TC10

Mission Statement

Vision

The vision of TC 10 concerns optical measurements, mainly spectrophotometric measurements, in the UV, visible and infrared ranges for the determination, by integration of the spectral transmittance and reflectance (T&R) curves, of the ultraviolet, luminous, solar and thermal parameters of glazing for building or automotive applications.

Mission

The mission of TC 10 is to improve the interlaboratory reproducibility of:

- the spectral measurements using the most convenient measuring procedure (reference standard, type of integrating sphere, etc.) and by suitable calibration of commercial instruments;

- the integrated parameters obtained according to the procedures recommended by ISO and CEN standards.

Accuracy is also checked by comparing the results of TC10 members with those obtained by metrological institutes.

Click to return to TC10 index

Planned Activities

A new interlaboratory comparison on translucent glass and plastic sheets has started and the same set of samples is circulating among the TC10 members. The homogeneity of each sample was previously checked and the area that has to be analysed during the measruement was established. The set includes a PTFE sheet, a sand-blasted glass and a laminated glass with a diffusing plastic sheet inside. A link has been established with the partners of the European RE-VIS project who are also involved in the goniophotometric and photometric characterisiation of similar diffusing glazing.

TC10 is trying to improve the interlaboratory reproducilibity of T&R measurements on translucent samples, by establishing a procedure by which these results are obtained from extrapolation of the experimental data and become independent of the sample geometry and the instrumental conditions of measurements (sample beam daimeter, thickness and diameter of the sample port of the integrating sphere, etc.)

Click to return to TC10 index

Next planned meeting

To be held in the CSTB, Paris on 15th October 2004.

Click to return to TC10 index

Charles Anderson was selected as a new chairman for the committee due to the tragic death of the former chairman, Pietro Polato. Dr. Polato posthumously received the William E.S. Turner Award from the ICG in September of this year. Dr. Polato was honoured for his many outstanding achievements to the ICG committees and will be missed by all.

 The technical activities of the committee centred around the near completion of two topics that have been ongoing for the last several years. Papers concerning the topics were published in the journal Rivista della Stazione Sperimentale del Vetro. Two meetings were held during the year to discuss the final drafts of the papers and to consider new topics for future activities. Also, another topic concerning colour and energy calculations has been advanced.

 The first publication was titled “Intercomparison of normal emissivity measurements on coated flat glass”. The committee decided several years ago to investigate this topic for several reasons. Low emissivity products are becoming very effective (emissivities less than 5%) and many countries are requiring new buildings to include these products. However, classical dispersive infrared spectrometers which are capable of measuring emissivity according to existing standards are no longer available on the market and are being replaced by FTIR spectrometers. Early experience with FTIR spectrometers indicated measurement problems and in addition, the wavelength range of these instruments is not sufficient to satisfy the requirements of the existing standards. The round robins performed by the committee and documented in the publication compared the two types of instruments and provided the starting reference for the European funded project THERMES which will make recommendations on using FTIR spectrophotometers for emissivity measurements with improved accuracy. One final paper will possibly be drafted to suggest a procedure for the self-assessment of the repeatability of spectrophotometric measurements.

The second publication was titled “Spectrophotometric determination of visible and solar parameters of sand-blasted glass panes and translucent glass laminates”. This topic was investigated because manufactures of highly diffusing products found large discrepancies between visible and solar parameters when measured with different types of instruments. The paper documents the discrepancies and a photon tracing program based on Monte Carlo techniques gave insight into the reasons for the discrepancies. Unfortunately, the committee was not able to determine measurement recommendations for highly diffusing samples measured on different instrumentation. Some members will continue work on this issue, with the aim to improve accuracy.

 The topic concerning the validation of colour and energy calculations has been ongoing for the last few years. The purpose of the investigation is to verify that members derive the same colour and energy values given the same spectrum. Also, we would like to determine calculation differences due to other parameters such as using different wavelength intervals. This work will continue in 2004 with the idea of creating “reference” calculation spreadsheets which include the necessary tables and formulas and will be source documented (CIE, ISO and so on).

