Speeding up Non-Destructive Testing for Rubber Hardness Measurements

Hardness measurement is a way to practically determine the degree of vulcanization of a rubber sample. One of the most important and commonly used measurements for non-destructive hardness testing is the IRHD (International Rubber Hardness Degree).

International testing standards specify a primary load of 5 seconds and then a secondary load of 30 seconds. H W Wallace has developed a predictive technique which allows these times to be reduced significantly.

Using a Wallace Cogenix Micro Hardness Tester, a program of research was undertaken on different rubber types (and thicknesses) at a variety of temperatures.

These tests were conducted to determine the relationship between indentation (IRHD) and time. A general equation was derived which meant that the 30-second hardness value could be calculated after a few seconds of the application of the “secondary” load. The effects of the “primary” load time were also examined.

The predictive technique has been incorporated into the newly launched Data Collection Software, giving an accurate, rapid prediction of the standard results, with a high level of confidence.


Hardness measurement has been in existence since the early 1900s and is one of the most commonly used methods of product and quality control. The standard, non-destructive, IRHD (International Rubber Hardness Degree) method (such as ISO 48:19941), holds the sample in place using an annular foot. A spherically tipped indentor is then applied for 5 seconds under a constant “primary” load.  A “secondary” load is then applied for another 30 seconds to give the indentor an incremental displacement. This is measured and then scale-converted to an IRHD reading.

The “secondary” load period was originally set at 60 seconds. However, it was felt that, especially when testing large numbers of samples, a reduction in this time would be preferable. Gurney2, and later Dock3, investigated the influence of the period of application of the load and, as a result of this work, the 30 second time was widely adopted.

It was noticed that a skilled operator could ‘predict’ the hardness of a sample much earlier in the measurement cycle. For this to be the case, there has to be a common family of hardness curves for a majority of sample types and ambient measurement conditions. This would mean that a general equation could be derived and this would in turn lead to a specific, predictable curve for any specified measurement. The curve could then be used to predict the 30-second value much earlier in the measurement cycle.

The standard specifies ambient temperatures to be 23 ± 2 °C. However, these conditions are not always observed in practice and so the effects of temperature on the prediction accuracy were also studied.


Both dead load and micro hardness instruments were used to perform tests on various samples in the hardness range of 30-95 IRHD. Results were recorded at 0.5 second intervals, producing 60 stored data points for each test. These were then logged on a computer. Standard load times of 5 and 30 seconds were used.

The standard1 specifies a temperature of 23 ± 2 °C. However, in order to determine what, if any, affect the temperature would make on predicting the end point of the sample, tests were conducted at varying temperatures. The sample and instrument were tested from 30 oC to 45 oC in 5 oC increments and also at 12 oC. The samples only were heated to 65 oC and tested on a room temperature instrument.

To determine the effect of sample thickness on the predictability of the endpoint of the sample, samples of thickness other than those specified in the standard1 (2 mm ± 0.5 mm for the micro test and 8-10 mm for the dead load test) were tested.

All data were plotted and then a random selection of measurements was chosen to get a variety of temperatures and samples. The data were normalized and superimposed to test the assumption of the similarity of the shape of the curves. Then, the normalized data were used to fit a general equation of the form:

Y = a + bt-c , where c is a constant (1)

Each sample will have a unique solution to equation (1). Two equations in the form of equation (1) can be created for each sample, using two t (time) values (t1 and t2) with their corresponding y (hardness) values (y1 and y2). These can be solved simultaneously to give unique values of a and b for a specific sample. A curve for that sample can then be created by substituting these values into equation (1).

The general equation was fitted at different t values to identify which two values of t gave the best fit for the majority of samples. Although it was preferable to fit as small a t value as possible, it was also important for the equation to fit as many samples as possible. This ensured a high degree of accuracy as well as the greatest possible reduction in time.  The equation and chosen t value were rechecked using a section of data.

For each sample, the equation and two y values at given t values, can be used to create a ‘predicted 30-second’ value. This was then compared, for each sample, to the actual measured value after the full 30 seconds during the same test on the same sample. This produced a measurement of prediction accuracy.

The indentation depth is measured from a datum which is provided by the primary load. The standard specifies that this should be 5 seconds. Investigations of this time were carried out.


It was possible to calculate a predicted 30-second value using the previously agreed t1 and t2 values with their corresponding y1 and y2 values because each sample was tested to the standard 30-second load time. This could be done after only 6 seconds (t2) of secondary loading time. This meant that the actual 30-second values and the predicted 30-second values could be compared. Out of the samples measured, 366/380 of the predicted values were in agreement with the actual 30-second values to within 0.5 IRHD. Samples within this agreement included samples from the complete hardness range (30-95 IRHD) and also those of non-standard thickness and those tested at varying temperatures. Of the 14 samples that did not agree within 0.5 IRHD, all agreed within 1 IRHD. These samples were not completely flat whilst being tested.

Table 1 lists the compounds that were tested and covers the full range of hardness tested, and includes the non-standard samples used. To ensure prediction repeatability, each sample was tested more than once.

Figure 1 shows a plot of a selection of samples of varying hardness. It shows their predicted 30-second value and their actual 30-second point. It is evident that there is a good agreement between the actual and predicted endpoints over the complete hardness range of 30-95 IRHD.

