Mercedes Power Unit Hardware & Software

[MrPotatoHead]

Every intake that I design and run through CFD I do both a static flow simulation through all of the runners open and then a dynamic simulation with a true time based simulation of per cylinder intake valve opening time.

It’s amazing the differences in flow that you will see between the two.

What do u mean by"static flow simulation"? U mean steady state flow...

Static flow as in no interruption, people forget combustion events aren't continuous, the valves close and open and this interrupts flow, this is dynamic.

Exactly.
You also have to factor in the air entering the plenum via the inlet. It gets incredibly complex.

[dren]

What do u mean by"static flow simulation"? U mean steady state flow...

Static flow as in no interruption, people forget combustion events aren't continuous, the valves close and open and this interrupts flow, this is dynamic.

Exactly.
You also have to factor in the air entering the plenum via the inlet. It gets incredibly complex.

Who designs the intake plenum? I'd assume it would be the PU manufacturer, but does the chassis maker take numbers from the PU maker and design it to fit the chassis?

[MrPotatoHead]

Static flow as in no interruption, people forget combustion events aren't continuous, the valves close and open and this interrupts flow, this is dynamic.

Exactly.
You also have to factor in the air entering the plenum via the inlet. It gets incredibly complex.

Who designs the intake plenum? I'd assume it would be the PU manufacturer, but does the chassis maker take numbers from the PU maker and design it to fit the chassis?

Intake manifold comes with the power unit.

[toraabe]

What is more interesting is the fuel consumption
https://www.auto-motor-und-sport.de/for ... 96761.html

In Mexico merc used 89 kg and ferrari 101 kg for the same race distance
It proves that merc normally are running their engine well below maximum output..

[MrPotatoHead]

What is more interesting is the fuel consumption
https://www.auto-motor-und-sport.de/for ... 96761.html

In Mexico merc used 89 kg and ferrari 101 kg for the same race distance
It proves that merc normally are running their engine well below maximum output..

I have a hard time believing those fuel consumption numbers if Lewis was allowed to drive on full power the whole race.

[Santozini]

What is more interesting is the fuel consumption
https://www.auto-motor-und-sport.de/for ... 96761.html

In Mexico merc used 89 kg and ferrari 101 kg for the same race distance
It proves that merc normally are running their engine well below maximum output..

I have a hard time believing those fuel consumption numbers if Lewis was allowed to drive on full power the whole race.

I might be wrong but I think Hamilton was allowed to drive on full power the whole race in Brazil, not Mexico..Sorry if I missed something

[MrPotatoHead]

What is more interesting is the fuel consumption
https://www.auto-motor-und-sport.de/for ... 96761.html

In Mexico merc used 89 kg and ferrari 101 kg for the same race distance
It proves that merc normally are running their engine well below maximum output..

I have a hard time believing those fuel consumption numbers if Lewis was allowed to drive on full power the whole race.

I might be wrong but I think Hamilton was allowed to drive on full power the whole race in Brazil, not Mexico..Sorry if I missed something

Whoops, my bad - i thought you said Brazil not Mexico, because you shared a link about Brazil.

[MrPotatoHead]

Here's a static (steady state) CFD simulation on a twin plenum V6 Intake Manifold that I did recently.
Hopefully this isn't off topic since it gives you an idea of the complexity of the flow and the velocities involved.
Velocity colours are shown on the top left.
You will notice differing velocity from the centre of the plenum to the edges as well.

[toraabe]

What is more interesting is the fuel consumption
https://www.auto-motor-und-sport.de/for ... 96761.html

In Mexico merc used 89 kg and ferrari 101 kg for the same race distance
It proves that merc normally are running their engine well below maximum output..

I have a hard time believing those fuel consumption numbers if Lewis was allowed to drive on full power the whole race.

I might be wrong but I think Hamilton was allowed to drive on full power the whole race in Brazil, not Mexico..Sorry if I missed something

He was ...

[gruntguru]

Here's a static (steady state) CFD simulation on a twin plenum V6 Intake Manifold that I did recently.
Hopefully this isn't off topic since it gives you an idea of the complexity of the flow and the velocities involved.
Velocity colours are shown on the top left.
You will notice differing velocity from the centre of the plenum to the edges as well.
https://i.imgur.com/wTzTkES.png

Nice image Mr P and a perfect illustration of our discussion. A couple of observations:
1. All the velocities are way too high. That doesn't matter much for the purposes of your sim but readers here should not take any notice of the actual values - perhaps divide them by 8 to 10.
2. The key take-away is the runner velocities are 2 - 3 times the plenum velocity. And of course most of the runner flow happens during about 1/4 of the cycle, so actual runner velocity is four times as high while the runner is flowing - and even higher at its peak.

So runner velocities at about 10x plenum velocity. This approaches the ideal of zero plenum velocity which is what you get with ITBs and open velocity stacks. (difficult on boosted motors)

[63l8qrrfy6]

I'm no CFD expert but surely in this particular case the low plenum velocities are an effect of the massive recirculation area ? This is hugely different from a divergent section that recovers the static head with minimal losses.

[PlatinumZealot]

Mr potatohead. Work is done for the day but your image just triggered flashbacks of the "management style" at my workpalce. Yes. Good work there. But i myself have done CFD work on intakes and that model you setup there looks snazzy... But is far from the complete article. Mutiphysics would needed to account for the accoustics. Also steady state give totally inaccurate results. To increase the fidelity a little bit you can use a math function for the exit pressure of each runner to represent the pulses of suction into the cylinders. At least it is just an approximation. If uour software has multiphysics you can set up that to see if there are any accoustic effects. But even without that.... The pressure pulses should show clear differences for each cylinder.

[63l8qrrfy6]

I'm pretty sure that's solidworks CFD. I've had the misfortune of using it about 7-8 years ago.

[MrPotatoHead]

Mr potatohead. Work is done for the day but your image just triggered flashbacks of the "management style" at my workpalce. Yes. Good work there. But i myself have done CFD work on intakes and that model you setup there looks snazzy... But is far from the complete article. Mutiphysics would needed to account for the accoustics. Also steady state give totally inaccurate results. To increase the fidelity a little bit you can use a math function for the exit pressure of each runner to represent the pulses of suction into the cylinders. At least it is just an approximation. If uour software has multiphysics you can set up that to see if there are any accoustic effects. But even without that.... The pressure pulses should show clear differences for each cylinder.

PZ - you are correct, a static flow is very different from a dynamic time based simulation.
I do have some videos of the time based / cam event based simulation for this same design. I'll think about posting those later. They do indeed show very different results - that is why i do simulation using both.

[MrPotatoHead]

I'm pretty sure that's solidworks CFD. I've had the misfortune of using it about 7-8 years ago.

It is. It keeps getting better and better, just like their FEA which I use a lot also.