Fritted glass task group Spring 2017

Jacob C. Jonsson

Windows and Envelope Materials Group

Building Technology and Urban Systems Division

Name and scope change We are changing name from Fritted Glass TG to

Diffuse Glazing TG to better represent that we are also dealing with diffuse interlayers and diffuse applied films

Find a way to obtain spectral data for diffuse glazing products in a manner that would allow WINDOW calculations of IGUs. Determine best practice how to obtain data and verify the accuracy of these practices.

2

Outline for today’s session Presentation of progress regarding 270 mm

spheres with 100 mm apertures

Plan for future work

Presentation of ballot work regarding to 300-series documents (full discussion of ballots scheduled for the afternoon so I will try to breeze through what we did and spend more time on what we wanted to do)

Discussion based presentation, interrupt with questions and comments as you see fit

3

Determine best practice how to obtain data

First step was to find a set of measurements we can use as gold standard

Four instruments Guardian 270 mm sphere with 100/50 mm aperture LBNL pgII goniophotometer Fraunhofer ISE pgII goniophotometer Fraunhofer 630 mm sphere with wide area illumination

Viracon made a laminate for each of the 3 labs 12 mm OptiWhite (low iron)+0.030“ Saflex Arctic Snow

(transluscent white)PVB+3 mm Optiwhite float glass Guardian got all 3 samples to see how identical they were

4

Sample visualization

5

Incident on 3 mm

Measurement problem Light not entering the

integrating sphere is a systematic error

Sample dependent • Bulk/surface scattering • Thickness

Instrument dependent • Detector response vs

angle of incidence • Sphere geometry

6

Measurement problem Aperture

studies made by OMT is the fundamental test that informs this design

7

Different ways to measure transmittance using 150 mm sphere

Standard port

Front Side1 2 3 1 3

Front Side

Large port Small aperture

0 1 3Front Side

1. Sample, here a fritted glass with the fritted side towards the sphere

2. Standard Labsphere port plate, easy to remove

3. Part of the sphere

0. Aperture plate (not included by Labsphere) to reduce beam size

V175 transmittance - 6 results

Current pursuit 270 mm sphere with large aperture commercially

available 100 mm aperture reduces the side loss to reasonable

levels 3 in use in Europe that participated in this activity Have shown in specular ILC 2011 and 2015

Wide area illumination and other special spheres can also work

10

The journeys of the L3 laminate sample

NSG at Lathom, UK, Lambda1050 with upward looking 270mm integrating saphere (Perkin Elmer - OMT Solutions)

INSIMa (INstitut Interuniversitaire des Silicates, Sols et Matériaux), Mons, Belgium, Lambda 750 with same sphere

OMT Solutions, Modified 270 mm sphere

All three participants measured with 100 mm and 50 mm apertures

11

All data transmittance 3mm case

12

Harder case

Smaller spheres bad

270 spheres issues from NSG and INISma

Focus on ISE, OMT, and Guardian

Comparison large spheres

13

270 mm spheres less loss

14

Difference front and back

15

Focus on 270 mm

16

OMT analyzed the issues with high scattering samples and included a modification that significantly reduced

Samples with higher R?

Mimic the design of the Guardian sphere

17

Specular measurement The least expensive 270mm sphere does not allow

for split between diffuse and total signal

However, the common 150 mm sphere can be used to measure and calculate the specular component as the light not captured (Tloss) is equal in both measurements and additive

𝑇𝑠𝑠𝑠𝑠 = 𝑇𝑡𝑡𝑡 − 𝑇𝑑𝑑𝑑𝑑 =

= 𝑇𝑙𝑡𝑠𝑡 + 𝑇𝑑𝑑𝑑𝑑 + 𝑇𝑠𝑠𝑠𝑠 − 𝑇𝑙𝑡𝑠𝑡 + 𝑇𝑑𝑑𝑑𝑑

18

Reflectance The samples

were designed to have a lower reflectance than transmittance

Difference between front 3mm (solid) and 12 mm (dashed)

19

Qualitative images

20

Without consistent camera settings it is very optimistic to try to draw conclusions from these images, but this does support the idea that the thin side first makes it harder to measure reflectance

The scattering particles are in an index matched medium?

21

Riddle from last call why R was lower when the diffusor was closer to the sphere

Goniophotometer measurements Captures angle resolved broadband signal

Side loss has to be on the order of 10 inches

Not trivial to back out parallax but it is small enough to be ignored

22

Angle-resolved data from ISE-pgII

23

Visible value

Almost no difference seen based on position of the diffusing interlayer, i.e. front/back of the sample

0 15 30 45 60 75 900.0

0.2

0.4

0.6

0.8

Tdh,

Rdh

[-]

angle of incidence [°]

Tdh thin toward light, thin-air on axis Tdh thick toward light, thick-air on axis Rdh thin toward light, thin-air on axis Rdh thick toward light, thick-air on axis

Angle-resolved data from lbl-pgII

24

Visible value

Almost no difference seen based on position of the diffusing interlayer

Very low absorption at 10 degrees, solved

Fine tuning LBNL pgII

25

Last years measurement marked by x

Black – transmittance Red - reflectance

Still not perfect agreement with ISE but very close

Silver edge tape added at LBNL

26

TracePro modeling approach For pursuit of correction factors to use with 150 mm spheres

