According to Andrew Hobbis of Atlanta-based rolling company Novelis, who also presented at the meeting, metal temperatures reach 300 to 500 C during hot rolling, as 2- to 500-millimeter strips are thinned out. A water and oil emulsion is typically used to cool and lubricate during the process.

Cold rolling temperatures, Hobbis continued, range from 20 to 200 C. The cold process further thins the aluminum sheets down to 0.1-0.4 mm thick or even less. Cold rolling oils are usually neat kerosene oil or water.

Deneuville described the challenges a rolling oil must overcome. Aluminum arrives at a rolling mill in large slabs covered in oxides and abrasive particles from the casting process. Rolling oils have to cope with these particles as well as different sheet sizes and entry angles to the roll bite-the point at which the aluminum sheet encounters the rollers.

As aluminum sheets pass through the rolls, a fresh, reactive surface is exposed, which can stick to and build up on the rolls. This transfer layer can then be deposited back onto the aluminum sheet, affecting uniformity of thickness and finish quality. One of the roles of the additive [in the oil] is to limit this layer on the rolls, added Deneuville.

Rolling emulsions must create a consistent film of oil and additive inside the roll bite. Selecting an oil with the right coefficient of friction is important not only to increase efficiency, but also to trap the metal in the roll bite and make it pass through. Viscosity plays an important part in avoiding bite refusal, as well as preventing build-up on the rolls.

However, properties that give rolling oils greater lubricity can make them more difficult to handle and maintain. You have almost two contradictory constraints, explained Deneuville. The coolant must be stable enough for storage and manipulation, yet a less stable emulsion has good friction reduction. Operators typically choose an intermediate emulsion, such as a white emulsion, to try to split the difference.

Choosing the Right Oil

The most important factor in rolling oil selection, Deneuville emphasized, is choosing the right laboratory tests that correspond to the intended application.

Deneuville listed oil concentration in water, pH, emulsion temperature, spraying conditions, rolling conditions such as speed, and the presence of any tramp oils as the main parameters to consider when selecting or designing tests.

An oils cooling capacity can be measured by spraying a heated plate with rolling emulsion, matching the flow rate and other parameters to the mill and calculating the heat exchange coefficient. Operators can then choose to adapt the oil, especially the surfactant, to improve heat exchange.

Measuring the coefficient of friction is typically done using tribological bench tests like pin-on-disc instruments. Ive done plenty of experiments on these types of tribometers, Deneuville reported. I believed in them. I do not believe in them anymore. The problem, he said, is that such bench tests do not create a fresh surface as often as happens in a mill, and therefore do not capture an oils real-world performance.

Devices such as a hot rolling simulator can be used to try to mimic conditions in a mill in order to test for specific properties. But in his experience, Deneuville lamented, performance in an actual mill was completely inverse from the lab test results.

Using a laboratory mill could provide a solution, but will only produce a few samples. The tool is useful, however, because parameters can be changed, such as metal temperature, the metal itself, and oil concentration and type. These tests are a first step in oil selection, Deneuville stressed, and only trials in an actual aluminum rolling mill will provide conclusive evidence of real-world performance.

The choice of rolling oil is normally the job of a mills lubricant engineer, who should be assisted by a chemist and a rolling specialist. Experience on the ground is the most important, he summarized. You have to use your mills as a reference and what you know in the past as a reference.

Pilot Testing

Conshohocken, Pennsylvania-based Quaker Chemical built a pilot mill at its facility in Shanghai, China, in order to test the performance of its oils. The facility, which has capacity for both hot and cold rolling processes, is used to develop new lubricant concepts, study emulsion properties and investigate new raw materials, wrote Bas Smeulders in a technical paper submitted to STLE.

The four-high, backup roll-driven reversing mill was commissioned in June 2016. It rolls aluminum strips of 100 to 200 mm in width to a gauge of 4 mm, and has a recirculating emulsion system of 3.1 cubic meters.

