Patty,
I thought it might be good to talk about where should the oil temperatures be on our engines. If we spend the time to install a gauge and a cooler, how do we know where to set the oil temps? Luckily for you, Holset gives you that information. Other manufacturers aren't always so giving.
As with any engineering exercise, there are multiple things going on and there is not always a clear answer. According to Holset,
your oil temperatures are below their recommendation of
90-100C. If we look at the behavior of the oil as the temperature changes, we see that it thins out and is often said to be like water at higher temps. The kinematic viscosity of 15W40 oil at 60 C is 45 mm^2/s, at 80 C it thins to 23 mm^2/s and at 100 C is 13.6 mm^2/s.
Now, in the friction power equation for a hydrodynamic bearing, the viscosity is a
linear term. When the bearing is in the fully hydrodynamic regime, which means it is fully floating on a "large" film (film thicknesses in the "wedge" vary, but they can be on the order of 20 micrometers), the
frictional power scales directly with the oil viscosity. So, if your viscosity doubles, you double the frictional power dissipated in the bearings for a given rpm. This suggests running the highest oil temperature that is practical for you to control.
Now the effective temperature of the oil in the wedge will be much higher due to the local shear stress on the fluid (150 C fluid temp in the wedge is a commonly cited value for full speed rotors and the value used for testing oils HTHS capability in the SAE J400 spec), which drops the dynamic viscosity (to 3.7 cP for 150 C (which converts to a kinematic viscosity of ~ 4.4 mm^2/s)). A lower oil inlet temperature should ideally give a lower temperature in the wedge, but due to the higher viscosity, the increased heating due to shear stress may eliminate the "gain" of trying to introduce the cooler oil. Similarly, increasing the oil flow rate (for a fixed temperature and pressure) also increases the bearing power consumption because the oil absorbs less heat, thus the oil temperature in the bearing is reduced which increases the fluids shear stress.
The power loss in a turbocharger due to the bearing friction is relatively low compared to the turbine power. For a smaller turbo than yours, a group in Paris (Podevin, P., et al, "Influence of Lubricating Oil Pressure and Temperature on the Performance at Low Speeds of a Centrifugal Compressor for an Automotive Engine", Applied Thermal Engineering, 2011, Vol. 31, pp. 194-201) measured the power loss due to the bearings of 475 W at 110000 rpm, at which the turbine power was between 3500 to 5500 W depending on the mass flow through the compressor. From this, the bearing loss is between 8.5 to 13.5% of the turbine power. Now this was only half way through the compressors map, which extended up to 240 krpms, due to the speed limitations of their torque meter, so the mechanical losses are usually larger at lower speeds and thus the percentage of turbine power will be smaller at full speeds, but they can still be significant. The plot of the bearing power was climbing exponentially as would be expected from the friction power equation and inspection of the Stribeck curve, so at full speed the bearing power consumption is much larger. For larger turbines, like the Garrett/Caterpillar ACERT engines, the bearing power consumption is more like 2800 W at 110000 rpms.
The same group in a separate paper also measured a bearing power loss of 580 W with 40 C oil temperature and 425 W for 60 C oil temperature. Furthering this, the University of Stuttgart measured torque frictional losses that show around 15% reduction in mechanical losses by increasing the oil temperature from 80 C to 100 C and almost a 25% reduction by increasing the oil inlet temperature from 60 C to 100 C.
As in all engineering problems, there may be a compromise...the bearing stiffness and damping vary directly with the viscosity too. Thinner fluids reduce the bearing stiffness and damping, which means that ability of the rotor system to handle imbalances, fluid driven excitation, etc., is changed. It's not possible to say if it is better able to handle these or worse, because that depends on the dynamics of the entire system. Increasing stiffness can sometimes case coupling of the rotor response to the excitation source. Reducing stiffness can do the same thing at times.
So this boils down to a determination of risk, how well do you believe the rotordynamics stay the same as you change this.
So is your 80 C oil temperature ok? If I were designing the engine, I would stick with the manufacturers recommendations. They have worked out the rotordynamics and component life with their values. But, I think you are safe at any temperature below the recommendation and maybe even benefit due to potentially increased bearing wedge clearance to a point, but if the manufacturer's data is available, I think it makes more sense to use it.
Now, if you look at the required oil pressures, you are at the very minimum for the full torque oil supply pressure. But, bearing losses also increase with oil supply pressure...so if you were going for minimum bearing losses, you would run at the minimum pressure. You could try to change the oil type to try to get a bit more high temperature operating pressure by going to an oil with a higher HTHS rating...so 10W50 or 10W60...the winter rating (15W or 10W) isn't a big deal since you will soon be hot!, the HTHS rating is where you need to be concerned.
The bigger issue is if you have enough flow rate...The flow rate of your set-up should be checked against Holset's recommendation of 3 l/min at max power if you wanted to be sure the lube system was up to specs. My guess would be that if you make the required pressures during operation, that the pump flow rate is sufficient since the bearing clearances should be the controlling area, but this assumption may be wrong and I would check this.
Finally a word on pumps with pressure reliefs. Having a pump with excess capacity is definitely a very desirable feature. Since gear, vane, gerotor (etc.) pumps are all constant volume, the flow rate is dependent on the rpm and the pressure is based on the controlling orifice area with which it has to "push" the volume flow rate through. If you have a bypass on your pump and the pump reaches the cracking pressure opening the bypass passage, you now have two controlling areas in series that form the effective controlling area. If the bypass area is not correctly sized, has a lower flow resistance, etc., the volume flow rate reaching you bearings could be below the required flow rate. Just something to be aware of if you are designing such a device into your system and definitely should be tested.
Good luck!
Chris
