Mercedes Power Unit Hardware & Software

[MrPotatoHead]

The transient response of a turbine engine is terrible.

"Then the transient response time will need to be improved, won't it?" says the executive paying the salaries.

You don't think they have done that as much as possible on modern turbines?
Even a small turbine fitted to a radio controlled airplane can take many seconds to spool up.

[PlatinumZealot]

Exactlty. You are building pressure in an open system everytime you need more torque. Go to the back of the turbine and you look right through the blades. Think about that. You esentailly have a chamber with some blades on either side and you have to build pressure inside it. The turbine is free flowing so the air has to basically compress against itself. Combustion occurs at one pressure as well. That response aint happening until that back pressure builds up with flow rate and heat spread! That takes seconds due to size of chamber speed of sound and what not. For a piston engine this is not a problem because the combustion chamber is very small, igntion is rapid and chamber is esentially a closed system. I am rusty when it comes to turbines.. But i know the response time aint pretty.

[dren]

Exactlty. You are building pressure in an open system everytime you need more torque. Go to the back of the turbine and you look right through the blades. Think about that. You esentailly have a chamber with some blades on either side and you have to build pressure inside it. The turbine is free flowing so the air has to basically compress against itself. Combustion occurs at one pressure as well. That response aint happening until that back pressure builds up with flow rate and heat spread! That takes seconds due to size of chamber speed of sound and what not. For a piston engine this is not a problem because the combustion chamber is very small, igntion is rapid and chamber is esentially a closed system. I am rusty when it comes to turbines.. But i know the response time aint pretty.

You still need the mass flow. And yes, response times can be quick depending on what your energy source is. In a combustion turbine you are instantly adding more fuel to be combusted.

[roon]

Regarding qualifying modes and potential oil burning: I'm curious about crankcase pressure dynamics. Assuming a 90* V6 with a three-throw crank, we have three cylinder pairs--one pair per crank throw. If the crankcase were not a continuous volume but rather segregated by piston pair, there would be three separate crankcase volumes. Each of these would experience a rise and fall in pressure as the cylinder pairs rise and fall. Inlets, outlets, and passive valves could allow this volume to act as a pump.

It could not move an equal volume to it's piston working volume, though. At the top and bottom 45* of crank rotation, the cylinder pairs are rising and falling at same time. No change in crankcase volume. So perhaps 2/3 of the swept volume of the cylinders could be pumped through the crankcase using this method. In this application that would mean the crankcase could move ~1 liter of air per crankshaft revolution through the crankcase.

Since these are 4-stroke engines, this means the crankcase could provide a massflow ~33% greater than the engine proper.

What could be done with such a 'crankcase charge?' How would it be routed?

Aside from the above, what else might be worth considering here? Would there be any benefit to running very high or low crankcase pressures in general? A highly pressurized crankcase would allow the pistons to transfer more work to each other through the crankcase atmosphere. Perhaps reducing frictional losses.

The transient response of a turbine engine is terrible.

"Then the transient response time will need to be improved, won't it?" says the executive paying the salaries.

You don't think they have done that as much as possible on modern turbines?
Even a small turbine fitted to a radio controlled airplane can take many seconds to spool up.

Has there been a need to? Primary applications remain aircraft and power plants. Automotive applications have been limited to concept cars, range extenders, one-off customs, drag racing, tractor pull and an Indy car. None of those required quick 'throttle response.' RC model aircraft engine manufacturing would not be a hotbed of research on the topic; regardless, that the engine is small does not alter the relationship between the mass flow it can produce and apply to its turbine. So it should not be expected to necessarily spool up faster.

Exactlty. You are building pressure in an open system everytime you need more torque. Go to the back of the turbine and you look right through the blades. Think about that. You esentailly have a chamber with some blades on either side and you have to build pressure inside it. The turbine is free flowing so the air has to basically compress against itself. Combustion occurs at one pressure as well. That response aint happening until that back pressure builds up with flow rate and heat spread! That takes seconds due to size of chamber speed of sound and what not. For a piston engine this is not a problem because the combustion chamber is very small, igntion is rapid and chamber is esentially a closed system. I am rusty when it comes to turbines.. But i know the response time aint pretty.

You still need the mass flow. And yes, response times can be quick depending on what your energy source is.

