Mercedes W13

A place to discuss the characteristics of the cars in Formula One, both current as well as historical. Laptimes, driver worshipping and team chatter do not belong here.
Post Reply
Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

Vanja #66 wrote:
25 Nov 2022, 21:17
Hoffman900 wrote:
25 Nov 2022, 21:07
Disagree Vanja, and I base that on talking to a PhD aero friend and hearing Peter Wright, James Allison, and Jean-Claude Migeot’s take. Lewis Hamilton also parroted the sidewall issue which would have come from his engineers. James Allison specifically called out the stall theory as wrong.

If stall caused this, the collapsing rear suspensions of years pass would have also caused this, and they didn’t.
So what's the mechanics of a car bouncing at high speed without oscilating levels of downforce?
The mechanism isn’t stall. It’s the car literally bottoming out and bouncing against the plank.

I listed three F1 aerodynamicists who have been / are involved in the design of tunnel cars. I’m not sure why else you guys want, the current one specifically said the stall theory is wrong.
Last edited by Hoffman900 on 25 Nov 2022, 21:22, edited 1 time in total.

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:16
AR3-GP wrote:
25 Nov 2022, 21:14
Hoffman900 wrote:
25 Nov 2022, 21:07


Disagree Vanja, and I base that on talking to a PhD aero friend and hearing Peter Wright, James Allison, and Jean-Claude Migeot’s take. Lewis Hamilton also parroted the sidewall issue which would have come from his engineers.

If stall caused this, the collapsing rear suspensions of years pass would have also caused this, and they didn’t.
All cars have the same tires and some of them didn't porpoise. Yes, the tire sidewall has deflection under load, but sidewall deflection is simply a lowering of the car, triggering the aero induced porpoising, which is a matter of a poor approach to a ground effect floor design.
It has to do with peak downforce. Merc was making so much that a stiffer rear suspension pushed the loads into the sidewall and over powered it. This is a big part of why this didn’t scale to the wind tunnel model and the CFD didn’t pick it up because they don’t have accurate or the ability to use accurate tire models (tires are incredibly hard to model) and why their monster of a downforce making design failed in the real world.
okay, so then what about porpoising on cars like the Alpha Tauri, Alfa Romeo, Haas, or Aston Martin?

Where is all this downforce coming from that made those cars "overpower" their tires?

Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

AR3-GP wrote:
25 Nov 2022, 21:20
Hoffman900 wrote:
25 Nov 2022, 21:16
AR3-GP wrote:
25 Nov 2022, 21:14


All cars have the same tires and some of them didn't porpoise. Yes, the tire sidewall has deflection under load, but sidewall deflection is simply a lowering of the car, triggering the aero induced porpoising, which is a matter of a poor approach to a ground effect floor design.
It has to do with peak downforce. Merc was making so much that a stiffer rear suspension pushed the loads into the sidewall and over powered it. This is a big part of why this didn’t scale to the wind tunnel model and the CFD didn’t pick it up because they don’t have accurate or the ability to use accurate tire models (tires are incredibly hard to model) and why their monster of a downforce making design failed in the real world.
okay, so then what about porpoising on cars like the Alpha Tauri, Alfa Romeo, Haas, or Aston Martin?

Where is all this downforce coming from that made those cars "overpower" their tires?
The design and von karman vortex street flow separation inducing large time sensitive peaks in load. That’s not stall / choke flow.

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:23
AR3-GP wrote:
25 Nov 2022, 21:20
Hoffman900 wrote:
25 Nov 2022, 21:16


It has to do with peak downforce. Merc was making so much that a stiffer rear suspension pushed the loads into the sidewall and over powered it. This is a big part of why this didn’t scale to the wind tunnel model and the CFD didn’t pick it up because they don’t have accurate or the ability to use accurate tire models (tires are incredibly hard to model) and why their monster of a downforce making design failed in the real world.
okay, so then what about porpoising on cars like the Alpha Tauri, Alfa Romeo, Haas, or Aston Martin?