Figure 2 shows a typical graph. In this case, the sample is neoprene, and it shows the predictability of the indentation curve with respect to time. The graph illustrates the standard 30-second data (fig. 2 from 2.5 seconds for clarity of the curve shape) alongside a predicted curve, which starts at 6 seconds and ends with a predicted 30-second value. This graph enables a comparison to be drawn between the actual and predicted data for one specific data series. In this case, the endpoints are within 0.5 IRHD.

The effect of reducing the primary load time was investigated. It was found that there was no significant difference to the endpoint when the primary load time was reduced from the standard 5 seconds to 1 second.


The investigation as carried out on both Micro Hardness and Dead Load instruments. To ensure a repeatable result, many samples were tested. Table 1 lists the compounds that were tested.

In all cases, including the non-standard tests, the curve shape was predictable. By running the predictive software and the standard test to conclusion and then comparing the endpoints, the hardness prediction accuracy can be determined. Consistency is achieved in this way because, with only on test being carried out, only one position on the rubber is used. Therefore, the sample variability does not interfere with the results which allows a true comparison of the accuracy of the prediction.

This method enables a significant reduction in the cycle time of the test for situations such as in-house comparative testing, where it is not necessary to strictly adhere to the standard. A time reduction may benefit users for two reasons – more tests can be carried out in the same time or the same number of tests can be carried out in less time.

The prediction quality can be improved by the user when a new sample type is used. A test can be carried out to the full 30 seconds as well as running on test to the reduced period and producing a predicted endpoint. A comparison can therefore be drawn as to the reliability of the prediction. However, care must be taken because each new test is in a new position on the sample. Surface hardness usually exhibits variations of a few IRHD across the surface when testing rubber. Therefore, a predicted test carried out in a shorter time in order to predict a 30-second value will also be subject to this variation. Agreement to within the normal amount of fluctuation is both expected and practically observed.

The newly launched Wallace Data Collection Software incorporates the equation and mathematics used to fit the curve and predict the hardness value.


By fitting a general equation of the form given in equation (1), the standard 35-second IRHD test time (5 second primary load and 30 second secondary load time) for a rubber sample can be significantly reduced. For each sample, the equation can be solved, producing an accurate prediction of the ‘final’ value with a high level of confidence. Moreover, the standard 5-second primary load time can be reduced to 1 second.

The minimum test time is determined by the prediction accuracy required. The total test time can be reduced to 7 seconds (1 second primary time and 6 seconds secondary load time). This ensures an agreement between the predicted value and the actual value to within 0.5 IRHD. The total test time can be further reduced to 5 seconds for less accuracy, such as agreement within 1 IRHD between actual and predicted values. This time can be reduced further still but with decreasing accuracy between the predicted and actual endpoints.


  1. ISO 48: 1994, Physical Testing of Rubber, Method for determination of hardness
  2. H. P. Gurney, India Rubber Journal, p17-22, (1921).
  3. E.H. Dock, J. Rubber Res., Vol. 13, p19-22 (1944)

Table 1. Samples Tested

Sample Type Micro Dead Load
Chloro compound 3 3
CR compounds 3 3
EPDM compounds 3 3
EPM compounds 3  
FKM compounds 3  
H W Wallace calibration blocks 3 3
H W Wallace samples 3 3
IIR compounds 3  
NR compounds 3 3
NBR compounds 3 3
Nitrile/PVC compound 3  
Polyurethane compounds 3 3
Q compounds 3 3
Thiokol compound 3  


Graph demonstrating the agreement between the actual and predicted 30 second endpoints.

Figure 1. Graph demonstrating the agreement between the actual and predicted 30 second endpoints.

Graph showing data collected from a sample of neoprene, with its predicted curve.

Figure 2. Graph showing data collected from a sample of neoprene, with its predicted curve.


S. Lackovic Bsc, PhD, R. Morgans Bsc and D. Reynolds*

1. H W Wallace & Co. Ltd
172 St James’s Road

2. University of Greenwich
School of Engineering
Medway Campus
Chatham Maritime

Presented at a meeting of the
Rubber Division, American Chemical Society Cleveland, Ohio
October 21-24, 1997


This information has been sourced, reviewed and adapted from materials provided by Wallace Instruments.

For more information on this source, please visit Wallace Instruments.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Wallace Instruments. (2019, July 26). Speeding up Non-Destructive Testing for Rubber Hardness Measurements. AZoM. Retrieved on October 14, 2019 from https://www.azom.com/article.aspx?ArticleID=18193.

  • MLA

    Wallace Instruments. "Speeding up Non-Destructive Testing for Rubber Hardness Measurements". AZoM. 14 October 2019. <https://www.azom.com/article.aspx?ArticleID=18193>.

  • Chicago

    Wallace Instruments. "Speeding up Non-Destructive Testing for Rubber Hardness Measurements". AZoM. https://www.azom.com/article.aspx?ArticleID=18193. (accessed October 14, 2019).

  • Harvard

    Wallace Instruments. 2019. Speeding up Non-Destructive Testing for Rubber Hardness Measurements. AZoM, viewed 14 October 2019, https://www.azom.com/article.aspx?ArticleID=18193.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback