Load Testing Machines
Important Features to Consider
M S Bayliss B.Eng
Senior Engineer, IST
As
manufacture’s of high quality spring load testing machines, we are often aware
that clients do not appreciate the factors that should be considered when
purchasing a spring load testing machine. It
should be appreciated that springs require special consideration and that
general purpose materials testing machines are not usually suitable.
This article will describe the major features that are required to ensure
that accurate and repeatable results are obtained.
Looking at the main component parts of a machine in turn:
i) The Loading frame
The loading frame must be as rigid as possible, to minimise the frame
deflection. Frame deflection which
can cause displayed length errors, and in some machines allow the spring platens
to become non parallel. Electronics and software can compensate for frame
deflection, but the compensation will not be perfect.
For the smaller spring sizes <500N load, most loading frames are of the
‘C’ type. This enables easy
access to the platens for loading and unloading springs.
This ‘C’ frame is the least satisfactory due to the bending moments
that are applied to the linear guidance system. For loads > 500N a ‘H’
type frame using a pair of ballscrews should be used.
This system ensures that the frame deflects symmetrically enabling better
compensation. In both the ‘C’
and the ‘H’ type frames, the ballscrews should be preloaded to remove
backlash.
a substantial purpose built
guidance system using well spaced bearings of either bronze or recirculating
type should be used for the crosshead.
ii)
Loadcells
Most commercial Loadcells are produce for weighing applications and are not
suitable for spring testing due to their sensitivity to the additional forces
and torque’s produced by springs. For
smaller springs (<500N) the double shear beam loadcell will usually operate
satisfactorily. Ideally, a cell
deflection of less than 0.1mm under full load is required for good dynamic
response. For the larger springs
(>500N) the shear beam pancake type cell has been shown to offer the best
performance. The overload
protection of the loadcell is also important, as inevitably during use of the
machine the load cell will be subject to abuse.
The signals from the loadcell need to be amplified processed and displayed on a
computer system. These electronics must be stable with temperature and time.
Switchable amplifier ranges which give extended measuring capabilities
are important to enable maximum machine utilisation.
iii)
Length measurement
Digital machine tool linear transducers are now universally used for spring
testing machines. These transducers typically have a resolution of 0.005mm and
an accuracy of+/-0.003mm/m however when built into a machine the systems
accuracy will be degraded. The
offset between the platens and the transducer will amplify the effect of any
bearing play degrading the length accuracy.
a displayed resolution of 0.01mm
is sufficient in all but the smallest of machines when a resolution of 0.005mm
can be usefully utilised. For physically larger springs, the resolution can be
usefully increased to 0.1mm if required.
iv) Defining machine accuracy
To ensure correct operation of the machine the load and length transducers
have to be calibrated at least every 12 months to prove their accuracy and
repeatability. For compression and
extension springs machines must be at least grade 1.0 or better still grade 0.5
as defined by BS 1610, DIN 51232 and ISO 7500-1 Class 0.5.
The grade number means the reading is accurate to that percentage of the
reading. Machines are quoted with
accuracy’s expressed as a percentage of full scale, a false impression of
accuracy. To ensure that a small
spring is not measured on a large test machine, the standards also stipulate
that the machines can only be calibrated down to a lower limit set by the
displayed resolution. This ensures
that a small spring is not measured on a large test machine. The machine must be calibrated in both compression and
extension mode.
Length transducers are not covered by such a standard however BS 1726-1: 2002 does
stipulate an accuracy of +/-0.02mm. Experience
has shown this not to be a realistic value for all but the smallest of machines.
It is considered that a value of +/-0.03/300mm of deflection is a more
achievable value.
v) Software
The quality of the software supplied with a load testing machine is
difficult to assess as people’s likes and dislikes are very subjective. The
controlling software should have the following features as a minimum:
·
Measurement of loads at length
·
Measurement of lengths at load
·
Measurement / Calculation of spring rate
·
Prestressing
·
Tolerancing of any measured parameter
·
Printing of test certificates
·
Storing of test sequence and data for recall later
Further features worth considering:
·
Measurement of free length
·
Measurement of initial tension
·
Measurement of hysteresis
·
Load-length or load-deflection graphs
·
Interfacing of external measuring equipment
(e.g. micrometers) to the
software
·
Spring rate-length or spring rate-deflection graphs
·
Measurement of solid length
·
Statistical analysis e.g.: X bar, range and histograms
·
Testing speed adjustment
vi) Safety considerations
The safety aspects of spring testing machines are often not considered. The
two main areas of concern are trapping points within the moving parts and the
possible ejection of a spring. A
correctly designed, fully enclosed and interlocked guard will protect against
these dangers.
All machines must meet safety regulations, and by law in Europe, all machines
must comply with safety driven CE regulations.