Where is all this downforce coming from that made those cars "overpower" their tires?
The design and von karman vortex street flow separation inducing large time sensitive peaks in load. That’s not stall / choke flow.
Adrian Newey:
“We knew it was a potential problem. The LMP cars had it for a very long time. It’s a very well-known problem. If you have an aero map which as you get closer to the ground generates more downforce eventually the flow structure breaks down and loses downforce, then it’s going to porpoise. With these regs you could see that was a possibility but whether they would and how you model that, was the difficulty.

“It was a bit of using experience as to what the causes of porpoising might be and trying to be mindful of that but at the same time we didn’t find a way of modelling it properly. In principle, you could do it in the windtunnel. There’s a thing called Strouhal number which is a bit like a Reynolds number, so it takes the speed and the size of the real thing, then applies a scaling factor based on speed and size.

“It’s much more aggressive than Reynolds number in that these cars are bouncing along at let’s say 6Hz then the frequency you have to achieve on a 60% model at 60 metres/second is very high. If you completely redesigned your model and beefed up everything and accepted less fidelity in the balance you might get there but it would be a big undertaking.”
https://the-race.com/formula-1/newey-in ... etirement/

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:23
AR3-GP wrote:
25 Nov 2022, 21:20
Hoffman900 wrote:
25 Nov 2022, 21:16


It has to do with peak downforce. Merc was making so much that a stiffer rear suspension pushed the loads into the sidewall and over powered it. This is a big part of why this didn’t scale to the wind tunnel model and the CFD didn’t pick it up because they don’t have accurate or the ability to use accurate tire models (tires are incredibly hard to model) and why their monster of a downforce making design failed in the real world.
okay, so then what about porpoising on cars like the Alpha Tauri, Alfa Romeo, Haas, or Aston Martin?

Where is all this downforce coming from that made those cars "overpower" their tires?
The design and von karman vortex street flow separation inducing large time sensitive peaks in load. That’s not stall / choke flow.
I'm struggling to follow you.

First you said it was the tire sidewall.
It has to do with peak downforce. Merc was making so much that a stiffer rear suspension pushed the loads into the sidewall and over powered it.
18" wheel has a smaller sidewall. It's much stiffer than last year making it even easier to predict.


Then you said it was because the car was bottoming and bouncing on the plank (as if we have not seen teams in the last 5 years dragging their plank all along the track sparking...).
The mechanism isn’t stall. It’s the car literally bottoming out and bouncing against the plank.
Now you say it's the Vortex street flow separation.

Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

AR3-GP wrote:
25 Nov 2022, 21:26
Hoffman900 wrote:
25 Nov 2022, 21:23
AR3-GP wrote:
25 Nov 2022, 21:20


okay, so then what about porpoising on cars like the Alpha Tauri, Alfa Romeo, Haas, or Aston Martin?

Where is all this downforce coming from that made those cars "overpower" their tires?
The design and von karman vortex street flow separation inducing large time sensitive peaks in load. That’s not stall / choke flow.
Adrian Newey:
“We knew it was a potential problem. The LMP cars had it for a very long time. It’s a very well-known problem. If you have an aero map which as you get closer to the ground generates more downforce eventually the flow structure breaks down and loses downforce, then it’s going to porpoise. With these regs you could see that was a possibility but whether they would and how you model that, was the difficulty.

“It was a bit of using experience as to what the causes of porpoising might be and trying to be mindful of that but at the same time we didn’t find a way of modelling it properly. In principle, you could do it in the windtunnel. There’s a thing called Strouhal number which is a bit like a Reynolds number, so it takes the speed and the size of the real thing, then applies a scaling factor based on speed and size.

“It’s much more aggressive than Reynolds number in that these cars are bouncing along at let’s say 6Hz then the frequency you have to achieve on a 60% model at 60 metres/second is very high. If you completely redesigned your model and beefed up everything and accepted less fidelity in the balance you might get there but it would be a big undertaking.”
https://the-race.com/formula-1/newey-in ... etirement/
James Allison:


Peter Wright:


Jean-Claude Migeot:

https://us.motorsport.com/f1/news/f1-te ... 98443/amp/


https://www.racecar-engineering.com/new ... ion-model/

My guess the original Merc designs had a high peak downforce, where the RB Cl map is a lot flatter over varying ride heights.
Last edited by Hoffman900 on 25 Nov 2022, 21:36, edited 1 time in total.

