6g acceleration occurs at the point where the crash structure is fully collapsed and the carbody begins to collapse. The peak acceleration is greater than 7.5g for just 3.5ms thus the modified vehicle meets the requirements of prEN 15227. Survival Space - prEN 15227 also requires that the loss of survival space shall be limited to 1% of the initial lengths. prEN 15227 also requires us to consider both passenger and crew survival spaces. The un-modified EUKV may carry train staff. The loss of survival space is compared for the un-modified and modified EUKV in Figure 6.
The un-modified EUKV shows a 0.641m loss of survival space (3.83%) failing the prEN 15227 criteria. The loss of survival space in the coaches is less than 1%. The total loss of survival space for the whole train is 0.879m which compares well with 0.899m estimated in the Excel calculations.
The crash structure in the modified EUKV is entirely crushed but the survival space is reduced by only 0.339m (2.03%). Most of this is lost towards the front of the EUKV. Clearly this has implications for the driver but his cab is rigid compared to the structure in the cab door area. This means that the driver’s survival space will be retained at the expense of the cab doorway structure. Immediately behind the cab is an equipment area with no passengers or crew. It is likely that most of the loss of survival space will occur here nearer the front of the vehicle rather than the passenger saloon area in the rear. The loss of survival space in the passenger saloon (final two thirds of the vehicle) is likely to be less than 1%. The loss of survival space in the coaches is less than 1%. It is concluded that the modified EUKV will pass the prEN 15227 survival space criteria. The total loss of survival space for the whole train is 0.386m which compares well with 0.399m estimated in the Excel calculations.
Input: Drawings supplied in PDF format by email.
Deadline: Three weeks after purchase order.
Output: Deveop strategy with the client assess a number of impact scenarios and prepare final calculation report suitable for third party scrutiny.
This analysis considered the structural performance of the dummy pod as it moves around the wheel in combination with exceptional wind loads experienced 140m above the ground. The structural assessment of the dummy pod considers the stress in members, the buckling of members, the stress in welded connections and the performance of bolted joints.This article describes a finite element analysis study supported additional calculations which are available for download from the www.ExcelCalcs.com site.
A Pod
FE Model
Stress in FE Model
Bolt Calculations
Design Code Calculations
The Detail - The dummy pod is constructed from two end
sub-assemblies called thimbles and a central cylindrical sub-assembly
called the barrel. The assemblies are fitted to a support ring. A bolt
passes through the thimble end plate, through the support ring and
through the barrel end ring. 32 M20 bolts per support ring are used
around the circumference to connect the parts together. The support
ring is fixed in three directions at the pin locations.
An ANSYS finite element model is constructed from beam elements.
Stresses are determined by linear elastic analysis and Euler buckling
analysis is also performed. There is no particular code of practise
applicable to the London Eye dummy pod. BS2573 Part 1: 1983 “Rules for
the Design of Cranes” is selected as cranes are mechanical structures
subject to wind and self weight loads. It is not a limit state code and
requires elastic assessment of the structure. It is similar to BS449 in
its methodology. The basic allowable stress is 60% of the material
yield (unless limited by buckling considerations). A fatigue assessment
is not considered necessary due to the low number of cycles. The model
has assigned the material density of steel multiplied by 1.2 as an
impact factor. All material is S355 steel with a yield stress of
355MPA. A number of load combinations were considered.
How to read ANSYS plots: The “STEP” number (and “TIME” number) is a
load step (e.g. self weight, wind load…). A “SUBSTEP” is of interest
for nonlinear analysis but since this assessment is linear elastic so
it is not relevant in this case. “SEQV” indicates that we are plotting
equivalent stresses (or Von Mises Stress) in MPa. For beam elements
SEQV is the addition of the axial stress and the bending stress in two
planes. The stress is displayed at all points around the beam section.
SEQV indicates the stress magnitude but not the sense (compressive or
tensile). “SMN” gives the lowest value of SEQV in the plot and “SMX”
shows the highest value. The stress legend shows how colours can be
related to the value of equivalent stress in the plot. “DMX” is the
maximum vector displacement in the model (in mm).
The worst axial force and the worst shear force are incorporated into a
detailed bolt assessment. The calculations check the following:
1) The joint axial strength.
2) The joint preload is sufficient to withstand the applied shear load even after embedding losses and bolt relaxation.
3) The tightening torque requirement.
4) The joint force diagram.
