Introduction

Purchasing goods and services has never been an easy task. Typically, the higher the price tag of a product, the more carefully the shopper must study the pros and cons of features and specifications. That each supplier highlights its own advantages over its competitors makes the job that much more difficult. Selecting an electromechanical static universal testing machine (UTM) is definitely not the exception. Many industrialists may be amazed at the enormous number of options available. There are over 60 UTM manufacturers in the world, all claiming to offer the exact product you are looking for.

This article reviews some of the most relevant features to consider when selecting a UTM test system.

Loadframe capacity and dimensions

These specifications are truly vital to the UTM selection process. Load frame capacity is established on the basis of the maximum force required to break the material to be tested.

Dimensions - clearances between columns, as well as vertical clearance - must be adequate for the product being tested. Some materials, such as elastomers and soft plastics, will elongate substantially. A sufficient vertical travel allowance will permit the material to stretch as far as necessary without running out of space. Care should also be taken to accommodate special grips, fixtures, and environmental chambers that may require additional space in both directions.

Frame stiffness

Many times this feature is overrated. The stiffness of the test frame can be an important factor where there is reliance solely on evaluation of crosshead motion, rather than on a separate extensometer or deflection measuring device. The fact is that most applications in compliance with international standards call for the use of an extensometer or deflection measuring devices.

Factors to be considered in calculating frame stiffness are: screw diameter, ball nut fit, crosshead stiffness, screw bearing fit, and frame stiffness (Figure 1). Of course, the compliance of the specimen itself, the pull rods, and the specimen gripping devices are possible sources of error. Thus, using the load frame as a deflection measuring device presents a challenge.

Figure 1: Frame stiffness factors

There are some applications where crosshead displacement should be used to measure deflection, but those are few and far between. Why, then, spend money on a machine that has high frame stiffness if this will not be a factor in the testing process?

Some units with lower frame stiffness have, as standard equipment, a data channel and a programme for the direct measurement of strain or deflection. This provides a correction mechanism if in fact the load frame were to be used for deflection measurement. This constitutes a more economical solution for those few applications that do require the use of crosshead displacement to measure deflection.

Drive system specifications

These specifications (speed accuracy, position resolution, position accuracy, and repeatability) are important and will assure that the system complies with the relevant international standards.

Lateral motion

Generally more important than frame stiffness is lateral motion of the crosshead. This can be a serious source of errors because it introduces bending motion into the test specimen. The occurrence of bending, in a tensile test, will cause the specimen to fail at lower than normal forces.

Many companies control the lateral motion with the use of round bars for crosshead guidance. But these lack the stiffness to prevent the lateral motion that can generate problems (Figure 2).

Figure 2: Deflection being measured on a round bar

Some companies use a very stiff two-column frame with crosshead guidance roller bearings to minimise lateral motion of the crosshead. This guidance system also prevents the crosshead from twisting in the front-to-back direction. With the appropriate steel gauge, formed into two large channels, the stiffness of these channels is about 13 times the stiffness of ball screws or the typical round bar columns used in many testing machine frames. This is illustrated in the figures, which show a deflection of 7 thousands with the columns after applying 200 lb. weights, and 46 thousands with two 2” round bars, typically used as guides in most universal testing machines. Two bars were necessary to hold the 200 lb. weights; one bar would have been insufficient.

Figure 3: Deflection measured on a column frame

Maximum speed at full load

Basically, this is another flashy specification with limited practical value. Many companies will assert that their machines can go full speed at full force. The question suggests itself, why would anyone want to do that, particularly with a high-capacity machine (100kN and over)? Most steel standards call for testing at load speeds under 50mm/min. Plastic material speeds can vary, depending on the type of plastic, from 2 to 50mm/min. Elastomers do require high speeds, but they very seldom demand full load at those speeds. Some few applications (e.g. springs, urethane) demand full speed at full load, but their materials normally are worked at low capacities (100kN and under).

Control electronics

In an engineering environment, designers always run into “contradictory criteria”. The objective, on the one hand, is to design something as complete and comprehensive as possible; on the other, to do so without making the hardware so sophisticated that it becomes either a nightmare to repair and/or extremely expensive to do so. This is called “optimising a design”. Some manufacturers with a control console go on to acquire a computer, interface boards, and signal conditioners, depending on the accessories to be used. The fewer components a system has, the fewer the parts that can fail - and the lower the likelihood of something going wrong.

For basic testing without data acquisition, a machine with a console should be considered. For applications requiring data acquisition, statistic analysis, report generation, etc., a control console should be avoided. It will be much easier and less expensive to repair or replace a PC than a high-priced, specialised console with high-delivery lead times. If a PC is already in the system, the console is redundant

Recording data on speeds and bandwidth

Some testing machine manufacturers suggest it would be a good idea to record 5,000 data readings per second during a test. For a test ten seconds long, that rate would produce 50,000 data points. A table of those data points would require more than 80 printed pages!