User avatar
Vanja #66
1329
Joined: 19 Mar 2012, 16:38
Contact:

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:20
The mechanism isn’t stall. It’s the car literally bottoming out and bouncing against the plank.

I listed three F1 aerodynamicists who have been / are involved in the design of tunnel cars. I’m not sure why else you guys want, the current one specifically said the stall theory is wrong.
So why doesn't a car with increasing downforce (as speed increases) keep pressing the plank instead of bouncing? RB18 did this on a few ocassions, minimal bouncing and very far from as violent as early W13 bouncing.
And they call it a stall. A STALL!

#Aerogimli
#DwarvesAreNaturalSprinters
#BlessYouLaddie

Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

Vanja #66 wrote:
25 Nov 2022, 21:33
Hoffman900 wrote:
25 Nov 2022, 21:20
The mechanism isn’t stall. It’s the car literally bottoming out and bouncing against the plank.

I listed three F1 aerodynamicists who have been / are involved in the design of tunnel cars. I’m not sure why else you guys want, the current one specifically said the stall theory is wrong.
So why doesn't a car with increasing downforce (as speed increases) keep pressing the plank instead of bouncing? RB18 did this on a few ocassions, minimal bouncing and very far from as violent as early W13 bouncing.
It would depend on the on the peakiness of the Cl map and thus the vertical velocity component of that ramp. None of us have that data and anything you would see publicly would be extremely rudimentary.

If we want to talk flow of the underfloors, these two are excellent papers:
https://arc.aiaa.org/doi/10.2514/2.1997

https://viterbik12.usc.edu/wp-content/u ... namics.pdf

You can also see the modern floor strake design is over twenty years old now.

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:33
AR3-GP wrote:
25 Nov 2022, 21:26
Hoffman900 wrote:
25 Nov 2022, 21:23


The design and von karman vortex street flow separation inducing large time sensitive peaks in load. That’s not stall / choke flow.
Adrian Newey:
“We knew it was a potential problem. The LMP cars had it for a very long time. It’s a very well-known problem. If you have an aero map which as you get closer to the ground generates more downforce eventually the flow structure breaks down and loses downforce, then it’s going to porpoise. With these regs you could see that was a possibility but whether they would and how you model that, was the difficulty.

“It was a bit of using experience as to what the causes of porpoising might be and trying to be mindful of that but at the same time we didn’t find a way of modelling it properly. In principle, you could do it in the windtunnel. There’s a thing called Strouhal number which is a bit like a Reynolds number, so it takes the speed and the size of the real thing, then applies a scaling factor based on speed and size.

“It’s much more aggressive than Reynolds number in that these cars are bouncing along at let’s say 6Hz then the frequency you have to achieve on a 60% model at 60 metres/second is very high. If you completely redesigned your model and beefed up everything and accepted less fidelity in the balance you might get there but it would be a big undertaking.”
https://the-race.com/formula-1/newey-in ... etirement/
James Allison:


Peter Wright:


Jean-Claude Migeot:

https://us.motorsport.com/f1/news/f1-te ... 98443/amp/


https://www.racecar-engineering.com/new ... ion-model/

My guess the original Merc designs had a high peak downforce, where the RB Cl map is a lot flatter over varying ride heights.
I think that you are needlessly overthinking/overcomplicating the matter.

First you said it was the tire sidewall. Then you said it was the "peak downforce" (while ignoring all of the backmarkers that were porpoising, while being much much slower than Mercedes and even Red Bull). Then you said the porpoising was because the car was literally just slamming onto it's plank and rebounding (ignoring that we have had cars with even more downforce riding into their planks for years). Then you started talking about vortex streets.

At the end of the day, everyone that you linked either had no idea what they were doing judging by the porpoising of the car they had a hand in (Allison), or is not a current F1 engineer (Migeot. Peter Wright).

Don't you think the guy who built the car that wasn't porpoising and claims to have predicted it, has the best understanding?
Last edited by AR3-GP on 25 Nov 2022, 21:50, edited 1 time in total.

