Nerd-O-Scope: Keeping The Energy In Turbos Using PTP Turbo Blankets


A PTP Turbo Blanket is seen here in an application for the BMW twin turbo S63 engine.

Using some very rough numbers and a ton of assumptions, we can guesstimate the heat loss by radiation. I took an average of the three measured surface temperatures with no turbo blanket resulting in a value of 584 K. The other assumptions are an ambient temperature of 298 K, a surface area of 0.17m^2, and an emissivity of oxidized cast iron of 0.64. Cranking out the numbers results in about 670 watts of heat being lost by radiation from the turbine housing. Keep in mind one horsepower is equal to 746 watts, so almost one horsepower of energy that could be going through the turbine wheel is lost. In an actual engine bay, there is going to be some heat transfer by convection from air flowing over the turbine housing too. This could be equal or greatly more than the heat lost by radiation depending on the speed of the airflow through the engine bay and the temperature of the air. I think it is safe to say a few horsepower are lost from the turbine due to heat transfer. Keep in mind these numbers are at a part load condition on a diesel. At full load, a diesel will easily have exhaust gas temperatures over 700 deg C, or 200 deg C more than this test. Heat transfer by radiation is a function of temperature to the fourth power, so a LOT more heat would be lost.

Let’s say with the 200 deg C hotter exhaust gas, the surface temperature increases by 100 deg C. Those changes result in the power lost due to radiation jumping from 670 watts to 1300 watts! Of course, the heat lost due to convection will increase also. If talking about gasoline engine temperatures, the exhaust gas temperatures can go over 950 deg C, or another 200 deg C hotter than diesel. Adding another 100 deg C surface temperature increases the heat loss due to radiation up to 2280 watts, or about three horsepower. Adding a handful of horsepower to a turbine wheel to spin it up will definitely improve spool up and response. For those of you in the Subaru crowd, you know Jeff Perrin will show dyno results with a hot header and a colder header. With a hot header, the dyno plots always show the turbo spooling faster. Why? Because when the header is cold, it is sucking heat out of the exhaust gas before it gets to the turbo. Fortunately for us, the research conducted by Bickle gives us further proof.


Turbo speed and boost pressure were measured at five different engine speeds and loads. In every case, as one would expect having the turbo blanket installed resulted in higher turbo speed and higher boost pressure. This of course means the turbine wheel is producing more power with the help of more heat which in turn puts more power into the compressor wheel generating the higher speeds and boost pressure. These measurements were taken at 25%-30% of maximum engine torque, so basically very lightly loaded. At higher loads, the differences would be more drastic.
One critical performance metric of turbocharger performance is time-to-torque, or basically how fast the turbo spools up. In this plot, the vanes of the VGT were set to a fixed position to simulate a wastegate turbo and the throttle tip-in began at 1500 rpm and 90 lb-ft of torque. With the blanket installed, turbo speed and boost pressure start off higher and appears to me to increase faster than without the blanket. However, the trends are close enough where it cannot be said conclusively that the rate of increase was faster.
The test was conducted again at 1750 rpm and 100 lb-ft starting condition. Here you can see the boost pressure does indeed increase at a faster rate with the turbo blanket installed. The time gap at 16 kPa is about half a second. At 22 kPa, that gap has increased to around one second.
This plot is from throttle tip-in tests beginning at 1000 rpm and 60 lb-ft of torque, so a very low load. The gains with the turbo blanket installed are immediate resulting in a 0.3 second time advantage and 0.3 bar of boost pressure advantage. Races are won by hundredths or sometimes even thousandths of a second, so 0.3 seconds can be an eternity.

Leave a Reply

Your email address will not be published. Required fields are marked *