We are involved in providing our clients with premium quality deflection of beams apparatus, advanced beam testing apparatus and torsion of bars for checking strength of the materials. All the apparatus are highly capable.

strength of material lab, civil engg. lab equipments, mechanical engg lab equipments trader, our new product, advanced technocracy inc.

**Features**

- Rigid base and supports
- Choice of end conditions

a) knife edge

b) built-in - Beams or cantilevers
- Deflection and slope measurable
- Three year warranty

**Range of Experiments**

- Verification of beam deflection formula
- Deflection and slope of beams and cantilevers
- Verification of both area - moment theorems

The bench mounted apparatus has a heavy steel base with a fixed support at one end and a moveable support at the other. The supports can be fitted with knife edges or clamp plates one of which permits horizontal movement for an encastre beam. A steel beam and two load hangers are supplied together with two dial gauges for measuring beam deflections and slopes.

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, and all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught. A complete instruction manual is provided describing the apparatus, its application, experimental procedure and typical test results.

**Features**

- Rigid base and supports
- Choice of end conditions

a) knife edge

b) built-in - Beams or cantilevers
- Deflection and slope measurable
- Three year warranty

**Range of Experiments**

- Verification of beam deflection formula
- Deflection and slope of beams and cantilevers
- Verification of both area - moment theorems

The bench mounted apparatus has a steel base with a fixed support at one end and a moveable support at the other. The supports can be fitted with knife edges or clamp plates. A steel beam and two load hangers are supplied together with two dial gauges for measuring beam deflections and slopes.

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, and all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught. A complete instruction manual is provided describing the apparatus, its application, experimental procedure and typical test results.

- Cost Effective Teaching
- Comprehensive theory of beams
- Simple and propped cantilevers
- Simply supported, fixed and continuous beams
- Three piers measure positive and

negative reactions - Piers include a re-leveling system
- Three dial gauges on stands
- Point loads and distributed loading
- Six test beams to verify all variables
- Two optional extra sets of selected beams
- Data logging option
- Three year warranty

**Range of Experiments**

- All variables in deflection of beams
- Slope and curvature of beams
- Support reactions of single span and continuous beams
- Effect of sinking supports
- Area moment theorems
- Super-position
- Clerk Maxwell's reciprocal theorum
- Flitched beams
- Non-uniform beams

The apparatus provided allows an unlimited range of beam experiments to be performed to measure support reactions and the deflections and rotations of simply supported, fixed and two span continuous beams. The end clamp also offers work on simple and propped cantilevers. In addition the effect of sinking supports on a continuous beam can be studied.

The experiments are assembled on a bench mounted twin beam base standing on end frames with levelling feet. Three load measuring piers with a digital read out in decaNewtons can be clamped to the base anywhere within its length of 1.2 m. These piers are equipped with a height correction system to compensate for the vertical deflection of the load indicator and are fitted with beam connectors which provide pinned conditions for both downward and upward beam reactions. A fourth pier is a simple clamp for supporting a cantilever or the fixed end of a beam.

Three dial gauges on stands can be clamped anywhere on the base. Four load hangers provide for point loads, while a set of slotted weights can be used to simulate a distributed load on a beam. The set of test beams affords the study of all the variables in the standard formula for uniform beams.

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, and all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught. A complete instruction manual is provided describing the apparatus, its application, experimental procedure and typical test results.

**Torsion Of Bars**

Advanced Technocracy Inc. is Manufacturer, Exporter & Supplier of Torsion Of Bars.

Specification

[1] elastic torsion of bars

[2] 2 movable bearing blocks with clamping chuck for mounting of bars, 1 fixed and 1 movable bearing

[3] 2 movable angle indicators clampable to the bar

[4] 4 bars: round bar with full cross-section, tube, longitudinally slotted tube, square tube

[5] application of load to the bar by a mass disc, a deflection roller and a set of weights

[6] storage system to house the components

[7] experimental set-up in frame SE 112

Technical Data

4 brass bars, l=695mm

- round bar, d=6mm

- tube, slotted tube

d=6mm, wall thickness: 1mm, slot width: 0,3mm

- square tube

WxH: 6mm, wall thickness: 1mm

Disc to apply the load

- effective radius: 110mm

Angle indicator

- measuring range: -90°...+90°, graduations: 1°

Weights

- 1x 1N (hanger)

- 4x 1N

- 3x 5N

**Features**

- Low cost, effective teaching
- Self-contained
- Bench-mounted
- Combined bending and tension
- Three eccentricities
- Three year warranty

**Range of Experiments**

- To measure the vertical bending deflection of the bar and to compare with theoretical predictions.
- To assess the effect of eccentricity of loading.