This is the important part. Automotive turbochargers can spool quickly due to a variety of different techniques, despite being able to "look right through the blades," PZ.

[Tommy Cookers]

what fundamentally limits the response rate is overtemperatures - the turbine is always a heat diluted engine
the F1 ICE could without fuel rate limit have a stronger response if using conventional mixture strength - till it blew up

the turbine-electric of course has no response issues
turbine Indy and F1 cars had 4 wd for braking to hold the turbine at power
turbine-electric is much better in this respect

the angular velocity eg in pitch over curbs and eg in yaw at Loew's hairpin is far higher than in turbine-powered aviation
due to conservation of angular momentum ie 'gyroscopic' reaction torques you'd presumably want eg a 2 spool engine
this (net) torque represents a force somewhere somewhen resisting the driver's intentions

[roon]

The composition of the lubricant is unknown. Would more-viscous, gelled gasoline or alcohol (Sterno) have sufficient lubricating qualities?

In 2007, NASCAR owner/driver Michael Waltrip and his team had severe penalties handed down by officials. Team Vice-President Bobby Kennedy and Crew Chief David Hyder of Waltrip's No. 55 were ejected from the Daytona International Speedway. Hyder was fined $100,000 for his involvement with lining fuel tanks and intake valves with Sterno. NASCAR also docked Waltrip 100 owner points and disqualified his qualifying speed for the Daytona 500. Mid-week, NASCAR determined that the then unknown substance was Sterno. When the highly regulated NASCAR fuel was added, the Sterno would liquefy, giving the car an added octane boost.[4]

Some options:

• The vapors emitted from the lubricant when heated or subjected to high shear or pressure, would render the sump breather gases flammable. Greater energy content in the air-fuel charge when injesting sump breather gases.
• The lubricant interacts with the air-fuel charge as a vapor introduced through sump gases or cylinder wall film.
• The lubricant is being mixed with the fuel directly in order to alter the properties of the fuel (octane, energy content, combustion properties). I don't see a regulation specifying that only fuel may exist within the fuel lines, or that only fuel may pass through the fuel injector.

A line linking the fuel injectors to the valvetrain lubrication would be internal and hidden from view.

If the lubricant is introduced pre-sensor, how would this influence the sensor's reading of the material passing through it? Only attempting to increase the fuel flow rate post-sensor is illegal. Altering the sensor's reading by varying the composition of the materials passing through it is not addressed in the rules.

5.10.5 Any device, system or procedure the purpose and/or effect of which is to increase the flow
rate after the measurement point is prohibited.

[PlatinumZealot]

This is the important part. Automotive turbochargers can spool quickly due to a variety of different techniques, despite being able to "look right through the blades," PZ.

Aahmm... An automotige turbocharger is not an engine! It has no combustion chamber. And it is a radial turbine which is definitively a quick response turbine with high pressure gradient per stage. That is the biggest reason why radial turbines are used on cars!

If you put combustion chamber in there... It would behave not like what you woulf expect either. They dont make much turbine engines only single stage radial turbine these days. Alot of them have a radial stage with other axial stages.. But it is not as like a turbocharger. And response is not as fast either.

[PlatinumZealot]

Regarding qualifying modes and potential oil burning: I'm curious about crankcase pressure dynamics. Assuming a 90* V6 with a three-throw crank, we have three cylinder pairs--one pair per crank throw. If the crankcase were not a continuous volume but rather segregated by piston pair, there would be three separate crankcase volumes. Each of these would experience a rise and fall in pressure as the cylinder pairs rise and fall. Inlets, outlets, and passive valves could allow this volume to act as a pump.

i am aware that street car engines made in the late 1990's early 2000's with partitioned crank cases such as the Toyota "AZ" engines do use a special passages joining the compartments to equalize air pressure under all the pistons as they reciprocate. These are inline engines.. v6 might not have such a big problem.. But still It reduces pumping losses and improves lubrication under the piston rings. It won't be a good idea to add significant back-pressure on this system though.