 Members were invited to submit new topics for future investigations. From the list of topics, two were chosen for investigation over the next few years. Determination of the precision of portable spectrophotometers that are used industrially was considered an important topic for investigation. A round robin will be organized and results will be compared to laboratory type instrumentation. We have also proposed to start a project that basically would provide a reference to correct methodology of transmission and reflection measurements over the solar range and that would eventually define a procedure for the self-assessment of measurement uncertainty.

 Finally, the committee decided that we would organize an archiving procedure for most significant activities dating back to 1989.

Click to return to TC10 index

MEETINGS in 2002: Murano-Venice on April 12 and September 23

The sad and unexpected death of the chairman of the TC, Pietro Polato, in October 2002 naturally disturbanced the activities of the TC. The second meeting in 2002 was chaired by Dr. Franco Geotti-Bianchini, a colleague of Dr. Polato. He and Mr. Roucour, the vice-chair of the TC, offered to act as an interim committee until a new chair has been found.

TC 10 has just completed two major activities:

• - Round robin test on emissivity measurements, especially dedicated to low emissivity coatings.
• - Measurement of the optical properties of diffusing glazings.

The reports of these activities were prepared by Dr. Polato and Dr. Geotti-Bianchini and will soon be published, in particular the subject on emissivity because the " THERMES" group is waiting for the conclusions.

Click to return to TC10 index

TC 10 has organised a Round Robin on emissivity measurements dedicated in particular to low emissivity coatings. Five samples of coated glass (covering the emissivity range from 0.03 to 0.4µm) were cut in specimens, tested in their reproducibility and distributed among the members of the Committee.
Results were received from 16 laboratories (21 infrared spectrophotometers on the whole) together with information on the instruments, accessories for the reflectance measurements and reference mirrors used.
The statistical elaboration was performed on the data measured by FTIR and dispersive instruments firstly by considering the two kinds of spectrophotometers separately. No significant differences in the mean emissivity values were obtained between FTIR and dispersive instruments. The FTIR instruments always had a larger standard deviation. For the FTIR instruments, the lower the emissivity, the lower the standard deviation (this was not always true for the dispersive instruments). By considering the instruments all together, an uncertainty on the emissivity measurements of 0.02 should be accepted.
Next year, for a complete statistical treatment of the data, each laboratory will evaluate the repeatibility of its instrument(s) by performing several measurements on each sample.

When commercial integrating sphere spectrophotometers are used for measurements on diffusing glass samples (sand blasted glass sheets, laminated glass panes with diffusing plastic sheets inside, etc.), a part of the transmitted or reflected radiation does not enter the small port of the integrating sphere.
Algorithms (e.g., ray-tracing based on Monte Carlo methods) could be used to predict the experimental values from the characteristics (surface roughness, size of the scattering particles, etc.) or diffusing properties of the samples and the geometry of the small integrating spheres (size of the beam incident on the sample, size and thickness of the entrance port, etc.).
The recorded curves show high discrepancies among measurements performed by different laboratories and the values of the integrated parameters (light transmittance, solar direct transmittance, etc.) are under-estimated. A procedure is under investigation in order to remarkably improve the interlaboratory reproducibility in the measurements and reduce the effect of the systematic errors on the final values of the luminous and solar parameters. This procedure is based on a photometric or spectral (only one wavelength) measurement on a large sample (e.g., with a size of 30x30 cm²) and a spectrophotometric measurement in the whole solar range on a small specimen (e.g., 5x5 cm²), the two specimens cut from the same diffusing glass sample.

The laboratories use their own software for computing the integrated parameters (UV transmittance, colour, light transmittance, solar factor, etc.) of flat glass starting from the transmittance and reflectance curves recorded in the solar range. Values computed by two labs might be different even if they use the same experimental results as input data. For this reason, a validation of calculation has been organised by TC10: spectral transmittance and reflectance curves of some selected glass sheets (a coated glass and a green glass) have been distributed among the members of the Committee which have to evaluate, for both the monolithic and double-glazed configurations:
- UV transmittance, Krochmann damage function, skin protection function;
- colour referred to illuminants A, C and D65 and observers 2° and 10°;
- colour rendering index (according to EN 410);
- solar parameters (solar direct transmittance, solar factor) according to different standards.