Sometimes in the design of a structure, a tension member has to be offset from the line of action of the force. The member then has to carry combined tension and bending loads, the latter increasing with the eccentricity of the load. The eccentricity is exaggerated to make visual appreciation of the phenomenon possible. When the load line is outside the middle third of a square tie bar, as in this experiment, the bending moment predominates and the bending deflection may be considerable.

The apparatus enables both the load and eccentricity to be varied. A 9mm square section by 800mm long specimen is provided, together with dial gauge and load hanger. Different shaped specimens can be manufactured in the college workshop as required.

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Simple determination of Young's modulus
- Verification of hooke's law
- Range of specimen material and thickness available
- 3 year warranty

Range of Experiments

- To determine Young's modulus of elasticity for the specimen wire
- To verify Hooke's Law

Description

Loaded wires form a simple experiment which produces excellent and easy to understand results. A single wire can be used to determine Young's Modulus of Elasticity for the material, and to confirm Hooke's Law.

Two brackets are secured to a wall minimum 2m apart in a vertical line; a top bracket from which to hang a specimen wire, and a slider bracket used to measure the extension of the wire. The slider includes a vernier for accurate measurement. For safety, the lower bracket should be reasonably close to the ground.

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Simple determination of Young's modulus
- Verification of Hooke's Law
- Range of specimen material and thickness available
- Investigation of stresses in compound

suspension - 3 year warranty

Range of Experiments

- To determine Young's modulus of elasticity for the specimen wire
- To verify Hooke's Law
- To evaluate the equivalent modulus of elasticity for the combined wire suspension
- To determine the load in the wire under conditions of equal strain in each wire.
- To compare experimental and theoretical results

Description

Loaded wires form a simple experiment which produces excellent and easy to understand results. A single wire can be used to determine Young's Modulus of Elasticity for the material, and to confirm Hooke's Law. With two wires, the experiment can be widened to investigate the effective characteristics of two different materials subjected to a common strain.

Two parallel sets of brackets are secured to a wall minium 2m apart in a vertical line; a top bracket from which to hang a specimen wire, and a slider bracket used to measure the extension of the wire. The slider includes a vernier for accurate measurement. For safety, the lower bracket should be reasonably close to the ground. The lower ends of each slide are connected by a link. A load hanger can be moved along the link until strains are equalised in each wire. The wires may be of different materials, but must be the same length.

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Demonstrates Hooke's Law
- Measurement of spring stiffness
- 3 year warranty

Range of Experiments

- To test the relationship between the load applied and the change in length of a spring (Hooke's Law)
- To determine spring stiffness
- For more advanced courses, the dependence of spring stiffness on the wire diameter, spring diameter, length, number of turns and material. Comparison with theoretical estimate.

Description

Springs are used in engineering to store energy or to provide restoring forces. Both compression and tension springs may be encountered. The deflection of a spring depends on the load applied to it, an observation enshrined in Hooke's Law. Applications of springs are found in spring balances which indicate loads by measuring spring deflections and in car suspensions where they absorb energy caused by wheel vertical movement due to potholes and bumps.

The equipment is designed to be fitted to a wall. It is used to test tension springs up to 200mm in length. The maximum spring diameter is 38mm.

A weight hanger is used to apply a load to the spring. Spring deflection is measured with a sliding scale which can be easily re-zeroed to suit the length of the spring. A spring, weight hanger and weights are supplied with each piece of equipment.

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Demonstrates Hooke's Law
- Measurement of spring stiffness
- 3 year warranty

Range of Experiments

- To test the relationship between the load applied and the change in length of a spring (Hooke's Law)
- To determine spring stiffness
- For more advanced courses, the dependence of spring stiffness on the wire diameter, spring diameter, length, number of turns and material. Comparison with theoretical estimate.