Eigenvalue buckling analysis can be performed by ANSYS. Eigenvalue
buckling analysis predicts the theoretical buckling strength (the
bifurcation point) of an ideal linear elastic structure. This method
corresponds to the textbook approach to elastic buckling analysis: for
instance, an eigenvalue buckling analysis of a column will match the
classical Euler solution. However, imperfections and nonlinearities
prevent most real-world structures from achieving their theoretical
elastic buckling strength. Thus, eigenvalue buckling analysis often
yields unconservative results, and should be used with caution. However
if the Euler buckling load shows a large margin it is very unlikely
that effect of real world imperfections will compromise the structures
resistance to buckling. In addition members are assessed against the
requirements of Eurocode 3.
Input: Preliminary drawings supplied by client.
Deadline: Two weeks after purchase order.
Output: Design change recommendations and a calculation report for final design suitable for third party scrutiny.
ASCE705W Wind Loading 1.0 ... for quite a while, finally here it is....."ASCE705W.xls" (version 1.0), a new spreadsheet workbook for wind loading analysis per the ASCE 7-05 Code. This workbook is obviously similar in form and fu... SimpleBoltCalc.xls This is a pretty simple bolt calculation from first principles. For a given bolt size it looks up the characteristics of a bolt from a 'Bolt Lookup Table'. A bolt preload (or clamping load) is estim... BoltedJoint.xls 02Purpose
of calculation: Determine force distribution in a preloaded bolted
joint. Calculation Reference: http://en.wikipedia.org/wiki/Bolted_joint
OR Schaum's Outline... Bolt spacing and edge distances.xls 1.0Calculate the minimum and maximum fastener spacing, end and edge distances for steel plates. BOLTGRP.xls v2.7BOLTGRP.xls"
is a MS-Excel spreadsheet workbook for analyzing bolt groups. There are
8 worksheets which replicate the AISC 9th Edition Manual Tables XI thru
XVIII to be used for analysis. EC3 Calculations V0001.xlsAssessment of hollow circular section under load to the requirements of Eurocode3. The following code checks are performed: EC3 5.4.4 (1) D:C ratio axial force EC3 5.4.5.1 (1) bending stress D: Clevis and lug design V0001.xls 1.1Calculation of stress concentration effects at a pin and clevis joint. This calculation requires the XLC addin and the XlXtrFun to be installed. Aerospace_lug_analysis.xls 1.0the proof and ultimate strength of lugs in an expedient manner. The spreadsheet has been produced by Abbott Aerospace an independent consultancy company offering structural engineering services the a...
Bolted joints are one of the most common elements in construction and machine design yet they are the root cause of half of the failures investigations undertaken by MoreVision. Every engineer experiences some difficulties with bolts at some point in his career. So what typically goes wrong and what can we do to prevent bolt problems? Follow our bolt failure checklist and find calculations to help you.
Bolt bench test using a loadcell.
Determine torque-tension relationship by test.
Tapped bolted joint bench test.
Strain gauged bolts for on-site testing.
Installation under a railway vehicle.
Harmonic anlysis of test results
Bolt thermal analysis
Random vibration analysis
Junkers vibration test.
Bolt load analysis
Automaton hold down bolts
Insufficient Clamp force? - Usually
by applying a measured torque load to the nut bolted joints are
tightened to achieve a specific clamp load. Even under the most extreme
applied loads, the clamping force must prevent joint movement between
clamped parts. Movement includes both opening of the joint to form gaps
and slipping. Loads applied to the joint may be axial forces (in the
direction of the bolt axis) and/or shear forces (perpendicular to the
bolt axis). If slippage occurs then the joint may fail by the bolt
loosening. If a gap in the joint opens then a bolt failure by fatigue
is more likely to occur. Typically bolt fatigue failures occur because
of insufficient preload rather than poor fatigue strength of the bolt.
Improving the method of tightening can reduce the scatter in bolt
preload and help guarantee the minimum required clamping force.
Excessive Applied Force?
- If a very high axial force is applied to a joint the bolt may be
loaded beyond its yield strength. If a direct tensile failure does not
occur immediately then plastic deformation will result in preload loss
and subsequent failure by loosening or bolt fatigue. Knowledge of the
applied loads may require calculations, finite element analysis or
strain gauge testing.
Joint Relaxation? - Joints
which loose bolt preload in service are said to suffer from joint
relaxation. There may be a number of reasons for joint relaxation.
Embedding occurs when high bearing pressure under the nut face exceeds
the compressive yield strength of the joint material. The resulting
plastic deformation gives rise to loss of bolt preload. Differential
thermal expansion in joints comprising dissimilar materials can lead to
loss of bolt preload and prolonged exposure to high temperatures will
increases the risk bolt creep.
Thread Stripping? -
Thread stripping is a shear failure of an internal or external thread.
Thread stripping tends to be gradual in nature and it may go unnoticed
at the time of assembly. This may have disastrous consequences on
product reliability.
