TC21: Furnace Design and Operation

Members Downloads

Download Manager

TC21 Overview Poster

The main activity of TC21 is to improve the quality and reliability of glass furnace simulation modeling and optimization of software packages of different suppliers and glass producing factories that describe heat transfer, flows and temperatures in glass furnaces (melt, batch & combustion space).

The most effective way to understand the strong and weak points is by simulating with the different participants the same well defined existing glass melting furnace and ideally with actual measured and validated data. This allows the different participants to compare and validate results with each other and also with real measured data.

Such a comparison activity is usually referred to as a Round Robin Test (RRT). In the past, TC21 has used several different so called RRTs and has now reached definition RRT5 for a formerly existing furnace with detailed measured data.

2019 Activities and Plans 2020

Dear Members and those interested  in TC 15&21, with the everyday acceptance of modeling TC21 (formerly modeling) combined with TC15 (sensors and controls) to form a new TC21 committee with the focus on Glass Furnace Design and Operation

2019 Activities and Plans

In 2019 TC 21 held meetings at both the International Seminar on Glass Furnace Design and Operation in Velke Karlovice CZ and the ICG Congress in Boston, USA.  TC21 also sponsered a session at the ICG Congress that had over 250 attendees for the presentations.

TC21 conducted Phase I of a round robin test (RRT) for modeling a small cold top electric glass melter with C-glass.  The geometry description and necessary data to develop a full 3D CFD model was compiled and distributed to the TC21 members. Results from several members were presented at the 2019 meetings and issues and modifications identified.  Summary of results and revisions to the modeling setup will be focus in 2020.

See the attached downloadable file for an overview of TC21 at the top of the page.

The 2020 planned meeting for TC21 will be:

20 September 2020 15:00-17:00 pm at the ICG Annual Meeting held in Krakow, Poland

All who are interested are welcome to attend (details such as meeting room etc will be sent out when assigned before the dates)

2018 Activities and Plans

TC21 is conducting a round robin test (RRT) for modeling a small cold top electric glass melter with C-glass.  The geometry description and necessary data to develop a full 3D CFD model was compiled and distributed to the TC21 members. Several members have responded that they are participating.  The first phase consists of modeling the melter and then presenting and discussing the results and observations at the scheduled May and June 2019 TC21 meetings listed below.

Phase 2 will be variations of this design as recommended by those participating in the discussion of phase 1.  We are looking forward to the discussion for this round robin test and will document with future presentations.

Also, TC21 is organizing a session for the ICG Congress in Boston USA.  The TC21 session is:  SYMPOSIUM III: GLASS TECHNOLOGY AND MANUFACTURING Session 2.  This session will focus on designs or new technology leading to improvements for glass furnaces and operations. Improvements in emissions, throughput, life, energy and quality are all important aspects for a glass furnace.  Papers presenting the concept or technology, demonstrating the difference it will make with modeling and implementation results have been requested and 4 invited papers are planned.  The session organizers are: Aaron Huber, Johns Manville, USA, and Scott Cooper, Owens Illinois, USA,

The 2019 planned meetings for TC21 will be:

21 May 2019 16:00-17:30 before the 15th international seminar on furnace design operation and process simulation held in Velke Karlovice, CZ  20 June 2019

9 June 2019 16:00-18:00 pm at the ICG Congress held in Boston MA USA

All who are interested are welcome to attend (meeting rooms etc will be sent out when assigned before the dates)

2017 Activity Summary and future plans

20 June 2017 the Technical committee 15/21 met in Velke Karlovice, Slovakia.  24 people were in attendance.  An overview of the committee was presented and discussion held on activities.  Discussion about modeling the proposed Glass Futures test furnace was held and efforts to keep in contact at they develop plans.

During the last half of the meeting 4 breakout groups were formed to discuss future activities and needs.

Each group reported what they discussed.  Improved batch modeling was discussed by 3 of the 4 groups with suggestions of RRT and to recommend projects to GlassTrend, universities, companies etc. to improve properties (TC18), simulation of redox, coverage (islands and log formation) etc.  Some of the other discussed topics were improving glass quality predictions, sensors, how to share results between companies, NOx control and emissions etc.

