Mercedes Power Unit Hardware & Software

[pb6797]

Two Mercedes cars given up the ghost in practice. Any ideas if there is an inherent weakness in the Mercedes engine that doesn't tolerate the banking? I thought the car wouldn't "know" it was on banking - the sideways travel converting into vertical forces, so not sure how or if it would affect the engine?

But they don't seem happy.

[dans79]

Two Mercedes cars given up the ghost in practice. Any ideas if there is an inherent weakness in the Mercedes engine that doesn't tolerate the banking? I thought the car wouldn't "know" it was on banking - the sideways travel converting into vertical forces, so not sure how or if it would affect the engine?

But they don't seem happy.

I'm betting these are the first PU's thats are being run purely on Friday because they are at the end of their lifespan. With that being said I didn't hear anything wrong with Lewis's engine, so it might just be a sensor issue or something else that's just annoying.

[ferrarifire]

I wonder though what MERC are doing to their PUs to have had 2 failures in 1 day? Rumor says they are exploiting some grey area and getting more power in the lower range of the engine(Acceleration) with the sensor temp trick....These PU failures just fan the flames of suspicion

Two Mercedes cars given up the ghost in practice. Any ideas if there is an inherent weakness in the Mercedes engine that doesn't tolerate the banking? I thought the car wouldn't "know" it was on banking - the sideways travel converting into vertical forces, so not sure how or if it would affect the engine?

But they don't seem happy.

I'm betting these are the first PU's thats are being run purely on Friday because they are at the end of their lifespan. With that being said I didn't hear anything wrong with Lewis's engine, so it might just be a sensor issue or something else that's just annoying.

[Hoffman900]

I wonder though what MERC are doing to their PUs to have had 2 failures in 1 day? Rumor says they are exploiting some grey area and getting more power in the lower range of the engine(Acceleration) with the sensor temp trick....These PU failures just fan the flames of suspicion

Two Mercedes cars given up the ghost in practice. Any ideas if there is an inherent weakness in the Mercedes engine that doesn't tolerate the banking? I thought the car wouldn't "know" it was on banking - the sideways travel converting into vertical forces, so not sure how or if it would affect the engine?

But they don't seem happy.

I'm betting these are the first PU's thats are being run purely on Friday because they are at the end of their lifespan. With that being said I didn't hear anything wrong with Lewis's engine, so it might just be a sensor issue or something else that's just annoying.

Likely old engines for practice.

Also, they have the most engines on the grid, so odds are if there is going to be an issue with anyone, it's going to be a Mercedes.

Fan the flames is exactly what RedBull is doing. Lack of evidence is why they are going it alone.

I would be shocked if the performance was down to a fluid inside the walls of the plenum. Not sure if I am buying that one.

[Big Tea]

I would just like to remind people that Mercedes are the masters of mis-direction.

[Hoffman900]

I would just like to remind people that Mercedes are the masters of mis-direction.

They all are. It's part of the game.

I don't believe half of what is reported, and the other half, I'm skeptical of. They know the media will amplify whatever they say, and then it gets all twisted and distorted (like a grown up version of whisper down the lane), then we get to the point of people just pulling information out of their butt on a forum, and then here we are.

Rinse and repeat.

[dans79]

As expected

https://www.racefans.net/2021/09/03/ham ... o-20-laps/

Mercedes confirmed Hamilton was running his highest-mileage power unit, which is his original one from the beginning of the season. The fault was traced back to unusual behaviour in its oil system.

[SmallSoldier]

https://www.motorsportmagazine.com/arti ... ngines-mph

[dans79]

https://www.motorsportmagazine.com/arti ... ngines-mph

I think this part is probably a lot more likely.

Of course the super-cooled walls theory may just be a red herring and the massive plenum may simply be to give variable size of the effective plenum by having different routes through it according to whether quick response or full capacity was required (a small plenum works best for response but will run out of capacity at high speed). This would give a better initial acceleration than a conventional ‘compromise size’ plenum, before then merging to normality just as would a super-cooled plenum.

Here is a several year old video that give greater detail. It also goes into altering the size of the runner. In the video using a valve is discussed, but I'm betting that's against the technical rules (haven't checked though). I bet a fluidic switch would be legal.

[SmallSoldier]

https://www.motorsportmagazine.com/arti ... ngines-mph

I think this part is probably a lot more likely.

