Increasing energy costs over the past few years have provided incentive for metallurgical thermal-processing companies to look for new ways to improve the efficiency of their furnaces.

Fig. 1. View of bottom of radiant tubes

The continuous annealing furnace at GalvTech has long been identified as a candidate for energy-efficiency improvements. Unfortunately, the required capital investment to purchase traditional recuperators and the increase in NOx emissions, along with the necessary downtime to retrofit the furnace for installation, has made recuperators less attractive. In 2005, GalvTech was presented with an energy-saving device, the SpyroCor™, a silicon carbide (SiC) radiant-tube insert – produced by Spinworks, LLC – that would increase radiant-tube efficiencies. These devices require a relatively small capital investment and an acceptable amount of effort to install and maintain. GalvTech made the decision to purchase the SpyroCor, and since installation in August 2005, the amount of natural gas consumed per ton of steel produced has decreased by over 14%.


GalvTech is one of three hot-dip galvanizing plants that make up The Techs Industries, Inc., all of which are located in the Pittsburgh, Pa., area. The furnace at GalvTech consists of four sections: the preheat furnace, the radiant-tube section or annealing furnace, the controlled-cool section and the jet-cool section. It is a horizontal furnace approximately 344 feet in total length. The area of focus for this paper is the radiant-tube section of the furnace, which is divided into four control zones.

Fig. 2. Radiant-tube burner and exhaust elbow


The radiant-tube furnace is approximately 98 feet long. There are a total of 60 radiant tubes in the furnace. Figure 1 shows a side view of the bottom tubes. The tubes are U-shaped with a 7-inch outside diameter, and they are approximately 77.75 inches long. The burners installed on the radiant tubes are natural gas fired with a theoretical maximum capacity of 565,750 BTU/hour. Figure 2 shows a close-up view of the radiant-tube burner and exhaust-stack configuration.

Over the past few years, fluctuations in natural gas prices have made it difficult to estimate the energy costs required per ton of steel produced. Going back to the summer of 2005, natural gas pricing fluctuated from $7.00/MCF to almost $14.00/MCF delivered. Natural gas is the single-largest utility expense at the GalvTech plant, and the primary consumer of natural gas within the plant is the strip furnace. In recent years, most of the attention has focused on concepts to increase furnace efficiency. The preheat furnace is the largest consumer section as it uses roughly four times as much gas as the radiant-tube section. With the understanding that the preheat furnace was the area with the most potential for efficiency increases, a recuperator was installed on this furnace section in December 2004. Savings of slightly over 20% were realized following the start-up. The next obvious target for savings was the radiant-tube section.

There is a large amount of unused energy that escapes out of the exhaust legs of the radiant tubes. Over the years, studies have been done at GalvTech to investigate using recuperative burners on the radiant tubes. It was then decided that the costs associated with purchasing and modifying the existing equipment and installing the necessary new equipment, as well as the line downtime associated with changing from non-recuperative to recuperative radiant-tube burners, outweighed the savings associated with them. The other considerations that negatively impacted the recuperative-burner decision were the increase in NOx emissions, the increased maintenance associated with the system and, in general, the complexity that would be added to what was a very straightforward process.

Fig. 3. U-tube firing leg and exhaust leg

SiC Inserts

In 2005, GalvTech was approached by Spinworks, LLC with an interesting concept that, if successful, would enable energy savings to be realized in the radiant-tube section of the furnace. What set this concept apart from ones that had been evaluated in the past was that the size of the capital investment for the equipment and the minimal conversion costs associated with installation would provide a less-than-one-year payback, even with gas prices at the bottom of the aforementioned price range. Also, there would not be any complexity added to the existing system.

The project would build upon technology and control schemes that were already in place. Spinworks proposed the utilization of their SpyroCor SiC inserts to replace the existing refractory tube inserts as a means to increase the amount of heat being transferred from the exhaust leg of the U-shaped radiant tubes to the inside of the furnace and ultimately to the load, which in this case is the steel strip being annealed.

Fig. 4. Side and frontal views of the twisted-Y-shaped SpyroCor SiC radiant-tube inserts

U-Shaped Radiant Tubes

The GalvTech radiant tubes are U-shaped and consist of a firing leg and an exhaust leg as depicted in Fig. 3. The firing leg is that part of the tube from the burner to the 180-degree elbow. The flame from the burner is adjusted so that its length just reaches the elbow and begins to make the bend. If the burners on the GalvTech annealing furnace are tuned correctly, the flame should not reach into the exhaust leg. A combustion analysis is done periodically to check the performance of the burners.

U-shaped radiant tubes are characterized by a heat-transfer imbalance between the firing leg and the exhaust leg of the tube. Inside the firing leg of the tube, the highly luminous flame transfers heat to the outer shell of the tube via both radiation and convection. Since there is no flame present in the exhaust leg of the tube – only hot gases with an emissivity of 0.05 or less – the only significant means of heat transfer is achieved through convection. This imbalance suggests that there is more heat transfer going on in the firing leg than the exhaust leg. Consequently, more heat is input to the furnace from the firing leg than the exhaust leg. In the past, refractory-type inserts have been installed in the exhaust leg in an attempt to improve the heat-transfer characteristics of that part of the tube.