With busy schedules it was challenging to find a time to meet in 2018.  A wbinar was also suggested and email requests to see who would attend the 2018 ESG meeting and 2018 ICG meeting.  TC15/21 is sponsering a session at the 2019 ICG Congress to be held in Boston along with a TC meetings at Boston and at the 2019 Furnace Design Seminar in Czech Republic.  Good participation is anticipated at those activities.

2016 Activity Summary

On 4 September 2016 the Technical committee 15/21 met in Sheffield UK.  There were 9 participants in attendance and 21 members who emailed that they would not be able to attend.

Erik Muijsenberg presented the agenda and reviewed activities of the committee (attached file).  Aaron Huber was introduced as the new chairman with Erik serving as vice-chair.

Alistair Wallace from Glass Futures was invited to review their plans with the committee.  Glass Futures was formed to address the need for attracting talent, training and development facilities for the glass industry.  The plan is to construct a 30 ton per day development furnace along with smaller pilot melters (from 5 kg and up) that can be used for training and development.  The presentation will be later forwarded

One beneficial collaboration between Glass Futures and the TC15/21 members would be to model the 30 ton per day furnace and compare results with the focus on batch models.  With batch modeling one of the weak parts of furnace modeling this would allow comparison of different modeling approaches and the actual batch shape obtained with the Glass Futures furnace.  This and other topics will be discussed at the next committee meetings.

An updated ICG website is now in use and in October 2016 TC15/21 committee members on record should receive an email with a user name and password for the TC site.  Files will be posted that can be download along with information on TC meetings.

Next meetings

1.     21 June 2017 the committee will meet after the technical sessions at the 14th International Seminar on Furnace Design – Operation & Process Simulation, 21-22 June 2017, Velke Karlovice, CZ.

2.     For those that can attend we are planning on meeting in November 2017 after the first day of technical sessions at the 78th Glass Problems Conference in Columbus Ohio.  (   Please pass this information on to anyone who might in interested in attending.

Aaron M. Huber, Ph.D.

Senior Manager, Furnace Research Group, Process Technology

Johns Manville | A Berkshire Hathaway Company

Johns Manville Technical Center

10100 West Ute Ave

Littleton, Colorado  80127

O:  303 978 2403


TC21/TC15 met during the ICG Congress in Prague 2013 and presented a paper, as part of the official ICG program, informing delegates of their activities.

Minutes of the TC21/TC15 Technical meeting:

The ICG conference in Prague was well attended and so the TC21/TC15 meeting on Sunday was also popular. However some people apologized that they could only come to the conference starting from Monday (or in some cases from Tuesday) and were unable or unwilling to travel and arrive on Sunday to attend the Technical meeting. It seems that those who travel further are more likely to attend a technical meeting on a Sunday.

The meeting was also attended by some TC15 members. TC21 and TC15 (Advanced Control & Sensors) decided to cooperate more closely in future and work on merging these 2 TCs.

In total 11 people attended this TC21/TC15 meeting. In the past numbers had ranged from 10 right up to 50 attendees.

Initially we discussed briefly some new results for RRT5. Mr. Muschick and Celsian had some updated results to share. They were asked submit these to the chairman so that later they could be included in a possible paper.

Andries Habraken gave a general presentation on how to do furnace modeling.

Mr. Linz presented the mission and targets of the former TC15 and discussed some new sensors. He commented that in a survey they did few years ago within the TC15 the interest in a possible foam sensor was very high.

The figure below is:

3D picture of TV panel melting furnace analyzed in RRT5


After the general discussion and presentations we changed the focus more to the new topic of furnace design and operation. The aim was to initiate a discussion forum on certain technologies used in glass melting such as forced bubbling for example. With forced bubbling, typically air is introduced via bubbler pipes into the bottom of the glass melting furnace to enhance convection and so heat transfer. But the bubbles created by these bubblers may also create extra defects, reduce the glass residence time in the melter and cause bottom corrosion.