Of course the super-cooled walls theory may just be a red herring and the massive plenum may simply be to give variable size of the effective plenum by having different routes through it according to whether quick response or full capacity was required (a small plenum works best for response but will run out of capacity at high speed). This would give a better initial acceleration than a conventional ‘compromise size’ plenum, before then merging to normality just as would a super-cooled plenum.

Here is a several year old video that give greater detail. It also goes into altering the size of the runner. In the video using a valve is discussed, but I'm betting that's against the technical rules (haven't checked though). I bet a fluidic switch would be legal.

Good video, thanks!

[Hoffman900]

https://www.motorsportmagazine.com/arti ... ngines-mph

I think this part is probably a lot more likely.

Of course the super-cooled walls theory may just be a red herring and the massive plenum may simply be to give variable size of the effective plenum by having different routes through it according to whether quick response or full capacity was required (a small plenum works best for response but will run out of capacity at high speed). This would give a better initial acceleration than a conventional ‘compromise size’ plenum, before then merging to normality just as would a super-cooled plenum.

Here is a several year old video that give greater detail. It also goes into altering the size of the runner. In the video using a valve is discussed, but I'm betting that's against the technical rules (haven't checked though). I bet a fluidic switch would be legal.

Good video, thanks!

Variable length manifolds have been around for just about forever. The Mazda R26B had them, all the Superbikes have had them in the EFI era, Formula 1 had them until the mid 2000s (rules), Ferrari street engines, Miata's, BMW's, etc. a friend even made them for an early Chrysler Hemi for the Popular Hot Rodding Engine Masters Challenge:


Here is a graph I made within a 1D pressure dynamics engine simulator for a project I was working on. Going through a torque shaping exercise for a friend's race engine at the moment, which relies on similar things:


Ferrari had a system that actually slid in and out over the rpm range to capture those waves and keep it above a fixed inlet length baseline. Not as simple as just making it shorter as rpm's grow.

The Engineering Explained videos are okay for the casual viewer. It all gets a lot more complex than that and he does get some things incorrect. The pressure waves in the intake port / manifold does not occur due to the air column "slamming into the back of the valve"... the origins for that, that I can find are in Superflow's manual and repeated adnaseum by magazines. If the air slams into the back of the valve, you closed it too soon .

It originates at valve open due to the shock of the pressure differential, between the port and cylinder as the piston starts to descend and creates a depression. An expansion wave occurs from that shock, travels up from the valve and is reflected at the end of the runner (or until a large change of area, which may be a plenum or atmosphere) and reflected back towards the cylinder as a compression wave. Ideally, you time the strongest of the reflection compression wave just as the valve is closing and pressure differential between the cylinder and intake port are approaching being equal, to get that last little bit in. Obviously, packaging has to be considered, so this is why it is usually something like the 3rd harmonic, and obviously it's going to change over rpm.

It is also contingent on air density. Honda with their V8 looked at the speed of the waves in the intake port (measureable) to look at fuel saturation / atomizing from different injector designs. Obviously the faster the waves move, the denser / cooler the air. IF Mercedes is doing what is being guessed at, which I don't think they are - at least not as described, then their tuned lengths will change from when the air is supposedly super cooled vs "normal" operating inlet temperatures.

For a layman's perspective, this is partly why single 4bbl carbureted engines with wet manifolds work super well over something like DCOE Webers. The wet plenum allows waves to reflect uninhibited, dampens poorly timed waves to the carburetor booster (which minimizes reversion / double fueling and helps with the fuel curve), and the wet plenum allows atomization to occur. A DCOE Weber (or the like) is much more sensitive because the full force of the expansion / and reflected wave is felt by the carburetor (and carburetors work off pressure differentials) and then the entire choke / booster creates a change in area and has its own microreflections / dampens the strength of the compression wave. The only issue with the single 4bbl carburetor type manifold is variability in runner length.. which isn't always a bad thing, but gets worst with growing bore size / centerlines. This is the only positive of a carbureted engine, is they are sensitive to pressure waves due to the function of a carburetor (call it "jet signal") so you become aware of these things.

Plenums can also have the effect of going into resonance, if sized / shaped properly to create just a hair more charging pressure. I think Honda covers that in one of their white papers.

Edit:
Development of Induction and Exhaust Systems for Third-Era Honda Formula One Engines
http://www.f1-forecast.com/pdf/F1-Files ... P2_10e.pdf

These are not cutting edge ideas / understanding, but this is a nice way to show it in practice.