Bolt Investigation Techniques
Bolted joints are one of the most common elements in construction and
machine design yet they are the root cause of half of the failures
investigations undertaken by MoreVision. All bolt problems can be
solved by understanding the mechanics of the bolted joint, the loads to
which it is subjected and the environment to which it is exposed. The
table below shows common techniques that we use to solve bolt design
problems.
Analysis/Simulation
Test/Verification
Bolted Joint Assessment
Joint modelling.
Dissimilar materials.
Thread mechanics.
Control of tightening.
Relaxation, embedding.
Finite element analysis.
Vibration analysis.
Thermal modelling.
In service testing.
Signal processing and analysis of results.
Our Bolt Investigation Team
John Doyle is a UK chartered mechanical engineer consulting in the
fields of simulation, finite element analysis and engineering
calculations. John’s website www.morevision.co.uk shows many bolt failure case studies. John also runs the engineering website www.ExcelCalcs.com
which includes many bolt/thread calculations. His analysis and testing
services have been used for railway vehicles, construction equipment,
oil and gas plant, cranes and mechanical items for theme parks.
MoreVision enlist the support of Bill Eccles, also UK chartered
mechanical engineer who researches and specialises in bolts. Bill’s
aptly named www.boltscience.com
website covers all variety of technical matters regarding bolts. Bill
wrote and developed BOLTCALC which is highly recommended for the
analysis of bolted joints. Rest assured that Bill is arguably the
world’s most eminent authority on bolts. It is unlikely that you have a
problem that Bill has not already encountered some way in his career.
The ‘simple bolted joint’ can be anything but simple. MoreVision and
BoltScience offer a unique skill set to solve your bolting problems.
Please contact John Doyle if you require our assistance with your own bolt problems.
– Email all the joint details (bolt specification, clamped materials,
tightening procedure, applied axial and shear forces, temperature) and
using our databases and analysis tools we’ll produce a report which
identifies any area of risk. Maybe you would just like an independent
review of your bolting procedures?
Bolt Loading Assessment
–
Not entirely sure what loads your joint will be subjected to? If
classical calculations are not enough we use finite element analysis to
calculate the loads to which a bolt is subjected. Analysis methods
available to us are inertia load assessment, dynamic load assessment,
random vibration assessment, thermal assessment and fatigue assessment.
Bolt Lab Test
– Send us
your bolt and in addition to the bolted joint design check report we’ll
measure the torque tension relationship for your joint using load
cells.
Bolt Value Engineering
–
We help clients replace expensive custom made bolts with cheaper
alternatives with no loss in structural performance. Our last value
engineering exercise save one of our clients over 1 million dollars.
On Site Bolt Testing
– If
your joint is too big to send to us we’ll come to you. It may be
appropriate to use pre-calibrated strain gauged bolts which we can
specially prepare for you. Our last test involved monitoring bolts on a
railway vehicle. Our measurements were linked to Google Earth via a GPS
to that we could identify parts of the railway infrastructure which
gave rise to high bolt loads.
Bolt Failure Investigations
– Ask our team for an independent failure investigation. Our combined
bolting experience and analysis tools and skills places you in the
strongest position to investigate the root cause of bolt failure. Given
the nature of bolt failures this service is offered on an independent
and confidential basis.
This is a pretty simple bolt calculation from first principles. For a
given bolt size it looks up the characteristics of a bolt from a 'Bolt
Lookup Table'. A bolt preload (or clamping load) is...
Showcase 
This crashworhiness assessment considered re-engineering the front end of a leading vehicle in a train to improve vehicle crash performance. The study provides an interesting comparison between a vehicle with no provision for crashworthiness (unmodified) and one with a 2MJ collapse structure on the leading vehicle (modified). Which train would you rather be in? The top one or the bottom one? Now play the video and see for yourself the benefits of the designing for crashworthiness. The article outlines the calculations for the train end crash structure. It considers kinetic energy and its conversion to strain energy and compares resulting forces and deceleration rates of the two trains.
Input: Drawings supplied in PDF format by email.
Deadline: Three weeks after purchase order.
Output: Deveop strategy with the client assess a number of impact scenarios and prepare final calculation report suitable for third party scrutiny.
ASCE705W Wind Loading 1.0
Bolted joints are one of the most common elements in construction and machine design yet they are the root cause of half of the failures investigations undertaken by MoreVision. Every engineer experiences some difficulties with bolts at some point in his career. So what typically goes wrong and what can we do to prevent bolt problems? Follow our bolt failure checklist and find calculations to help you.
SimpleBoltCalc.xls
Bolt spacing and edge distances.xls
Ultimate shear capacity of bolt group.xls
