Variable nozzle turbines (VNT) were supposed to be the killer turbo technology - the wastegateless design able to be matched to engines to give excellent response throughout the rev range. And, while they have been adopted in huge numbers in diesel engine applications, problems associated with the high exhaust gas temperatures of petrol engines have continued to prevent their more widespread use in passenger cars.
Ball bearing turbos? Yes, it's a technology that's now being applied widely - although it hasn't seemed to have made quite the impact that was tipped for it.
But now, waiting in the wings, are two turbo technologies that have the potential to absolutely revolutionise turbocharging in passenger car petrol engine applications.
Air bearings and electrically assisted turbos.
Major turbo manufacturers are feverishly working on turbos that have the potential to produce boost at literally idle revs, and to spool up to speed in a shorter time than would be possible with any conventional turbo design. Not only that, but the electrically-assisted turbos are designed to also function as generators for the car's electrical system, potentially removing the need for what is becoming an increasingly expensive and heavy alternator.
Now if a turbo manufacturer were to make a presentation to an OE car company showing a device that needs no lubrication, has a life longer than that of the car, gives the car no weight penalty (the alternator possibly gets left off, remember), improves bottom-end and transient torque very substantially, has the ability to improve fuel economy and emissions - well, put that list of attributes together and you can see why the technology is so exciting!
For this two-part special series we've scoured resources across the world - studying patent applications, trawling through scientific and engineering papers, and talking to turbo company personnel. The result is a story like you'll find nowhere else - a detailed look at the technologies that you can expect to see in turbos five years down the track.
This week - air bearings.
Turbos and Oil
Current turbos use one of two centre bearing designs. (The centre bearing supports the shaft which has the compressor at one end and the turbine at the other.) Sleeve bearings like the pictured design are still the most common, while more modern turbos use ball bearings. One of the major advantages of ball bearings is that they can survive with a far smaller oil supply - and less oil means less drag. In fact, it is suggested by turbo engineers that the reduction in oil flow through the bearing is at least as significant as the lower friction of the ball bearings in giving quicker spool-up onto boost.
But having any oil in the centre bearing gives rise to a number of significant disadvantages:
Another point is that in that in a fuel cell application, the supply of air to the cell has to be scrupulously clean - less contaminated than can be achieved with a conventional oil-bearing turbo.
Quoted in Automotive Design and Production, Robert Gillette, president of Garrett Engine Boosting Systems, says, "Preventing oil from entering the intake charge is important in any gasoline or diesel engine, but vital in low emission vehicles. Any amount of oil in the system potentially could foul the system."
So a turbo bearing system that doesn't require any oil supply has some potentially major benefits - and air bearings look like being the best way of achieving that.
Air bearings are being developed for high-speed turbomachinery not just because that will give us a better spool-up time on the road in our turbo cars, but because the new bearings are very suited to aircraft turbine engines. In aircraft applications, studies have shown that there could be a 15 per cent weight saving if the requirement for a bearing oil supply was removed.
A number of companies - including Capstone Turbines, Pratt & Whitney, Mohawk Innovative Technology Inc and Honeywell - are currently working on the development of high-speed, high-temperature air bearings.
And Honeywell owns Garret, one of the best-known names in turbos.
There is also major government research interest, with the Oil-Free Turbomachinery Program at the NASA Glenn Research Center performing major work on the bearings.
So, what actually is an air bearing?
An air bearing consists of a shaft surrounded by very thin shaped foils. The foils both provide support for the shaft when it is stopped and also direct and control the airflow when the shaft is spinning. At high speeds, foil air bearings maintain the air film between moving parts by actually pumping air between the rotating and stationary surfaces. Air is drawn in and adheres to the bearing surface, no matter how fast it is moving. The pumping of air between the moving surfaces creates an air pressure that generates a load-carrying capacity. At high speeds, a very thin layer of air, just under one-thousandth of an inch thick, can support hundreds of kilograms.
This animation shows that as the journal shaft (grey) starts to rotate, it drags a film of air between it and the top foil (purple). As the hydrodynamic pressure increases, a force is exerted on top foil. This pressure pushes this foil away from the journal, moving it back a tiny distance against the backing bump foil. After the lift-off speed, the journal 'floats' on this hydrodynamic film of air without touching the top foil.
In one example of the forces involved, an electric motor was used to turn the shaft. Prior to lift-off, the shaft required a torque of about 4Nm to turn it. After lift-off, this dropped to just 1.5Nm, then on touchdown the torque peaked at about 6Nm.
While the technology has been known for many years - air bearings are used in the hard-drive read/write heads in computers - it is only recently that there has been sufficient development to produce bearings that can cope with the speeds, loads and temperatures associated with turbos.