[zibby43]

Two Mercedes cars given up the ghost in practice. Any ideas if there is an inherent weakness in the Mercedes engine that doesn't tolerate the banking? I thought the car wouldn't "know" it was on banking - the sideways travel converting into vertical forces, so not sure how or if it would affect the engine?

But they don't seem happy.

[SmallSoldier]

I think this part is probably a lot more likely.


Here is a several year old video that give greater detail. It also goes into altering the size of the runner. In the video using a valve is discussed, but I'm betting that's against the technical rules (haven't checked though). I bet a fluidic switch would be legal.

Good video, thanks!

Variable length manifolds have been around for just about forever. The Mazda R26B had them, all the Superbikes have had them in the EFI era, Formula 1 had them until the mid 2000s (rules), Ferrari street engines, Miata's, BMW's, etc. a friend even made them for an early Chrysler Hemi for the Popular Hot Rodding Engine Masters Challenge:
https://img-s-msn-com.akamaized.net/ten ... =t&o=f&l=f

Here is a graph I made within a 1D pressure dynamics engine simulator for a project I was working on. Going through a torque shaping exercise for a friend's race engine at the moment, which relies on similar things:
https://live.staticflickr.com/65535/514 ... 376d_c.jpg

Ferrari had a system that actually slid in and out over the rpm range to capture those waves and keep it above a fixed inlet length baseline. Not as simple as just making it shorter as rpm's grow.

The Engineering Explained videos are okay for the casual viewer. It all gets a lot more complex than that and he does get some things incorrect. The pressure waves in the intake port / manifold does not occur due to the air column "slamming into the back of the valve"... the origins for that, that I can find are in Superflow's manual and repeated adnaseum by magazines. If the air slams into the back of the valve, you closed it too soon .

It originates at valve open due to the shock of the pressure differential, between the port and cylinder as the piston starts to descend and creates a depression. An expansion wave occurs from that shock, travels up from the valve and is reflected at the end of the runner (or until a large change of area, which may be a plenum or atmosphere) and reflected back towards the cylinder as a compression wave. Ideally, you time the strongest of the reflection compression wave just as the valve is closing and pressure differential between the cylinder and intake port are approaching being equal, to get that last little bit in. Obviously, packaging has to be considered, so this is why it is usually something like the 3rd harmonic, and obviously it's going to change over rpm.

It is also contingent on air density. Honda with their V8 looked at the speed of the waves in the intake port (measureable) to look at fuel saturation / atomizing from different injector designs. Obviously the faster the waves move, the denser / cooler the air. IF Mercedes is doing what is being guessed at, which I don't think they are - at least not as described, then their tuned lengths will change from when the air is supposedly super cooled vs "normal" operating inlet temperatures.

For a layman's perspective, this is partly why single 4bbl carbureted engines with wet manifolds work super well over something like DCOE Webers. The wet plenum allows waves to reflect uninhibited, dampens poorly timed waves to the carburetor booster (which minimizes reversion / double fueling and helps with the fuel curve), and the wet plenum allows atomization to occur. A DCOE Weber (or the like) is much more sensitive because the full force of the expansion / and reflected wave is felt by the carburetor (and carburetors work off pressure differentials) and then the entire choke / booster creates a change in area and has its own microreflections / dampens the strength of the compression wave. The only issue with the single 4bbl carburetor type manifold is variability in runner length.. which isn't always a bad thing, but gets worst with growing bore size / centerlines. This is the only positive of a carbureted engine, is they are sensitive to pressure waves due to the function of a carburetor (call it "jet signal") so you become aware of these things.

Plenums can also have the effect of going into resonance, if sized / shaped properly to create just a hair more charging pressure. I think Honda covers that in one of their white papers.

Edit:
Development of Induction and Exhaust Systems for Third-Era Honda Formula One Engines
http://www.f1-forecast.com/pdf/F1-Files ... P2_10e.pdf

These are not cutting edge ideas / understanding, but this is a nice way to show it in practice.

Thank you! Digesting all of the information!

[Jolle]

I think this part is probably a lot more likely.


Here is a several year old video that give greater detail. It also goes into altering the size of the runner. In the video using a valve is discussed, but I'm betting that's against the technical rules (haven't checked though). I bet a fluidic switch would be legal.

Good video, thanks!