User avatar
Vanja #66
1329
Joined: 19 Mar 2012, 16:38
Contact:

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 21:36
It would depend on the on the peakiness of the Cl map and thus the vertical velocity component of that ramp. None of us have that data and anything you would see publicly would be extremely rudimentary.
Yes, it would depend on the -C_l levels, the higher the downforce the bigger the damping of the rebound, ie the faster it would stick down, stop osciallating and remain level.

Not sure if you heard the very beginning of Wright's explanation, he clearly refers to mechanical movement of the car (heave and pitch) as reasons for varying aerodynamic forces within flutter comparison. Later he mentions stall "somewhere" on the car as addittional reason, which is clear as front wing stall e.g. would cause all kinda of trouble for floor downforce.
And they call it a stall. A STALL!

#Aerogimli
#DwarvesAreNaturalSprinters
#BlessYouLaddie

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

I also think that it's worth mentioning that while Migeot is correct that flutter of aero surfaces can induce a forcing function that would promote a bouncing mode of the sprung mass, Mercedes used extensive floor stay treatments and it never resolved their problem. Flutter of aero surfaces wouldn't seem to be the problem in the W13 (nor was it really the issue with any other team. It CAN be an issue, but given we saw many teams stiffen their floors, it didn't appear to be the dominant issue.) .

Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

Vanja #66 wrote:
25 Nov 2022, 21:49
Hoffman900 wrote:
25 Nov 2022, 21:36
It would depend on the on the peakiness of the Cl map and thus the vertical velocity component of that ramp. None of us have that data and anything you would see publicly would be extremely rudimentary.
Yes, it would depend on the -C_l levels, the higher the downforce the bigger the damping of the rebound, ie the faster it would stick down, stop osciallating and remain level.

Not sure if you heard the very beginning of Wright's explanation, he clearly refers to mechanical movement of the car (heave and pitch) as reasons for varying aerodynamic forces within flutter comparison. Later he mentions stall "somewhere" on the car as addittional reason, which is clear as front wing stall e.g. would cause all kinda of trouble for floor downforce.
Well the issue is when it bounces off the floor from bottoming out, you've now introduced the Von Karmen Vortex street component (ie: flutter) and it can be self perpetuating. Model the car bouncing hard off the floor. This is the same reason hitting a bump for Mercedes would also introduce it.

A PhD Aerodynamicist friends who worked for one of the major aerospace companies, in a very high up level, gave me examples of this, but I can't share them.

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 22:05
Vanja #66 wrote:
25 Nov 2022, 21:49
Hoffman900 wrote:
25 Nov 2022, 21:36
It would depend on the on the peakiness of the Cl map and thus the vertical velocity component of that ramp. None of us have that data and anything you would see publicly would be extremely rudimentary.
Yes, it would depend on the -C_l levels, the higher the downforce the bigger the damping of the rebound, ie the faster it would stick down, stop osciallating and remain level.

Not sure if you heard the very beginning of Wright's explanation, he clearly refers to mechanical movement of the car (heave and pitch) as reasons for varying aerodynamic forces within flutter comparison. Later he mentions stall "somewhere" on the car as addittional reason, which is clear as front wing stall e.g. would cause all kinda of trouble for floor downforce.
Well the issue is when it bounces off the floor from bottoming out, you've now introduced the Von Karmen Vortex street component (ie: flutter) and it can be self perpetuating. Model the car bouncing hard off the floor. This is the same reason hitting a bump for Mercedes would also introduce it.

A PhD Aerodynamicist friends who worked for one of the major aerospace companies, in a very high up level, gave me examples of this.
So someone who has no F1 experience then and is only speculating?

Hoffman900
163
Joined: 13 Oct 2019, 03:02

Re: Mercedes W13

Post

AR3-GP wrote:
25 Nov 2022, 22:06
Hoffman900 wrote:
25 Nov 2022, 22:05
Vanja #66 wrote:
25 Nov 2022, 21:49


Yes, it would depend on the -C_l levels, the higher the downforce the bigger the damping of the rebound, ie the faster it would stick down, stop osciallating and remain level.