Description

Springs are used in engineering to store energy or to provide restoring forces. Both compression and tension springs may be encountered. The deflection of a spring depends on the load applied to it, an observation enshrined in Hooke's Law. Applications of springs are found in spring balances which indicate loads by measuring spring deflections and in car suspensions where they absorb energy caused by wheel vertical movement due to potholes and bumps.

The equipment is designed to be fitted to a wall. It can use compression springs up to 150mm long. The maximum spring diameter is 38mm.

A weight hanger is used to apply a load to the spring. Spring deflection is measured with a sliding scale which can be easily re-zeroed to suit the length of the spring. A spring, weight hanger and weights are supplied with each piece of equipment.

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Simulates strains for a bolt stressing

a tube - Determination of stiffness of tension and compression springs
- 3 year warranty

Range of Experiments

- To determine the stiffness of springs in tension and compression
- To investigate a self-straining system similar to a bolt in a tube. In particular to measure the reduction in length of the "tube", and the forces in the springs
- To measure the increase in length and forces in the system due to applying an external tensile load

Description

The apparatus is a self straining system analogous to a bolt stressing a tube, enabling the final overall deflection of the system to be determined. It consists of a frame in which there are two springs. A tension spring with means of adjusting its length has a disc at its lower end. Between this disc and the top of the frame is fitted a compression spring. A weight hanger attached to the disc enables the two springs to be loaded. The stiffness of the two springs is different so that an overall deflection is induced.

The two springs can be installed separately in the frame so that the stiffness of each can be determined. In the case of the compression spring, it is necessary to apply loads through a cord and pulley arrangement. A graduated scale alongside the disc shows the deflection of the spring(s). Excellent results are achieved due to the low friction of the equipment.

Features

- Universal machine
- Compact, bench mounted
- Four specimens supplied: Circular ring,

Semi-circle, Quadrant and Davit - Measurement of oscillation frequency
- Measurement of horizontal and vertical deflections by dial gauges
- Demonstrates strain energy concepts
- Three year warranty

Range of Experiments

- To experimentally determine the vertical and horizontal deflections of various curved bars whose cross sectional dimensions are small compared with the bar radius.
- To compare with theoretical estimates using strain energy theories such as Castigliano's first theorem.

Description

The theoretical deflections of curved shapes are most easily found by applying strain energy ideas, such as Castigliano's first theorem. The shapes chosen provide a relatively easy introduction to the use of such techniques, which students often seem to find difficult to grasp.

A bench mounted base supports a curved bar formed into a ring, semi-circle or quadrant/davit. Loads are applied by specially designed weight hangers so that the specimen bends. Horizontal and vertical deflections are measured by dial gauges rigidly attached to the base. The bars can be readily changed and the position of the dial gauges relocated to measure the deflections of the new configuration. Bars, weight hangers and a set of weights are supplied.

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, and all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught.

A complete instruction manual is provided describing the apparatus, its application, experimental procedure and typical test results.

Features

- Low cost effective teaching
- Self-contained
- Bench-mounted
- Range of specimen materials
- Introduction to theories of failure
- Bending and torsional loading ratios

variable - 3 year warranty

Range of Experiments

- To determine elastic failure of a specimen subjected to several ratios of bending and torsion simultaneously
- To compare the results with the established theories of failure

Description

Much of the design of parts in mechanical and civil engineering is complicated by there being biaxial or triaxial stresses for which some failure state has to be determined. Obvious examples are high pressure cylinders containing liquids or gases and concrete hinges for large bridge bearings. For more than a century, physicists, mathematicians and engineers have been proposing various theories of failure. Some theories have been attempts to explain observed failures while a few have tried to base a mechanism on fundamental properties of materials.

It is evident that there is a considerable difference between the behavior of ductile and brittle materials. That apart, it is quite difficult to determine failure with sufficient accuracy in experiments designed to show which failure theory is most applicable. Hence, it is frequently found that codes of practice lay down what appears to be a somewhat empirical design method which experience has proved to be workable.