[MrPotatoHead]

Regarding qualifying modes and potential oil burning: I'm curious about crankcase pressure dynamics. Assuming a 90* V6 with a three-throw crank, we have three cylinder pairs--one pair per crank throw. If the crankcase were not a continuous volume but rather segregated by piston pair, there would be three separate crankcase volumes. Each of these would experience a rise and fall in pressure as the cylinder pairs rise and fall. Inlets, outlets, and passive valves could allow this volume to act as a pump.

i am aware that street car engines made in the late 1990's early 2000's with partitioned crank cases such as the Toyota "AZ" engines do use a special passages joining the compartments to equalize air pressure under all the pistons as they reciprocate. These are inline engines.. v6 might not have such a big problem.. But still It reduces pumping losses and improves lubrication under the piston rings. It won't be a good idea to add significant back-pressure on this system though.

Almost every engine on the road today has "windows" between the adjacent cylinder main webbing to help with cylinder to cylinder breathing.
On most race engines these windows are missing as a Dry Sump system is used and they are not needed.

99.9% of the time an engine will make more power with a negative pressure in the crankcase because of the increased ring sealing.

[MrPotatoHead]

The composition of the lubricant is unknown. Would more-viscous, gelled gasoline or alcohol (Sterno) have sufficient lubricating qualities?

In 2007, NASCAR owner/driver Michael Waltrip and his team had severe penalties handed down by officials. Team Vice-President Bobby Kennedy and Crew Chief David Hyder of Waltrip's No. 55 were ejected from the Daytona International Speedway. Hyder was fined $100,000 for his involvement with lining fuel tanks and intake valves with Sterno. NASCAR also docked Waltrip 100 owner points and disqualified his qualifying speed for the Daytona 500. Mid-week, NASCAR determined that the then unknown substance was Sterno. When the highly regulated NASCAR fuel was added, the Sterno would liquefy, giving the car an added octane boost.[4]

Some options:

• The vapors emitted from the lubricant when heated or subjected to high shear or pressure, would render the sump breather gases flammable. Greater energy content in the air-fuel charge when injesting sump breather gases.
• The lubricant interacts with the air-fuel charge as a vapor introduced through sump gases or cylinder wall film.
• The lubricant is being mixed with the fuel directly in order to alter the properties of the fuel (octane, energy content, combustion properties). I don't see a regulation specifying that only fuel may exist within the fuel lines, or that only fuel may pass through the fuel injector.

A line linking the fuel injectors to the valvetrain lubrication would be internal and hidden from view.

If the lubricant is introduced pre-sensor, how would this influence the sensor's reading of the material passing through it? Only attempting to increase the fuel flow rate post-sensor is illegal. Altering the sensor's reading by varying the composition of the materials passing through it is not addressed in the rules.

5.10.5 Any device, system or procedure the purpose and/or effect of which is to increase the flow
rate after the measurement point is prohibited.

You are really reaching here... The fuel flow regulations are quite clear:

5.1.4 Fuel mass flow must not exceed 100kg/h <--- It doesn't matter what fuel, that's it.

5.14.2 Other than engine sump breather gases, exhaust gas recirculation, and fuel for the normal purpose of combustion in the engine, the spraying of any substance into the engine intake air is forbidden.

Once again the new rule added for 2018 tells you all that you need to know about "oil burning":

7.9 Oil injection: The use of active control valves between any part of the PU and the engine intake air is forbidden.

[MrPotatoHead]

Regarding qualifying modes and potential oil burning: I'm curious about crankcase pressure dynamics. Assuming a 90* V6 with a three-throw crank, we have three cylinder pairs--one pair per crank throw. If the crankcase were not a continuous volume but rather segregated by piston pair, there would be three separate crankcase volumes. Each of these would experience a rise and fall in pressure as the cylinder pairs rise and fall. Inlets, outlets, and passive valves could allow this volume to act as a pump.

The question would be why even try? A shortage of pressurized air is not the problem in any fashion - fuel is what is limited.

[PlatinumZealot]

Regarding qualifying modes and potential oil burning: I'm curious about crankcase pressure dynamics. Assuming a 90* V6 with a three-throw crank, we have three cylinder pairs--one pair per crank throw. If the crankcase were not a continuous volume but rather segregated by piston pair, there would be three separate crankcase volumes. Each of these would experience a rise and fall in pressure as the cylinder pairs rise and fall. Inlets, outlets, and passive valves could allow this volume to act as a pump.

i am aware that street car engines made in the late 1990's early 2000's with partitioned crank cases such as the Toyota "AZ" engines do use a special passages joining the compartments to equalize air pressure under all the pistons as they reciprocate. These are inline engines.. v6 might not have such a big problem.. But still It reduces pumping losses and improves lubrication under the piston rings. It won't be a good idea to add significant back-pressure on this system though.

