The Official 2017 Spanish Grand Prix Thread **Spoilers**
Discussion
Hungrymc said:
Catatafish said:
It's not impossible at all. This kind of daft, elaborate aero is a direct symptom of the state of computer modelling. You don't need supercomputers, although that speeds things up. CFD tools are widely available and usable on a fairly average, recent pc.
But it must get ever harder to be accurate and get good correlation, the more complex, and the more interaction between more features there are? The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
Merc suspension arm:
This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
Otispunkmeyer said:
You can be as complex as you like.... If you have a normal PC it will just take a while. We used to do CFD at uni on in cylinder swirl and tumble, we'd leave the computers running overnight to finish. If the solver converged on an answer then great if it it didn't it was a whole day gone! Lol
The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
It's nothing now to do with the lack of computing power. I'm sure the teams use GPUs and FPGAs and I'm aware of an Intel development in my field that will massively increase performance on the CPU. This means that there is no limit to the simulations that can be run with as many variables as the teams choose. The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
Then there's the use of statistical analyses.
From memory, I noticed McLaren shortlisted and I think winning a BI industry prize for their development of real-time analytical tools. And has anyone noticed the Tibco decal on the front of the Mercedes drivers' helmets. If that doesn't come with a technological input I'd be very surprised indeed.
Otispunkmeyer said:
You can be as complex as you like.... If you have a normal PC it will just take a while. We used to do CFD at uni on in cylinder swirl and tumble, we'd leave the computers running overnight to finish. If the solver converged on an answer then great if it it didn't it was a whole day gone! Lol
The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
That's kind of my point, it's one thing to have a good analysis of the flow around these features, but scale models are very risky (suddenly resonance and flex changes - whole system changes) and ambient conditions. You get slightly different performance from an element at the front of the chain and the whole system is compromised.The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
If it were simple and automatically accurate, there would be no difference between the teams. It's the fine margins, the variable situations where we see differences emerging. I'd be surprised if that isn't partly because they have different algorithms and more reliable correlation in there models?
ash73 said:
Just tuned in to P3 and I think I heard Crofty say there was a problem with Bottas' engine so he's reverted to the lower power spec for the rest of the weekend; Lewis now a shoo-in for pole and this thread will hit 50 pages when his engine grenades mid race.
Bottas's car had an electrical problem, the mechanics spent all night stripping the car to bits and rebuilding it. They elected to fit an entire new engine. When they started it up the engine had a problem, so they had to take it apart again and refit the original engine he's used thus far.Not sure what engine Lewis is using.
Hungrymc said:
Otispunkmeyer said:
You can be as complex as you like.... If you have a normal PC it will just take a while. We used to do CFD at uni on in cylinder swirl and tumble, we'd leave the computers running overnight to finish. If the solver converged on an answer then great if it it didn't it was a whole day gone! Lol
The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
That's kind of my point, it's one thing to have a good analysis of the flow around these features, but scale models are very risky (suddenly resonance and flex changes - whole system changes) and ambient conditions. You get slightly different performance from an element at the front of the chain and the whole system is compromised.The major thing with CFD is, it's just pretty pictures until you have something real world to validate it against. Not necessarily a full real world model mind (else there was no need to do the CFD), but you can do scale models or just models of certain sections to help you validate the flows. For a lot of stuff there's plenty of literature to validate against.
If it were simple and automatically accurate, there would be no difference between the teams. It's the fine margins, the variable situations where we see differences emerging. I'd be surprised if that isn't partly because they have different algorithms and more reliable correlation in there models?
ash73 said:
Crafty_ said:
Bottas's car had an electrical problem, the mechanics spent all night stripping the car to bits and rebuilding it. They elected to fit an entire new engine. When they started it up the engine had a problem, so they had to take it apart again and refit the original engine he's used thus far.
Not sure what engine Lewis is using.
On the commentary they said Merc refitted a used engine in Bottas' car, the one he used in Russia where they had cooling problems. Lewis is still using the new spec higher power engine.Not sure what engine Lewis is using.