How many readings need to be recorded to get all the important information from a test? ASTM E 1856 appendix X2 shows a maximum bandwidth requirement of 20/(event duration in seconds) and a sampling rate of 31 times the required bandwidth. So, for a ten-second test you will need bandwidth of 2Hz and a sample rate of 62 samples per second. If the test is run very fast (so that the sample breaks in one second), then you need 20Hz of bandwidth and a sample rate of 620 samples per second.

Some computerised testing systems have a bandwidth of 20Hz. This is enough to capture a one-second-long test with acceptable accuracy, but below the frequency of the power system to exclude noise from that source. If analog-to-digital converters are synchronous, there is no time skew between channels. A data-sampling rate adjustable from 1 per second to 1,000 per second allows collection of data at a rate appropriate to the test being performed.

The question arises, why do some companies offer such fast sampling rates (up to 5kHz) if most of the data collected will not be used? Obviously, the specification will effectively eliminate some competitors. Additionally, some companies take advantage of their dynamic UTM technology and apply it to the static UTMs, even though there may be very little practical value in doing so.

Force measurement system

This system calls for accuracy and repeatability. Accuracies of ±0.5% of reading to 1% of capacity, and repeatability of 0.25% of reading, will cover 95% of the applications. Self-identifying load cells can be convenient when multiple load cells are to be used on one system.

Automatic calibration

The push-button “automatic calibration” function touted by some makers of testing machines is a misrepresentation. It is really only a single-point check of the readout system. By definition, calibration requires that the device being calibrated be compared with a traceable standard source of what is being measured. Traceable standards can be weights, calibration rings, and load cells. “Auto-calibration” means that a button in the readout device allows for adjustment. It also creates a high risk of bumping the span knob, or mistakenly turning it, during use of the machine. This could potentially yield erroneous data that would be impossible to correct.

An alternative is to make the system very stable, allowing for no adjustments except those made by a qualified calibration technician using proper standards for comparison. Again, the more buttons to push, the more chance of something going wrong.

Strain measurement system

Most international standards such as ISO, ASTM, JIS, DIN, and BS call for similar specifications. Accuracy here should be 0.5µm; repeatability 0.25µm; and resolution 0.0004% of range. In selecting a strain measurement system, it should be verified that the specifications meet the corresponding standards.

Rockwell hardness tester series

The Tru-Blue series of Rockwell hardness testers from United Testing Systems Inc. meets or exceeds ASTM and ISO requirements and NIST recommendations. All testers in the series are computer controlled and feature the new NTEP-approved load cell, a depth sensing device, a high-resolution A/D converter, and a Mach Z CPU that allows hardness numbers to be displayed from one point down to one hundredth of a point. With over 18 years’ experience with load cell technology, the company has optimised the Tru-Blue series of Rockwell hardness testers for superb performance.

The Tru-Blue series of Rockwell hardness testers

The Tru-Blue series offers hardness testers for a wide variety of applications and requirements:

• Model R for standard Rockwell testing;
• Model II for Rockwell and Rockwell superficial testing of metals and other materials;
• Model II/36 for large parts such as camshafts, crankshafts, etc. This model features vertical capacity of up to 91mm with a motorised elevating assembly;
• Model U: a versatile tester that performs Rockwell, Rockwell superficial, Brinell, and Vickers tests;
• Model II/URF retrofit test head, which upgrades existing Wilson dial testers to digital computerised units.

Conclusion

The first step in an evaluation process is to carefully analyse machine features and specifications in light of the particular application. Weeding through sales pitches and “specsmanship” to identify the legitimate useful features and specifications is a daunting task, but an essential one. Many buyers take the easy way out by purchasing the best-known product, which very often is also the most expensive. The buyer ends up paying more than necessary. There is, of course, the opposite situation, with suppliers offering low-price options for the “same solution”. Here, the money saved initially may be lost in the long run due to poor quality and service, unreliability, excessive downtime, etc.

Most of the time, “low-end” units will have noticeably low-end features and specifications. Thus a good price/feature comparison table would be a useful purchasing tool. The key to making a useful comparison is to determine exactly what features are important for the particular application. When a supplier offers a machine that satisfies the specifications, at the best price, then a product with genuine value will have been found.

With so many options available in a highly competitive global business environment, “name buying” has become a thing of the past. The UTM selected must comply with the major applicable standards. Additional but unnecessary “special” features, however flashy, will only inflate the purchase price, to no useful purpose.