Almost every engine on the road today has "windows" between the adjacent cylinder main webbing to help with cylinder to cylinder breathing.
On most race engines these windows are missing as a Dry Sump system is used and they are not needed.

99.9% of the time an engine will make more power with a negative pressure in the crankcase because of the increased ring sealing.

Yes.Today those little windows are pretty common, but back in the 1990's it was a special feature.

[Tommy Cookers]

btw
iirc 1950s Bill Wilshire ? made 'Wilhelmthing' ? a 4 stroke sprint motorcycle supercharged from under-piston displacement
a conventional parallel/inline twin ie the pistons were in phase and displaced on each downstroke the charge required

similarly a V6 might have 90 deg banks and a 'zero degree' crank ie each bank's pistons in phase, and maybe 6 throws
or c. 50 deg banks with a 3 throw crank

[roon]

You are really reaching here...

Pretty low risk environment here, so I don't mind reaching.

The fuel flow regulations are quite clear:

5.1.4 Fuel mass flow must not exceed 100kg/h <--- It doesn't matter what fuel, that's it.

I'm more curious about how the sensor functions and if it is tuned toward a specific fluid composition. Alteration of the fuel beyond the testable, regulated components, via an additive downstream of the fuel tank, might influence how the sensor reads. Just speculating.

5.14.2 Other than engine sump breather gases, exhaust gas recirculation, and fuel for the normal purpose of combustion in the engine, the spraying of any substance into the engine intake air is forbidden.

I don't think this precludes introducing substances into the fuel lines or the direct-injectors. Engine lubricant in this case.

Once again the new rule added for 2018 tells you all that you need to know about "oil burning":

7.9 Oil injection: The use of active control valves between any part of the PU and the engine intake air is forbidden.

The wording is interesting. To me this implies the sump breather had an active valve emptying into the intake plenum(s). But this should be normal/expected. By regs, they have to route breather air through the engine. They want this to be a passive valve instead? Releasing gases only once a certain pressure level is reached.

[MrPotatoHead]

You are really reaching here...

Pretty low risk environment here, so I don't mind reaching.

The fuel flow regulations are quite clear:

5.1.4 Fuel mass flow must not exceed 100kg/h <--- It doesn't matter what fuel, that's it.

I'm more curious about how the sensor functions and if it is tuned toward a specific fluid composition. Alteration of the fuel beyond the testable, regulated components, via an additive downstream of the fuel tank, might influence how the sensor reads. Just speculating.

5.14.2 Other than engine sump breather gases, exhaust gas recirculation, and fuel for the normal purpose of combustion in the engine, the spraying of any substance into the engine intake air is forbidden.

I don't think this precludes introducing substances into the fuel lines or the direct-injectors. Engine lubricant in this case.

Once again the new rule added for 2018 tells you all that you need to know about "oil burning":

7.9 Oil injection: The use of active control valves between any part of the PU and the engine intake air is forbidden.

The wording is interesting. To me this implies the sump breather had an active valve emptying into the intake plenum(s). But this should be normal/expected. By regs, they have to route breather air through the engine. They want this to be a passive valve instead? Releasing gases only once a certain pressure level is reached.

The rules are very explicit about introducing anything post sensor. This would be a big big no no. Again a reach.

The sensor is Ultrasonic:

https://www.gillsc.com/products/flow-se ... w-meter-2/

http://www.racecar-engineering.com/tech ... ters-work/

The wording is interesting. To me this implies the sump breather had an active valve emptying into the intake plenum(s). But this should be normal/expected. By regs, they have to route breather air through the engine. They want this to be a passive valve instead? Releasing gases only once a certain pressure level is reached.

The wording is new for 2018 and it is because of the Mercedes engines breather system.
As much as the "oil burn" stuff was exaggerated the only way they could control it differently in different sessions of the weekend was through the use of an Active control valve.