The First Generation of foil bearings used either flat leaves...
...or bump foils. The bumps on this First Generation foil bearing are uniformly spaced across and around the bearing. Note that the foils are only 100µm thick!
The Second Generation of air bearing foils have variable compliance characteristics, and are split circumferentially (ie around the shaft).
In this Second Generation design, a foil backing spring uses a variable pitch.
The Third Generation design (the one that is currently being used) has an even more complex foil with variable-pitch bumps and a circumferential split.
The maximum weight that the bearing can support (ie its load capacity) is dependent on the design of the top and bump foils; and the diameter, length and speed of the rotating shaft. A faster shaft speed increases the load capacity by pumping more air between the top foil and the shaft, while a larger bearing increases load capacity by providing a bigger area over which the load can be spread. The design of the foils has a different affect - it determines how quickly the load capacity increases with speed for a particular bearing.
The First Generation of foils used a uniformly-corrugated bump foil, which resulted in air leakage at the edges. As a result, this type of bearing had a relatively low load capacity. The current Third Generation design uses features which are better at trapping and maintaining the air film between the shaft and top foil. These refinements have resulted in great improvements in the weight that can be supported by foil air bearings. This graph shows the improved performance of the latest design of bearings when compared with the First Generation.
One of many interesting aspects of these bearings is that not only does the load carrying ability alter with shaft speed, but so do other characteristics. Mohawk Innovative Technology, Inc suggests that under light loads and speeds their foil air bearing, "maintains its softer character, while under high load and speed, it stiffens up to provide stability and high load capacity. Additionally, the higher the amplitude of vibration, the greater the damping value becomes to dissipate vibrations and make for a smoother operating system; as vibrations are reduced and operation becomes smoother, the bearing system damping automatically reduces as well."
Until recently, a major limitation of foil air bearings was that their use was limited to applications where they would be subjected to temperatures below 300 degrees C. This was the case because during startup and shutdown, the bearing and shaft make rubbing contact until sufficient rotational speeds pump air into the bearings to develop the running clearance. To reduce wear when contact is being made, a dry lubricant coating is applied to the foil and shaft. Most current foil-bearing applications use a polymer coating such as Teflon® that is limited to about 300 degrees C.
However, the researchers at the NASA Glenn Research Center have pioneered new surface coatings that have lifted the upper temperature limits to over 650 degrees C. One of the coatings (US Patent No 5,866,518 'Self-Lubricating Composite Containing Chromium Oxide'), consists of chemicals acting as a binder, hardener, high-temp lubrication and low-temperature lubrication.
The lubricant has been successfully used in a foil bearing application with more than 100,000 start-up/shut-down cycles successfully completed! Making the conditions even harder is that the test was carried out with a turbine inlet temperature of 650 degrees C and 2.5 times the normal static start-up loading. The test mule was a 150hp turbocharger equipped with the air bearings.
One hundred-thousand hot start/stop cycles..... with no oil!
A plasma spray, as shown above, is used to apply the solid lubricant coating. The actual process of coating the shaft is shown in the steps below.
Firstly, the shaft is undercut.
Then it is sandblasted, providing a surface with good grip characteristics.
The plasma spray is then applied...
...before it is ground back to a 0.005-0.0010 inch thickness. In normal use, wear particles fill any surface pits or voids, resulting in a mirror smooth surface.
Further development of the coating is expected to provide a lubricant which:
Test air bearing-equipped turbos have already been taken to speeds of 121,500 rpm - a limit that was set not by the turbo, but by the testing equipment! In this test, the journal bearings operated at 4.4 million DN (diameter in mm times speed in rpm), which is well beyond the capability of even advanced oil-lubricated ball bearings. (At high speeds, balls and rollers are forced away from the shaft and place large stresses on the bearing.)
The test turbo (shown here) was based on a 150hp Caterpillar diesel engine turbo that normally spins to 95,000 rpm. It has a rotor 2kg in mass and 254mm long. The special turbo used a 95.5mm thrust runner and two double-acting foil thrust bearings. Journal bearings consisted of a pair of 36.5mm air foil assemblies.
In addition to turbochargers and aircraft gas turbines, the air bearing technology also has potential benefits for flywheel energy storage mechanisms, aircraft auxiliary power units and gas expanders used for cooling. Given the wide applications of the technology and the progress that has already been made, it's very likely that air bearing turbochargers for car applications aren't far away.
Garret personnel unofficially hint that air bearing turbos are in fact much closer than you might imagine....
Next week: electrically assisted turbos