This simple machine uses inexpensive test specimens made from round bar. The specimen is clamped at one end to the base bracket and at the other to a counterbalanced circular loading plate. This plate is graduated in 15° intervals. A special hanger enables pure bending, pure torque or combined loads to be applied depending on the position of the plate. The specimen deflection is measured by a dial gauge mounted diametrically opposite the load point. In the event of a specimen failure safety is ensured by set screws

Features

- Low cost effective teaching
- Self-contained
- Wall-mounted
- Seven mild steel struts supplied
- Extra strut available, eccentrically loaded
- Tests pivoted or built-in ends
- Longitudinal and lateral loading
- Comparison with theoretical predictions
- 3 year warranty

Range of Experiments

- Determination of Young's modulus of Elasticity for specimen material
- Struts with pivoted ends, but varying lengths
a) to assess the effect of slenderness ratio on crippling load for the same specimen material

b) to compare with Euler and Perry-Robertson formula predictions - Struts of same length, but different end fixings
a) to assess the effect of end constraint on crippling load

b) to compare with Euler and Perry-Robertson formula predictions

c) to observe the shape of each critically loaded strut - Slender strut with eccentric loading (optional accessory)
a) to investigate how the lateral deflection of an eccentrically loaded strut varies with the applied load and eccentricity and to produce a Southwell plot.

b) to compare the experimental and theoretical values for maximum lateral deflection

Description

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught.

A piece of material in compression is called a strut. If it is short and stubby it will fail by compressive stress, but if it is slender the failure mode is that of buckling. The load at which the strut buckles depends on the way in which the ends are restrained. Built-in ends resist buckling more than ends which are free to move. The apparatus shows how the buckling mechanism occurs, and the influence of the end restraint.

The apparatus is rigid and wall mounted. It can test struts between 0.75 m and 1 m in length with either pivoted or built-in ends. Axial load is applied to a load hanger linked by cables to the yoked ram whose travel can be pre-set to prevent permanent damage to the strut. A lateral load can be applied at any position to the strut. Seven mild steel specimens are supplied as standard. A dial gauge is supplied to measure strut deflection.

With this equipment, an in depth study can be made of the factors that effect the buckling of a strut; its length, cross section, material and end restraint.

Young's modulus for the strut material is derived in a secondary experiment, using the same equipment but with a specimen loaded as a beam.

Features

- Low cost, effective teaching
- Self-contained
- Bench-mounted
- Demonstration of shape of a deflected

strut - Direct loading gives highly visual impact

of Euler theory - All possible end constraints
- Comparison with theoretical predictions
- Three year warranty

Range of Experiments

- To observe the behaviour of four struts of the same length but with different end constraints when subjected to buckling loads.
- To compare the result with theoretical predictions, such as Euler's formula.

A piece of material in compression is called a strut. If it is short and stubby it will fail by compressive stress, but if it is slender the failure mode is that of buckling. The load at which the strut buckles depends on the way in which the ends are restrained. Built-in ends resist buckling more than ends which are free to move. The apparatus shows how the buckling mechanism occurs, and the influence of the end restraint.

The apparatus comprises a sheet metal frame which supports four slender spring steel struts having loading platforms at their top ends. Each strut has a different end constraint so that comparisons can be instantly made in a highly visible way.

a) Both ends pinned

b) One end pinned, the other end fixed

c) Both ends fixed

d) Base fixed, top free

For the first three, the ends move inwards as the strut buckles. The loading platforms act through relatively friction free guide bushes.

Features

- Low cost, effective teaching
- Self-contained
- Wall-mounted
- Measurement of torsional stiffness
- Demonstration of Hooke's law for

torsional spring - Comparison with theoretical predictions
- Three year warranty

Range of Experiments

- To compare the experimental stiffness of a plane spiral spring with theoretical predictions.
- To observe if the spring exhibits a linear elastic behavior.

Spiral springs are used to provide a resisting or restoring torque to a shaft when it is rotated through an angular displacement. They exhibit similar stiffness characteristics to linear springs, except that the effect is one of torque rather than force. The stiffness of a spiral spring depends on its physical dimensions and the rigidity of the steel strip from which it is formed. The student can easily calculate the theoretical stiffness of the spring, and compare the value with simple experimental results.

The wall mounted unit consists of a spiral spring coiled from a length of 25 x 0.6mm steel strip to give an effective length of 2 metres, attached to a shaft mounted in ball bearings. A cord carrying a weight hanger is wound round the shaft, and a load applied to twist the spring. Spring deflection is measured with an attached 360° scale. A cord and weight hanger are supplied.