An activity planned for 2002 is the evaluation of uncertainty of spectral transmittance measurements in the visible range using commercial spectrophotometers. A procedure will be established in order to evaluate all possible components of the overall uncertainty: from random effects from human operators to metrological characteristics of the instrument used. A summarising table will be filled in by each TC 10 member and the uncertainty of his instrument will be evaluated starting from the reported data.

Meetings in 2001

Murano-Venice, March 23, 2001
Murano-Venice, September 28, 2001
 

Click to return to TC10 index

Measurements on diffusing glass samples (sand blasted glass sheets and laminated glass panes with diffusing plastic sheets inside) have been performed using integrating sphere spectrophotometers and large integrating spheres. Two specimens were prepared for each sample: a small one, 5x5 cm², for spectrophotometers and a large one, 30x30 cm², for large spheres. By comparing the results using entrance ports with variable diameters for both kinds of instruments the following conclusions were highlighted for the commercial spectrophotometers:
- the small integrating sphere looses a portion of the radiation transmitted (or reflected in the case of reflectance measurements) by the sample;
- a small size is required for specimens and, consequently, edge effects perturb the measurements.
Both effects depend on the scattering properties of the sample. Moreover, the lost radiation is a function of the orthogonal cross-area ratio between the light beam and the sample port of the integrating sphere. Finally, the thickness of the port should be as thin as possible.
The experimental values obtained by each TC 10 partner using his integrating sphere spectrophotometer have been compared to values calculated using a Monte Carlo method starting from the characteristic of the sample and the geometry of the sphere. A laminated glass pane containing a diffusing plastic sheet has been considered for this comparison. The discrepancies between experimental and calculated values decrease by implementing the algorithm (e.g., knowing diameter and refractive index of the scattering particles contained in the plastic sheet).

A new activity of TC 10, which started in 2000, consists of a validation of calculation. Different software is used by the laboratories for computing the integrated parameters (UV transmittance, colour, light transmittance, solar factor, etc.) starting from the experimental results. Therefore, the values computed by two labs may be different, even if they use the same experimental results as input data. For this reason, spectral transmittance and reflectance curves of some selected glass sheets (a coated glass and a green glass) have been distributed among the TC 10 members, which have to evaluate, for both the monolithic and double-glazed configurations:
- UV transmittance, Krochmann damage function, skin protection function;
- colour referred to illuminants A, C and D65 and observers 2° and 10°;
- colour rendering index (according to EN 410);
- solar parameters (solar direct transmittance, ..., solar factor) according to different standards.

Finally, a Round Robin on emissivity measurements is scheduled for 2001. Technical Committee 10 had already performed this kind of RR several years ago (results were published in 1988). Infrared dispersive spectrophotometers used at that time are now replaced by FTIR spectrophotometers. Moreover, new products are now on the market with emissivity values, which can be very low: a coating with two silver layers may have an emissivity of 0.03, or even lower. Errors in the measurements performed with infrared spectrophotometers may cause completely incorrect evaluations for such a low value (depending on the type of instrument, the specular accessory and the calibrated standard used). In order to check and, possibly, improve the interlaboratory reproducibility of measurements on low emissivity coatings among the TC 10 members it was decided to start this Round Robin. The samples have already been distributed.

Meetings in the last year

Grenoble, May 18, 2000
Paris, October 4, 2000

Next meeting:  Murano-Venice, March 23, 2001
 

Click to return to TC10 index

1999 Annual report

ACTIVITIES IN 1999 AND PLANS FOR 2000

Activities very much concentrate lately on diffusing glass samples. The first Round Robin on diffusing samples was completed in 1998, while during a later meeting of TC 10 it was decided to interrupt the second one. 'The second Round Robin (Transmittance measurements on sand blasted glass sheets) failed because of the bad inter-specimen reproducibility.
A new interlaboratory comparison has now started and this time the same set of samples circulates among the TC 10 members for transmittance measurements in the visible range using integrating sphere spectrophotometers.
The set includes a PTFE sheet, a sand-blasted glass and a laminated glass with a diffusing plastic sheet inside.
All samples have a square shape with 5 cm sides. The recorded data will be correlated with the orthogonal cross areas of the light beam and of the entrance port of the integrating sphere. By using a suitable mathematical model, theoretical values will be obtained for the same cross areas and compared with the experimental results.

