Cord Defects.

Cord is the name for a vitreous (glassy) defect in a glass article that has a different composition and hence different properties from the glass being manufactured. Cord defects can be due to the following:

• Problems with the composition of the raw materials.
• Incorrect weighing or mixing of the raw materials during batch preparation.
• Segregation of the batch during transport to or charging into the furnace. • Inadequate homogenisation during the melting process.
• Loss of volatile components of the glass from the glass surface during melting and conditioning. • Dissolution of refractories into the glass.

“Cat Scratch” Cord:

• “Cat scratch” cord is a surface cord originating from the dissolution of refractories.
• It exhibits as a line or series of lines down the sidewall of the container on the surface of the glass, having the appearance of the scratch of a cat’s claw.
• If severe it can be felt with the fingernail or tip.
• The glass of the cord is enriched in alumina or zirconia or both. • It has been a very common problem in glass manufacture for about 80 years or more.

“Cat Scratch” Cord Origin.

Virtually all tableware and container manufacturers have this problem to varying degrees. It is a visual defect in normal quality bottles which can be tolerated if limited in extent. However, it can be unacceptable if it becomes extreme or is present in premium quality bottles. It can be a particular problem in tableware as the surface cord contributes to a greater percentage of the overall glass thickness resulting in a highly visible defect unacceptable for good quality tableware.

In the 1970’s and 80’s fusion cast refractory suppliers indicated that the majority of “cat scratch” cord originates from zirconia containing refractories such as fusion cast AZS or “lower quality” zircon-mullite bonded materials used in the working[1]end (distributor) and forehearths and recommended that zirconia containing materials should not be used downstream from the throat. They recommended that fusion cast alpha-beta alumina material be used.

Some glass companies still believe that this is the case. However, “cat scratch” cord regularly occurs on forehearths where no zirconia containing materials are used. It can also occur on the first day of production on a new furnace when no significant corrosion of any refractories could have occurred.

“Cat Scratch” Cord Source.

The source of ”cat scratch” cord is the vitreous (glassy) phase present in the fusion cast AZS refractories used extensively in the furnace melting-end. This glassy phase, which amounts to about 20% of the crystallographic composition of the fusion cast AZS refractory, exudes from the refractory surface on initial heating. This explains why the “cat scratch” cord defect can be present at the very start of a furnace campaign when no significant corrosion of the refractories could have taken place. The vitreous (glassy) phase also comes out of the refractory during subsequent corrosion by the glass during the campaign.

If this “cat scratch” cord material remains dispersed in the glass, then it is not visible and will only register as a general low increase in the alumina and zirconia content of the base glass chemical analysis. However, if the exudation or corrosion is excessive, or if the material is allowed to settle out in the distributor and forehearth, the material will be concentrated in the glass and result in the visible “cat scratch” cord defect. The fusion cast AZS vitreous (glassy) phase is much more viscous than the base glass with a different viscosity versus temperature relationship and is therefore more likely to settle out at the lower temperatures in the distributor and forehearths due to the greater difference in viscosity. In settling out it will travel along the bottom of the forehearth, and feeder spout so will be present on the surface of the gob which forms the surface of the bottle.

Fusion cast AZS refractories used for distributor and forehearth glass contact refractories could be a source of the “cat scratch” cord defect due to the glassy phase present. However, the warmup and operating temperatures in the distributor and forehearth are much less than the furnace and so the degree of exudation of the glassy phase and the rate of corrosion of the refractory is much less. Due to the difference in operating temperatures the relative corrosion conditions for the refractories in the furnace melting-end are up to 32 times greater than in the distributor and 64 times greater than in the forehearths. Nevertheless, fusion cast alpha-beta alumina refractories perform better than fusion cast AZS refractories in the distributor and forehearths with regard to the “cat scratch” cord defect, not because they contain no zirconia, but because they contain minimal glassy phase (typically 2%) and exhibit no exudation on initial heating.

Bonded Refractories for the Forehearth and Distributor.

Bonded zircon-mullite and bonded high alumina refractories also contain no glassy phase and are not a source of “cat scratch” cord. PSR 993, a 99∙7% dense bonded alumina material has equivalent corrosion resistance to fusion cast alpha-beta alumina. Bonded zircon[1]mullite materials such as PSR 333, PSR 315 and PSR 930 are widely used for feeder spouts and feeder expendable parts without causing any “cat scratch” cord defects and will generally dissolve “gracefully” into the glass creating neither seed, blisters, stones, or cord.

Cord Dispersal System (CDS) – A Guaranteed Solution.