Variable length manifolds have been around for just about forever. The Mazda R26B had them, all the Superbikes have had them in the EFI era, Formula 1 had them until the mid 2000s (rules), Ferrari street engines, Miata's, BMW's, etc. a friend even made them for an early Chrysler Hemi for the Popular Hot Rodding Engine Masters Challenge:
https://img-s-msn-com.akamaized.net/ten ... =t&o=f&l=f

Here is a graph I made within a 1D pressure dynamics engine simulator for a project I was working on. Going through a torque shaping exercise for a friend's race engine at the moment, which relies on similar things:
https://live.staticflickr.com/65535/514 ... 376d_c.jpg

Ferrari had a system that actually slid in and out over the rpm range to capture those waves and keep it above a fixed inlet length baseline. Not as simple as just making it shorter as rpm's grow.

The Engineering Explained videos are okay for the casual viewer. It all gets a lot more complex than that and he does get some things incorrect. The pressure waves in the intake port / manifold does not occur due to the air column "slamming into the back of the valve"... the origins for that, that I can find are in Superflow's manual and repeated adnaseum by magazines. If the air slams into the back of the valve, you closed it too soon .

It originates at valve open due to the shock of the pressure differential, between the port and cylinder as the piston starts to descend and creates a depression. An expansion wave occurs from that shock, travels up from the valve and is reflected at the end of the runner (or until a large change of area, which may be a plenum or atmosphere) and reflected back towards the cylinder as a compression wave. Ideally, you time the strongest of the reflection compression wave just as the valve is closing and pressure differential between the cylinder and intake port are approaching being equal, to get that last little bit in. Obviously, packaging has to be considered, so this is why it is usually something like the 3rd harmonic, and obviously it's going to change over rpm.

It is also contingent on air density. Honda with their V8 looked at the speed of the waves in the intake port (measureable) to look at fuel saturation / atomizing from different injector designs. Obviously the faster the waves move, the denser / cooler the air. IF Mercedes is doing what is being guessed at, which I don't think they are - at least not as described, then their tuned lengths will change from when the air is supposedly super cooled vs "normal" operating inlet temperatures.

For a layman's perspective, this is partly why single 4bbl carbureted engines with wet manifolds work super well over something like DCOE Webers. The wet plenum allows waves to reflect uninhibited, dampens poorly timed waves to the carburetor booster (which minimizes reversion / double fueling and helps with the fuel curve), and the wet plenum allows atomization to occur. A DCOE Weber (or the like) is much more sensitive because the full force of the expansion / and reflected wave is felt by the carburetor (and carburetors work off pressure differentials) and then the entire choke / booster creates a change in area and has its own microreflections / dampens the strength of the compression wave. The only issue with the single 4bbl carburetor type manifold is variability in runner length.. which isn't always a bad thing, but gets worst with growing bore size / centerlines. This is the only positive of a carbureted engine, is they are sensitive to pressure waves due to the function of a carburetor (call it "jet signal") so you become aware of these things.

Plenums can also have the effect of going into resonance, if sized / shaped properly to create just a hair more charging pressure. I think Honda covers that in one of their white papers.

Edit:
Development of Induction and Exhaust Systems for Third-Era Honda Formula One Engines
http://www.f1-forecast.com/pdf/F1-Files ... P2_10e.pdf

These are not cutting edge ideas / understanding, but this is a nice way to show it in practice.

Plus, since 2015 all F1 cars have variable inlets again powered by the onboard hydraulics. It’s also just air, not a fuel mixture like other applications could do, because of the direct injection (in the V8 days they had shower jets above the trumpets).
For today’s fuel limited engines, there are only very small gains for colder intake air. Not only isn’t all the oxygen used to burn the fuel, it would only result in needing a bit less boost pressure. As suggested this would be a boost for qualifying laps, the same could be gained by running the H unit a bit harder.

[gruntguru]

. . . . The pressure waves in the intake port / manifold does not occur due to the air column "slamming into the back of the valve"... the origins for that, that I can find are in Superflow's manual and repeated adnaseum by magazines. If the air slams into the back of the valve, you closed it too soon .

It originates at valve open due to the shock of the pressure differential, between the port and cylinder as the piston starts to descend and creates a depression. . . .

There is also the phenomenon of exhaust tuning (pulses are much stronger in the exhaust runners due to the high cylinder pressure at EVO). Race engine manifolds are tuned so that a strong negative pressure wave arrives at the exhaust valve just before it closes. This pulls the last bit of exhaust out of the cylinder at TDC and sends a negative wave out the intake valve (which is also open at TDC). This is the main driver of waves in the intake manifold.