Features

- Cost, effective
- Self-contained
- Calibration of strain gauges to
- Determination of gauge factor
- Introduction to calibration and standards
- Introduction to probability of production

errors of batch made strain gauges - Three year warranty

Range of Experiments

- To study the application of structural theory in strain gauge calibration
- To asses the accuracy of calibration techniques
- To introduce the application of probability theory in production quality control

Based on BSI Draft for development 6:1972 this gauge factor test rig is a precision item specially designed for measuring the gauge factor of an electrical resistance strain gauge. It also demonstrates how structural theory is used to determine the strain on the surface of a test bar for calibration purposes.

The apparatus is based on a four point loading system which produces circular bending in the center section of a precision ground steel beam. A device for measuring the curvature over a length of 300mm has been calibrated to give direct readings of strain up to 1000 micro stain.

For demonstration purposes a pair of electrical resistance gauges have been bonded to the beam, but for calibration work users will bond their own gauges in accordance with DD6/1972.

An extension from the normal technical experiment is to introduce students to probability theory to assess likely differences in gauge factor due to batch manufacture.

Features

- Low cost effective teaching
- Bench mounted
- Self contained
- Wheatstone bridge and temperature compensation dummy gauge included
- Introduction to strain gauges
- Bending and Torsion included
- Optional extras for Tension and

Compression - Three year warranty

Range of Experiments

- To show the application of strain gauges in the measurement of stress, due to bending and torsion
- To demonstrate the use of a Wheatstone Bridge in measuring change of resistance.
- With the optional extras to show other methods of temperature compensation in conjunction with tension and compression specimens.

Description

The apparatus has been designed to illustrate the basic features of electrical resistance strain gauges and their application to measurement of strain and the derivation of stress levels, in bending, torsion, tension and compression.

An alloy cantilever has a single gauge bonded onto its surface, and an identical gauge is fixed to an unstressed piece of the same material for temperature compensation. The two gauges form part of a Wheatstone Bridge which has an apex or balancing potentiometer, and whose meter is calibrated directly in microstrains. The cantilever is loaded by weights hung from its free end, a weight hanger is included.

To extend the scope of the apparatus the cantilever can be replaced by a torsion bar having two gauges bonded orthogonally at 45º

For a complete study of strain gauging two optional extra accessories demonstrate averaging techniques for tension and compression specimens.

This equipment is part of a range designed to both demonstrate and experimentally confirm basic engineering principles. Great care has been given to each item so as to provide wide experimental scope without unduly complicating or compromising the design. Each piece of apparatus is self-contained and compact. Setting up time is minimal, and all measurements are made with the simplest possible instrumentation, so that the student involvement is purely with the engineering principles being taught.

A complete instruction manual is provided describing the apparatus, its application, experimental procedure and typical test results.

Features

- Self contained
- Bench mounted
- Fully guarded
- Digital readout of revolutions to failure
- Motor stops when specimen fails
- Necked specimens, steel, aluminium alloy

and brass - Ideal introduction to fatigue
- Optional extra for alternating bending

fatigue - Three year warranty

Range of Experiments

- To make an introductory study of fatigue using a Wohler rotating fatigue apparatus, including the time to failure caused by various stress levels and materials
- The accessory, HSM19X affords bending fatigue of a cantilevered strip of metal or plastic in modes varying from alternating to fluctuating stresses

Description

This machine has been designed to introduce students to the effects of fatigue. A simple cantilever specimen rotates at about 5700 or 1425 revs/min, inducing a sinusoidal variation of bending stress. At the faster speed, a third of a million stress reversals occur each hour, so failure should occur within a day. Failure can be hastened by using a specimen with a stress raiser.

The loading system cancels its own self weight enabling any desired value of bending stress to be applied, ten mild steel specimens are supplied. Axiality has been ensured, and care has been taken to reduce the effects of vibration. When failure occurs, a microswitch stops the motor and the cycles to failure are registered on a 5 digit revolution counter.

All rotating parts are shielded and a safety guard is provided to restrain the broken specimen. The apparatus is mounted on a heavy steel base plate and is designed to overhang the bench or pedestal on which it is placed. Ideally a heavy pedestal (eg concrete), isolated from the floor by rubber matting, should be used to minimise shock loads.

An additional accessory for alternating bending fatigue and additional specimens are available.