When we introduced our Cord Dispersal System (CDS) as a solution to “cat scratch” cord in 2002, many customers stated that they had already tried stirrers and they did not work. Other suppliers also stated that stirrers were not effective and promoted the use of bottom drains as the solution. However, our experience indicated that correctly designed and configured stirrers were the best solution to this problem. Such was our confidence that we decided to offer a money-back guarantee such that if the CDS failed to eliminate or minimise the defect to the customer’s satisfaction, we would work to modify the system as necessary, at no additional cost to the customer, to achieve the required results. Ultimately, if we were unsuccessful, then we would take the system back and re-imburse the full cost of the equipment. To date we have supplied 300 CDS on that basis with complete success and have not been asked to provide a refund.

We regularly install this equipment on forehearths already equipped with another stirrer system or a bottom drain which had been unsuccessful at satisfactorily minimising the “cat scratch” cord and we achieve complete success with our CDS.

In addition to eliminating or minimising the “cat scratch” cord defect the CDS has also been found to improve the glass thermal homogeneity. They have also proved successful in dispersing colour streaks following furnace colour changes and other persistent colour streaks, such as white flint streaks in amber glass, amber streaks in white flint glass and black streaks in dark green glass.

We have supplied 300 CDS with a money-back guarantee and have not been asked to provide a refund.

Cord Dispersal System Developments

The Heli-paddle Stirrer Design.

In 2017 we introduced the heli-paddle stirrer which incorporates a helix into the centre of the stirrer as an extension of the paddle blade. This increases the stirring action through the depth of the glass, providing the possibility of dispersing any other cord that is not on the bottom of the forehearth channel. The first field trial of these heli-paddle stirrers in an existing system resulted in breakage of the stirrers after only a few hours of operation, which was found to be due to the additional torque created by this stirrer design. Exhaustive testing was carried out in our Workshop, stirring one of our refractory raw materials in a box to achieve close to the maximum full load current of the stirrer drive motor. This re-created the problem, and the cause was corrected by a change in design to the mechanism. This was thoroughly tested in the same way, before offering the heli-paddle stirrers for installation in new systems. Existing CDS supplied before that time can be modified to be suitable for heli-paddle stirrers.

The heli-paddle stirrers have been found to provide approximately double the stirring effect of normal paddle stirrers, with approximately half the stirrer rotational speed required to provide the same stirring effect. This should result in reduced stirrer wear and longer stirrer operating life. They have been found not only to be suitable for dispersing “cat scratch” cord and colour streaks but also cord due to inadequate homogenisation of the glass during the melting process.

One customer was experiencing cord on a very old furnace operating with two forehearths, and the furnace tonnage had to be significantly reduced to minimise this cord. The cord was producing stress in the glass which could not be reduced to acceptable levels by the annealing process. A CDS was supplied for one of the forehearths as soon as possible. Following installation and operation of the system, the cord was dispersed sufficiently in that forehearth so that the furnace tonnage could be increased, but the other forehearth glass condition still deteriorated. A second system was subsequently supplied for the other forehearth, and the tonnage could then be increased so that the furnace could operate satisfactorily with production maintained until the planned furnace rebuild.

CDS Re-configuration and Upgrades.

As we have now been supplying our CDS for 22 years, many systems have been re-used following furnace rebuilds and, in some cases, we have found that changes have been made to the forehearths without consideration of the stirrer systems. To successfully operate, the stirrer configuration must be suitable for the forehearth channel width and design, and the stirrer access opening must be in the correct position with the correct dimensions. In cases where the superstructure refractories have been replaced hot during the furnace campaign to install a CDS, the new forehearth design must have the same stirrer opening configuration if the stirrer system is to be re-installed. If changes are being made to the forehearth, then it is advisable for PSR to be consulted at the design stage to determine any modifications required for correct re-installation of the stirrer systems to avoid problems during or after the rebuild.

Earlier CDS used a chain drive mechanism which has since been replaced with a shaft drive system together with other improvements. A furnace rebuild may also be a suitable time to upgrade.

The upgrade eliminates the chain drive mechanism and any associated maintenance, such as the need for re-tensioning due to stretching and lubrication. The shaft drive mechanism provides a smoother operation which is not reliant on chain tension and does not produce any side thrust on the stirrer shaft nearest the motor reducing the potential for carbon bearing wear. A telescopic drive shaft connection allows easier setting up and replacement of the drive motor / gear box as well as positioning of the stirrer back plate. The original bevel gear trantorque couplings have been replaced with adjustable shaft mounted bevel gears to allow easier and more reliable setting of the relative stirrer paddle positions. A new back plate guard provides access to the bevel gears during stirrer replacement and set[1]up via a lockable access door without having to remove the entire back plate guard. Remote grease points are provided for the high temperature bearings on the drive shaft.

As the glass industry looks at new solutions for melting glass to reduce carbon emissions, perhaps the root cause of “cat scratch” cord will be addressed with different refractories or methods of melting, or alternatively maybe this will result in new defects emerging. Whatever the case PSR’s Cord Dispersal System will be ready to provide a solution.

Click here for more information about PSR Cord Dispersal Systems – a proven and guaranteed solution to the problem of cat scratch cord.