Two specimens (a small one 5 x 5 cm² and a large one 30 x 30 cm²) prepared from the same diffusing laminated glass are used to perform transmittance measurements by changing the diameter of the entrance port of small and large integrating spheres. The experimental results will be interpolated by the mathematical model and the extrapolation to an infinitive port diameter will supply the “true” value.

As glass manufacturers often need to report in their brochures the light transmittance and solar factor of double- or multiple- glazed units, transmittance measurements have also to be performed using a large integrating sphere on a double-glazed unit, the composing glass sheets of which are the large laminated diffusing specimen and a non-diffusing glass sheet.

In 2000 TC 10 will start to prepare a paper summarising the results of the measurements on diffusing glass sheets, with recommendations on measurement procedures.

Next meeting: the next TC 10 meeting is scheduled in May 2000 in Grenoble.

Click to return to TC10 index

1998 Annual Report

ACTIVITIES IN 1998 AND PLANS FOR 1999

Most of the members of' TC 10 will participate in an industrial Round Robin at (near) normal and angular transmittance and reflectance measurements of coated glass of commercial production for automotive or building applications. This Round Robin is promoted by the partners of the European ADOPT project. The purpose of this activity is to improve the interlaboratory reproducibilty and accuracy of spectrophotometric measurements.

Click to return to TC10 index

1997 Annual Report

Meetings:

TC 10 had two meetings in 1997, one in Murano-Venice in February and the second one in Paris, in August.

Collaborative work:

The investigation on transmittance and reflectance (T&R) measurements in the UV range on uncoated and coated glass sheets of commercial production is finished: a paper has been sent to Glastechnische Berichte for publication.

A first round robin on T&R; measurements on diffusing samples for building applications (sand blasted glass, laminated glass containing PC or opal PMMA sheets) using integrating sphere spectrophotometers has shown high discrepancies among the participating laboratories, mainly due to the use of instruments with different irradiation and observation geometry (spectrophotometers, colorimeters, spectrocolorimeters). As small sizes are required for the sample to perform spectrophotometric measurements the results are also perturbed by border effects and depend on its geometry.

In the case of transmittance measurements on diffusing laminated glass the interlaboratory reproducibility is very poor if the results of each TC10 member are referred to the open port of the integrating sphere (standard with 100% transmittance). The reproducibility increases if one of the samples is used as a standard in the measurements on the other ones. Analogously the reflectance measurements should be referred to one of the laminated glass sheet used as a standard (after being calibrated by a metrological Institute) instead of using a different type of reference standard.

The results of the measurements on a sand blasted glass depends on the surface which is in contact with the entrance port of the integrating sphere. The higher values are obtained when the smooth side is against the port. Sand blasted glass samples with a good interspecimen reproducibility are difficult to be prepared. The cleaning procedure has a strong influence on the measurements.

Click to return to TC10 index

Recent Publications

F Nicoletti, F Geotti-Bianchini, P Polato

Spectrophotometric determination of the normal emissivity of coated flat glass.

Glastech Ber 61 (1988), p127-139

F Nicoletti, P Polato, J Roucour

Interlaboratory comparison of solar range transmittance and reflectance of coated and

reflectance of coated and uncoated flat glass.

Glastech Ber Glass Sci Technol 67 (1994) p31-44

P Polato, J Roucour, J Simons

Interlaboratory comparison of ultraviolet transmittance and reflectance of coated and

uncoated flat glass.

Glastech Ber Glass Sci Technol 71 (1998) p129-140

P Polato, A Danco, J Roucour, A Conti

Glazing solar parameters evaluated by ISO CEN and ASTM standard procedures.

Glastech Ber Glass Sci Technol 69 (1996) p131-139.

Click to return to the Technical Committee Index

Click to return to the main index