Lauda just confirmed that the two engines in terms of power output are exactly the same, the updates have been for reliability.
Crafty_ said:
Merc suspension arm:
This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
It's a beautiful bit of engineering. It is impressive, almost awe inspiring. The fear is that it is killing F1, bit by bit. This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
I'm in a quandary. I don't want a one-make series. In fact I think that will kill F1 for me. On the other hand the image above shows why teams need the backing of a massive constructor or multinational conglomerate in order to compete. In the case of Honda, even that is not enough. There must be some middle line, not necessarily in the middle of course, that will either satisfy both extremes or at least not irritate them too much.
Too many Sochies will turn fans away.
Crafty_ said:
ash73 said:
Crafty_ said:
Bottas's car had an electrical problem, the mechanics spent all night stripping the car to bits and rebuilding it. They elected to fit an entire new engine. When they started it up the engine had a problem, so they had to take it apart again and refit the original engine he's used thus far.
Not sure what engine Lewis is using.
On the commentary they said Merc refitted a used engine in Bottas' car, the one he used in Russia where they had cooling problems. Lewis is still using the new spec higher power engine.Not sure what engine Lewis is using.
Lauda just confirmed that the two engines in terms of power output are exactly the same, the updates have been for reliability.
I very much doubt Merc would introduce a new spec ICE/PU with no performance increase. A couple of years back when they introduced a new spec for 'reliability' and then proceeded to put some gap between themselves and the rest of the field. Could be Monza 2015? Was it then that they'd introduced TJI? Rumours? Rosberg suffered an engine failure the same weekend too, running on the old spec engine, IIRC.
Lewis would be the only one running the new spec this weekend, no customers have got it incidentally.
Dr Z said:
Lauda would say that, wouldn't he?
I very much doubt Merc would introduce a new spec ICE/PU with no performance increase. A couple of years back when they introduced a new spec for 'reliability' and then proceeded to put some gap between themselves and the rest of the field. Could be Monza 2015? Was it then that they'd introduced TJI? Rumours? Rosberg suffered an engine failure the same weekend too, running on the old spec engine, IIRC.
Lewis would be the only one running the new spec this weekend, no customers have got it incidentally.
Why does everything have to turn in to a conspiracy I very much doubt Merc would introduce a new spec ICE/PU with no performance increase. A couple of years back when they introduced a new spec for 'reliability' and then proceeded to put some gap between themselves and the rest of the field. Could be Monza 2015? Was it then that they'd introduced TJI? Rumours? Rosberg suffered an engine failure the same weekend too, running on the old spec engine, IIRC.
Lewis would be the only one running the new spec this weekend, no customers have got it incidentally.
Crafty_ said:
Why does everything have to turn in to a conspiracy
No conspiracy, managing expectations. 
It's alright, I'm sure no malice was intended from the team.
I'd like some competition at the front, and if Merc dominate this race then it's a relatively easy win for Hamilton...same as last year when there were some issues for Hamilton and Rosberg had a clean run for a few races at the front.
Shame that Vettel too is having issues.
Crafty_ said:
Merc suspension arm:
[img]https://i.redditmedia]
This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
Yes a complete was of time effort and resource have said for year the cars should have standardised wings choice of 3 front and rear, same for every team stop pissing around wasting money on stupid winglets etc.[img]https://i.redditmedia]
This sums up F1 to me. Someone has wondered about just a touch more downforce / air management in this area. Figured out a winglet design, prototyped it, tested it in CFD/wind tunnel, refined it, finally its been produced and put on the car.
Blows my mind the attention to detail and effort put in to something that can surely only bring such a small benefit. The numbers of hours and cost must be immense for the benefit.
And yet, it gets done.
At the same time the neverending pursuit of perfection means enormous cost
F1 in a nutshell.
Gassing Station | Formula 1 | Top of Page | What's New | My Stuff