Smeulders told the group gathered in Atlanta that Quaker conducted 15 trials for hot rolling oils to examine the effects of lubricant type and emulsion condition on rolled products. Three oils were tested, each at different roll forces and in different conditions: a fresh emulsion, an emulsion aged for one week, and an aged emulsion dosed with a phosphorous based extreme pressure additive.

Two thousand liters of emulsion with oil concentration at 4 percent was prepared one day before the trial to allow for overnight circulation at 50 C. The emulsion was used to heat the work rolls before the trial while the mill was running idle.

The aluminum strip was a 5052 alloy with a width of 200 mm. The trial began with coil temperature at 480 C, and the strip was rolled for three passes, reducing thickness by 15 percent, then 25 percent, then 21.6 percent to 2 mm at exit of the third pass. Speed progressively increased from 100 to 200 meters per minute.

Surface finish on the milled product was inspected to observe the performance of the oils.

Smeulders reported that oil 2 performed best as measured by roll force and forward slip values. Oil particle size for this emulsion was the largest, and consequently it had the lowest stability index.

The oils performance was reproduced with pin-on-disc tribometer testing at 316 C. Results showed a lower coefficient of friction for oil 2 and corresponded with the ranking for the oils in the pilot mill trial.

Aging improved rolling performance for all three emulsions, corresponding with increased oil particle size after aging. Adding the EP additive improved performance for oil 1, but not the others, particularly oils that already contained significant amounts of EP additives.

The emulsions performance was also visible on the rolled surface, which was investigated with a scanning electron microscope. On the second pass, oil 1 produced significant scooping and much higher roll force. Scooping could be evidence of a particular type of lubrication mechanism dominating for that oil, Smeulders explained. The researchers believe this mechanism involves significant contribution from wedging abrasive wear. Further study is needed to determine whether there is a general correlation between wedging wear and higher roll forces, said Smeulders.

Oil 2 had less scooping in the second pass. The researchers pointed to more plowing type abrasive wear with this oil, caused by the plowing action of work roll asperities on the workpiece surface.

Wedging wear is seen when interfacial shear stress is very high and sliding occurs within the bulk metal, while plowing is encountered when sliding occurs over interfacial layers, causing deformation of the top-most surface by intra-metallic shear.

Surface cracking was observed with oil 1 after the third pass, but was not reflected in higher rolling force or higher forward slip. More study is needed to understand this phenomenon, Smeulders concluded.

Application Design

As a representative of an industrial aluminum producer, Hobbis of Novelis emphasized to meeting attendees the importance of lubricant application methods and described his investigation into the way aluminum rolling coolant is applied to rolls.

In a mill, lubricant is applied by several rows of spray nozzles spaced one to three inches apart across the width of the roll. Valves can turn the nozzles on and off individually or in zones to selectively cool the roll. Nozzles spray coolant in two shapes: flat or full cone.

I think cones on average are slightly more efficient, said Hobbis, but the flat nozzles are much easier to nest because you can overlap them without them interfering. Interference between nozzles can cause stripes on the roll and finished product.

The amount of oil sprayed can be controlled, as well as the pulse of application. Pulse control is a more modern way of adjusting oil application, according to Hobbis. An older approach is to simply switch the nozzles on or off.

Hobbis colleagues measured the cooling efficiency of spray using a heated rotating roll rig, examining the effects of flow rate, pressure, nozzle type, spray geometry, roll speed and coolant type.

What we found with the older level control was that the actual amount of cooling is not monotonically related to the flow rate, Hobbis reported, so you could get into situations where you increase the coolant level but the amount of cooling actually goes down.

The most significant factors are the extent of the spray coverage area, flow rate and roll speed. Optimal cooling results from large spray coverage, achieved by spacing spray bars around the circumference of the rolls. Hobbis recommended cooling the exit side of the rolls, where they are hottest. He acknowledged that this is not always possible, however, since coolant carryover can be a problem for some products.

Hobbis cautioned against positioning the spray bars too close to the roll, recommending they be aimed almost perpendicular to the roll so the nozzles will be less sensitive to changes in roll diameter as they wear. You dont want the geometry of the spray system to be too dependent on the roll diameter, he explained. Target 10 to 50 percent overlap for the full range of spray, he concluded.