Not sure if you heard the very beginning of Wright's explanation, he clearly refers to mechanical movement of the car (heave and pitch) as reasons for varying aerodynamic forces within flutter comparison. Later he mentions stall "somewhere" on the car as addittional reason, which is clear as front wing stall e.g. would cause all kinda of trouble for floor downforce.
Well the issue is when it bounces off the floor from bottoming out, you've now introduced the Von Karmen Vortex street component (ie: flutter) and it can be self perpetuating. Model the car bouncing hard off the floor. This is the same reason hitting a bump for Mercedes would also introduce it.

A PhD Aerodynamicist friends who worked for one of the major aerospace companies, in a very high up level, gave me examples of this.
So someone who has no F1 experience then?
Are we going to hang our hat on what someone who has no F1 experience, did some rudimentary Cfd work on a random forum says? Vanja is a smart guy, but I'm not sure where you're coming from piggy backing off something he said.

He worked on projects that had budgets of the entire F1 field combined and the issues if not resolved would have resulted in death. The physics are the same. :lol:

Likely the issue is Mercedes had a high peak downforce with a steep C_l curve, that as the car got sucked down, the velocity of it caused it bounce off the ground. This bouncing introduced von karmen vortex street components (ie: flutter) that it could never drive out of. Stiffening the car to keep it from bouncing off the floor only goes so far unless the forces involve overpower the sidewall, which they did. They're undampened. That isn't "stall" / "choke flow".

I forgot how pedantic this place is and how much it's not about learning but a **** measuring contest of purported knowledge.

AR3-GP
313
Joined: 06 Jul 2021, 01:22

Re: Mercedes W13

Post

Hoffman900 wrote:
25 Nov 2022, 22:09
AR3-GP wrote:
25 Nov 2022, 22:06
Hoffman900 wrote:
25 Nov 2022, 22:05


Well the issue is when it bounces off the floor from bottoming out, you've now introduced the Von Karmen Vortex street component (ie: flutter) and it can be self perpetuating. Model the car bouncing hard off the floor. This is the same reason hitting a bump for Mercedes would also introduce it.

A PhD Aerodynamicist friends who worked for one of the major aerospace companies, in a very high up level, gave me examples of this.
So someone who has no F1 experience then?
Are we going to hang our hat on what someone who has no F1 experience, did some rudimentary Cfd work on a random forum says?

He worked on projects that had budgets of the entire F1 field combined and the issues if not resolved would have resulted in death. The physics are the same. :lol:

Likely the issue is Mercedes had a high peak downforce with a steep C_l curve, that as the car got sucked down, the velocity of it caused it bounce off the ground. This bouncing introduced von karmen vortex street components (ie: flutter) that it could never drive out of. Stiffening the car to keep it from bouncing off the floor only goes so far unless the forces involve overpower the sidewall, which they did. They're undampened.

I forgot how pedantic this place is and how much it's not about learning but a **** measuring contest of purported knowledge.
None of this has to do with personal knowledge. I've only presented to you Adrian Newey's comment on the matter as well as pointed out your inconsistent explanations relating to tire sidewalls, then the car literally slamming to the ground and "rebounding" from the plank as the entire porpoising phenomena, the vortex streets. I am not a source of reference. I can only gather what other's have said and access their credibility.

When it comes down to it, your PHd friend has no race car experience while Adrian Newey just won the championship. James Allison did not get the W13 right therefore it stands to reason that his understanding is incorrect. The other two fellows do touch on plausible explanations (aero flutter is a realistic phenomena, because like anything, a forcing function applied to a sprung mass has the potential to cause it to bounce, but again it is not mentioned by Adrian, so one presumes it's only a tertiary concern, and not the primary mechanism which Migeot proposes).

To your credit, you do post a link to https://www.racecar-engineering.com/new ... ion-model/, which along with Kyle Engineers explanation, and Adrian Newey's comment do seem to all circulate around the cause of porpoising.

The porpoising is entirely related to the shape of the downforce curve as a function of ride height. All 3, 1 who built the car which won the championship, the 2nd who worked for Mercedes but left, and the third (some sim experts) all point to the same thing.
Last edited by AR3-GP on 25 Nov 2022, 22:19, edited 3 times in total.

Post Reply