Different experts made the following comments:

  • For green bottle glass production bubbling is highly interesting and may increase bottom temperatures by 25°C,
  • For clear tableware production, bubbling does not make a lot of sense, more likely bringing a risk of defects,
  • The experience in fibre glass is very positive, also as the bubblers break open the foam,
  • A comment was made that bubble size is important,
  • Larger bubbles resulting in fewer bubbles/minute would reduce the bubbling effect,
  • For some cases bubbling with oxygen may be better or more safe than bubbling with normal air,
  • It was suggested that bubbler rows might be placed in different positions to those traditionally chosen, e.g. to create cross recirculation. Also fiber glass producers often use more than one row,
  • It was commented that even so, bubbling for fiber glass is seen as very positive,
  • Today the trend is to use bubblers at lower flow rates.

This could be an interesting parameter to investigate further: what is the optimal position of bubblers to enhance glass melting?

The discussion was followed by a question from Aaron Huber: what is the more effective barrier, bubbling or boosting?

  • Şişecam commented that bubbling can maintain better the batch line,
  • Schott commented this decision depends on many parameters,
  • Toyo Glass Japan commented that they had had a bad experience in the past with barrier boosting which had increased their seed level,
  • The chairman made the comment that a half barrier is similar to no barrier.

General conclusion of the group:

Bubbling can be very positive in glass melting, but many parameters should be considered to decide if it should be applied or not.

During the last and partly the previous meeting we discussed the next steps to be followed and suggested:

  • Update RRT5, the 2004 case with increased convection, as was already done for the 2002 case by Glass Service;
  • All participants model also the 2002 case (process data situation 2002, with increased air cooling) and send results to the TC21 chairman;
  • Supply data for the calculated glass convection velocities using the same TC checkpoints along the length as used by the participants of RRT5;
  • Complete evaluation with calculations for both cases, using tracers (“particle tracing”) resulting in minimum and residence time distribution curves (send to the chairman);
  • Trace bubbles from batch blanket and refractory (definition will follow);
  • Compare so called glass melt quality indicators.

Alternative ways in which TC21&TC15 might continue working together were discussed. What seems to work, and already is done, is to organize special lecture sessions during ICG conferences (congresses & annual meetings). This was done for the first time at the ICG meeting in Salvador, Brazil and also during ICG/ESG/DGG in Maastricht; such sessions were well attended, with over 100 people present.

Papers from TC21 and TC15 members are most welcome for any future event.

Plans for 2014:

NEXT REGULAR TC MEETING of TC21/TC15 Glass Furnace Design & Operation: ICG meeting during the ICG/ESG meeting in Parma 21-24 September, to discuss progress and actions.

Co-organize the Special “Hot Sensor Symposium” on Instrumentation and Control Innovation in Glass Manufacturing, November 6, 2014 in Columbus, Ohio following the Glass Problems Conference (WWW.GLASSPROBLEMSCONFERENCE.ORG):

Description: The Hot Sensors Symposium on Instrumentation and Control Innovation in Glass Manufacturing is focused on the latest technologies in the market to support critical processes and address current challenges in the glass manufacturing process.  It provides a forum for the audience to gain technical knowledge and exchange experiences with each other in support of the advancement and application of sensors and controls technologies.

Committee Members

Bessette, Didier
Bouillet, Fabien
Eisenga, Menno
Gul, Burcin
Fasilow, Fabrice
Habraken, Andries
Hofmann, Otto
Huber, Aaron Committee Position: Chair
Jochem, Klaus
Johnson, William
Kang, Guosheng
Kuhn, Wolf
Lindig, Matthias
Linz, Wilfried (Mr)
Maehara, Terutaka
Meuleman, Rene
Mueller-Simon, Hayo
Muijsenberg, Erik
Muschick, Wolfgang
Philipp, Gerd
Pierrot, Laurent
Polcyn, Adam
Purnode, B.
Sato, Keizo
Solis, I.
Unwin, Graham
Wright, C.

Committee Contact Details

TC21: Furnace Design & Operation
Johns Manville Technical Center
10100 West Ute Ave.
Littleton, Colorado USA 80127

Tel: 303-978-2403