Features

- Self-contained
- Bench-mounted
- Fully guarded
- Digital readout of revolutions to failure
- Determination of gauge factor
- Motor stops when specimen fails
- Specimens from strips of plastics or metals
- Special setting details supplied
- Optional extra for rotating fatigue
- Three year warranty

Range of Experiments

- Bending fatigue of a cantilevered strip of metal or plastic in modes varying from alternating to fluctuating stresses
- The accessory HSM20X allows an introductory study of fatigue using a Wohler rotating fatigue test, including time to failure caused by various stress levels and materials

To extend the range of fatigue testing to strips of plastic or metal, this variant of the popular rotating fatigue machine HSM19 has been developed. Using the drive mechanism and base plate of the new HSM19 Mk.3 with a heavy steel portal straddling the width of the base an alternating displacement can be imposed on the free end of a cantilever. The frequency of the reciprocating force is around 24Hz for plastics or 100Hz for metals.

A rotating faceplate carries an adjustable eccentric bearing driving a connecting rod attached to the cantilever. The bridge to which this test piece is clamped can be adjusted vertically so that the imposed displacement can be varied. A counter with a 50:1 reduction gear is driven by the electric motor, offering a 1:100 or 1:25 count depending on the drive ratio to the faceplate. Microswitches detect failure of the specimen and stop the motor.

To test a specimen a special dial gauge enables a calculated deflection to be set for the actual maximum bending stress of the specimen. An instruction manual containing a set of nomograms is provided to assist the user.

Great care has been taken to minimise extraneous vibration. All moving parts are shielded within a protective cover which can be removed during setting up. A guard surrounds the connecting rod to prevent damage when the specimen breaks.

It is possible to add extra parts (HSM20X) to this machine so that the rotating fatigue test can be carried out as an alternative.

**Specification**[1] investigation of the deflection and strain of a thin disc under internal pressure [2] strain gauge configured as half-bridge[3] strain gauges measure the radial and hoop strain[4] dial gauge with scale, adjustable mounting[5] measurement of deflection possible at any radius[6] hydraulic cylinder with hydraulic pump to generate pressure[7] hermetically sealed hydraulic system, maintenance-free**Technical Data** Aluminium disc- outer diameter: d=230mm- diameter used in experiment: d=200mm- thickness: 3mmStrain gauge application- 8 strain gauges: half-bridges, 350 Ohm- gauge factor: 2,00 +/-1%- supply voltage: 10VDial gauge- 0...20mm, graduations: 0,01mmPressure gauge- 0...1bar, accuracy: class 1.0System pressure: max. 0,6bar

Advanced Technocracy Inc. is Manufacturer, Exporter & Supplier of Methods to Determine the Elastic Line

Specification

[1] comparison of different methods to determine the elastic line

[2] statically determinate or indeterminate beam

[3] 2 supports with clamp fixing, optionally as articulated support with measurement of angle of inclination or clamp fixing

[4] articulated support with force measurement dial gauge

[5] device to generate a bending moment

[6] dial gauge with generation of moment to measure the angle of inclination

[7] dial gauge to record the deformations of the beam

[8] weights to subject the beam to point loads or moment

[9] weights to determine the clamping moments on the supports with clamp fixings

[10] storage system to house the components

[11] experimental set-up in frame .

**Vibrations on Machine FoundationsSpecification**[1] representing and investigating vibrations on machine foundations[2] vibration generator generates vibrations by imbalance[3] vibration-free operation due to additional vibration isolation of foundation[4] 2 brushless high power servo motors as the “machine” drive[5] eccentricity, rotation frequency, direction of rotation, phasing and frequency ratio adjustable[6] variable arrangement of vibration absorbers for absorption of vibrations[7] vibration measurement using acceleration sensors[8] inductive position sensor records the eccentricity of the imbalance masses Experiments- familiarisation with vibrations occurring on machine foundations- tuning foundations for various excitation forces- investigation of vibration absorbers- investigation of the influence of additional damping- difference between metal springs and rubber springs- measuring and evaluating vibrations- measuring operating vibration forms

Corporate Information

Atico Export

Mr. Kunal Chopra
(CEO)

Atico House 5309, Grain Market, Near B.D School

Ambala - 133001

Haryana, India

Atico House 5309, Grain Market, Near B.D School

Ambala - 133001

Haryana, India

© Atico Export. All Rights Reserved (Terms of Use)

Developed and Managed by IndiaMART InterMESH Limited

Developed and Managed by IndiaMART InterMESH Limited