Fig. 5. Side and frontal views of the refractory-type radiant-tube inserts

Benefits of SiC Inserts

What makes the SpyroCor SiC inserts work better than the traditional refractory insert is fairly straightforward. The new inserts are made from silicon carbide in the shape of a twisted-Y pattern. The geometry of the twisted-Y shape allows for the surface area to be increased significantly over the same length. The greatly increased surface area of the SiC insert combined with the material’s emissivity of 0.95 are the characteristics that allow it to do a more effective job of heat transfer – via radiation – between the hot insert and the radiant-tube shell. Figures 4 and 5 show a couple of views of the different inserts. The end result is that the overall heat transfer of the radiant tube is improved. Less energy is wasted out the exhaust stack, and more energy goes into the furnace. Therefore, the radiant-tube furnace doesn’t have to fire as hard to achieve the same heat input. This is how energy savings were realized in the GalvTech radiant-tube furnace. Figure 6 summarizes the difference in heat-transfer methodology with and without the SiC inserts.

Fig. 6. Heat-transfer methods taking place in each leg of the U-shaped radiant tube – with and without SpyroCor SiC inserts

The Installation Project

GalvTech was limited to the amount of line downtime that could be set aside for the installation of the equipment. After reviewing the proposed installation and the amount of work involved, it was determined that getting the project completed on a regular 12-hour repair turn was achievable if the work was planned properly. Four maintenance technicians were assigned the task of getting the job completed in 12 hours. The job tasks were fairly simple in nature, but care had to be taken because the furnace was still hot for the length of time it took to get completed.

The job tasks included the following:
1. Remove the stacks and elbows from all 60 radiant tubes.
2. Remove the two insulating-brick sleeves at the end of the exhaust leg.
3. Remove the existing refractory inserts.
4. Insert three SpyroCor SiC inserts into the exhaust leg at the prescribed distance.
5. Re-install the insulating sleeves, elbows and exhaust stacks.

The job was completed on time, and the line was able to start back up on schedule.

Fig. 7. Natural gas usage comparison – refractory insert vs. SiC insert


Since the installation of the SiC inserts, two key factors have been monitored and evaluated. They are:
  • Radiant-tube furnace natural gas consumption
  • Furnace production efficiency
The primary metric that was used to evaluate the performance of the new SiC inserts was a comparison of the natural gas consumption by the radiant-tube furnace per ton of steel produced before and after the SiC inserts were installed. Figure 7 displays the comparison of the two scenarios in graphical form. The refractory-insert data displayed by the dark-blue line shows that for the seven months prior to doing the project, the average gas-usage rate in the radiant-tube section (pink line) was 0.239 MCF of natural gas/ton of steel produced. For the seven months following the SiC insert installation, averaging the actual gas usage data (light-blue line) shows that the average gas-consumption rate was reduced to 0.204 MCF of natural gas/ton of steel produced. This data comparison shows that there was a 14.6% improvement in the rate of gas consumption in the radiant-tube furnace after the SiC inserts were installed.

Fig. 8. Natural gas usage vs. feed rate for refractory insert and SiC insert

Figure 8 shows another performance comparison of SiC inserts versus refractory inserts. This chart uses the natural gas flow rate and production feed rate to show the efficiency improvement gained through use of the SiC inserts. As Table 1 also demonstrates, for both sets of conditions, there appears to be a trend that as feed rate increases, MCF/ton has a tendency to decrease and the furnace would appear to become more efficient.


In this trial, GalvTech established benchmarks for measuring the effectiveness and success of the SpyroCor™ SiC inserts:
  • Decrease the natural gas consumed per ton of steel in the radiant-tube section of their galvanizing furnace a minimum of 10%, which would result in a one-year payback. The SiC insert delivered over 14% fuel savings.
  • Retrofit of the furnace should not cause any additional downtime. The SiC inserts were installed over a 12-hour repair turn while the furnace was still “hot,” resulting in the furnace going back to production on schedule.
Reducing the cost of manufacturing is essential for all successful manufacturing organizations. In the field of metallurgical thermal processing, energy is a large part of the cost of manufacturing and an obvious place to look for savings. There are proven technologies, such as recuperative burners, for reducing the energy usage in the radiant-tube sections of galvanizing furnaces. Unfortunately, the high capital expense combined with the difficulty of the retrofit often leads to an unacceptable return on investment and prevents the realization of efficiency improvements.

A technology that has been designed for maximum heat transfer, engineered for ease of installation on existing furnaces and supplied to provide a less-than-one-year payback would be a welcome alternative to existing recuperators. As this paper has described, such an alternative exists with the SpyroCor SiC radiant-tube insert.IH

For more information:Roy Hardy - Spinworks, LLC, 5451 Merwin Lane, Erie, PA, tel: 440-899-7153; fax 440-899-7152; e-mail:; web: or Chris Winger - GalvTech, The Techs Industries, 300 Mifflin Road, Pgh., PA

Additional related information may be found by searching for these (and other) key words/terms via BNP Media SEARCH at recuperator, radiant tube, firing leg, exhaust leg