Accurately modeling interface between the three parts of the sample

Records energy for each surface, i.e. resolves direct-hemispherical transmittance, d-h reflectance, and side-loss for each layer

2-parameter bulk scattering model can be applied to center layer while retaining interface refractive index information of the PVB

27

TracePro modeling approach

To do

Detailed report with exit position and direction of each ray to track side shift and angle-resolved scattering for comparison with goniometers

Fit bulk scattering model to yield accurate results

28

Sample Refl A1 (3mm) A2 (lam)

A3 (12 mm) Trans R+T+A SL1 SL2 SL3 Total

Clear 0.077 0.011 0.004 0.035 0.873 1.000 0 0 0 1.000

Scattering 0.487 0.054 0.015 0.084 0.344 0.984 0.003 0.0003 0.013 1.000

IR properties of rough samples

29

# Sample Description Side Gloss* Emissometer Reading FTIR Emissivity Decrease when movi 1 Uncoated V175 No Frit 74 0.827 0.846 0.019

(White) Frit 64 0.81 0.836 0.026Decrease as frit is added 0.017 0.01 0.036 <-moving d

2 Uncoated V907 No Frit 75 0.822 0.846 0.024(Black) Frit 67 0.813 0.84 0.027

0.009 0.006 0.033

3 V175 with VE-85 No Frit 64 0.096 0.089 -0.007(White) Frit 24 0.109 0.31 0.201

-0.013 -0.221 -0.02

4 V907 with VE-85 No Frit 64 0.095 0.088 -0.007(Black) Frit 31 0.12 0.25 0.13

-0.025 -0.162 -0.032

5 V175 with VE-2M No Frit 56 0.043 0.038 -0.005(White) Frit 22 0.053 0.307 0.254

-0.01 -0.269 -0.015

6 V907 with VE-2M No Frit 56 0.042 0.037 -0.005(Black) Frit 33 0.055 0.183 0.128

-0.013 -0.146 -0.018

7 V175 with VNE-63 No Frit 79 0.03 0.025 -0.005(White) Frit 43 0.036 0.263 0.227

-0.006 -0.238 -0.011

8 V907 with VNE-63 No Frit 82 0.03 0.026 -0.004(Black) Frit 51 0.046 0.231 0.185

-0.016 -0.205 -0.02

Comparison of emisso-meter and FTIR for fritted glass with coatings

Need to include IR sphere

True value?

Ballot strategy No intentional language regarding diffuse glazing in

the current ballot

Decided to try to tidy up documents with editorial changes in separate ballot to not be distracted by these for when the new diffuse language is added

30

Main components to include in 300 partially completed

Definition of how to measure diffuse and specular component

Definition of correction factors for diffuse signals

31

Main components to include in 301 partially completed

Define difference between surface and bulk scattering

Mirror 300 requirements for integrating spheres

Validate emissometer measurements as a viable alternative to IR integrating spheres

Include language for emissometer measurements

32

Main components to include in 302 partially completed

Reporting requirements

Tolerance for light scattering

33

NFRC 300 goals Section 2: Updated the products covered/not

covered for clarity.

Section 3: Updated the year on ASTM E 903, ISO 9050, ISO 15099 and updated the LBNL website.

Section 4 : Updated and added definitions both for clarity of old properties and for new.

Section 6 : Updated wavelength requirements to what is used for the IGDB and filled in gaps in the process

Section 7: Specified specular in one case 34

NFRC 300 short comings We added large thickness as a sample property that

requires special care. A specific number was left out as it would depend on the instrument geometry and the sample properties Planned solution: Change the word definition in 2.2.3 where it

references 6.1. Possibly extend the note in 6.1

We lost a footnote in 2.1 when we moved solar heat gain coefficient, that should stay in

The word Lambertian merits a definition in the terminology

ASTM E275 was updated recently to cut out NIR Update the Referenced documents to contain current titles and year

on the standards

35

NFRC 301 goals and short comings Editorial changes of the blackbody temperature

used, old number “23ºC (75ºF)” replaced by 27ºC

Added standard E408 for FTIR instruments Unintended consequence as it is not recommended for

specular samples Forgot to update year on standards

Added reporting requirements to prepare for other use of integrating spheres and emissometers

36

NFRC 302 goals and short comings

Continue the 302 TG work from 2013

The ASTM standard talking about ILCs has been discontinued Replace reference with outline for the NFRC ILC

procedure

37

Future tasks Update 300 series documents (deal with current

comments) and include work almost finished in

Multi layer calculations need more work but premature if we do not have single layer data

Computational means exist in Berkeley lab WINDOWS and Radiance but require models for angle dependence (incident and outgoing)

Models exist but would require some validation

ILC for measurement of diffuse glazing

38

One pager to research SC Carry out inter-laboratory comparison (ILC) with a

range of samples

Define a data format and implement parsing of that format for inclusion in the CGDB and to use in WINDOW

Generate virtual scattering models using TracePro of the samples used in the ILC. Data used to generate these models will be based on data from the ILC and by using LBNLs pgII goniophotometer.

39