F1 Teams Simulators - how useful are they?

[Woody]

In recent weeks Ferrari's Luca Di Montezemelo has had a pop at the simulators and has called for proper testing to be re-introduced.

How useful/acurate are these simulators to the teams?

I'm guessing a lot of it is down to the programming and accuracy of data input?

What do they consist of/look like?

Thanks

Chris

[joewilliams]

I'm going to stick my neck out and guess that his doesn't work as well as Red Bull's.

[uktrailmonster]

In my personal experience of the Jaguar and Williams F1 car simulators, not that useful in the overall scheme of things. Great for generating PR bullst, but they don't really make a significant contribution to making the cars go faster.

[Jasandjules]

It may be useful for things like learning a track but I can't honestly believe that they can possibly be close to actually being in the car, feeling how she responds to throttle/steering input etc.

[supertouring]

I was watching the Williams Superfactory program the other day and it showed a front wing change going from initial concept, through design, cfd, wind-tunnel, simulator and production (may be wrong order).

One question I had was, ok the design is all in the computer, but how easy is it to take a design and plug it into the simulator and then verify any results?

Would the driver notice the difference or does the simulator simply assist in gathering data for analysis?

I can understand the PR stuff, but I hear test drivers constantly talking about "all day on the simulator" so it can't all be PR can it?

[Vaud]

I'm going to take a guess that the good ones are good for:

Letting a driver practice with all of the various wheel stetting based on direction from the pits, in a high demand environment (noise, light, some movement, etc)
Simulating (roughly) tyre wear and pitstops
Providing approximations of fuel load strategies (given you can mimic early throttle off, etc)
Scenario modelling for various downforce settings
Scenario modelling for various throttle maps

Anything else?

[Graham]

no use what so ever thats why all the teams spend millions on them


the physics models are very advanced, and the circuits laser mapped. they have full tyre models, and all the cfd and load info for all the aero parts etc. The graphics arnt amazing but they dont need to be IIRC they are based on an advanced version of the rfactor software..

[allegerita]

A driving simulator only serves a limited amount of functions. A simulator is always a performance trade off and should be used as such. Before you invest in a simulator you need to consider what you want to achieve with the device. Any results should be treated in the light of those considerations.

Some simulators are usefull for track familiarisation, other are a development tool in aero/ suspension/ ergonomics optimisation. I have seen sims for model valiadation. Some simulators are pure hardware-in-the-loop, other have a driver controlling the device and providing input. And sometimes they are PR things.

As a result simulators can look like anything from a simple desktop device to a full-blown, large multi-axis dynamic device bigger than a full flight simulator.

The driving sims I have built (for F1, fast racing cars and road cars) were all very usefull.

[MocMocaMoc]

I remember Martin Brundle asked Jenson Button how accurate the McL sim was - and if it simulated the tyres accurately. Button quickly said 'yes' but everything about it suggested he was lying - call it body language, call it a panic look on his face or my woman's intuition (I'm a dude) but there's no way that sim went into that much detail.

[MocMocaMoc]

I remember Martin Brundle asked Jenson Button how accurate the McL sim was - and if it simulated the tyres accurately. Button quickly said 'yes' but everything about it suggested he was lying - call it body language, call it a panic look on his face or my woman's intuition (I'm a dude) but there's no way that sim went into that much detail.

[Vaud]

MocMocaMoc said:

I remember Martin Brundle asked Jenson Button how accurate the McL sim was - and if it simulated the tyres accurately. Button quickly said 'yes' but everything about it suggested he was lying - call it body language, call it a panic look on his face or my woman's intuition (I'm a dude) but there's no way that sim went into that much detail.

Well there is accurate and accurate. Does it technically mimic tyre degradation? Possibly. But can any simulator give a highly aware, highly trained driver a suitably comparable experience (g loadings, the vehicle dynamics through vibration, the body, etc) that is is for a driver, always going to a massive compromise.

I think they are probably very useful for new tracks, mildly useful for simulations and not a true comparison for real testing - but then you can do far more in the simulator than you ever could at the track, even at the peak of testing.

I think Alonso told Ferrari to hire PDLR as the best test/simulator guy, if I recall correctly?

[eps]

Graham said:

no use what so ever thats why all the teams spend millions on them


the physics models are very advanced, and the circuits laser mapped. they have full tyre models, and all the cfd and load info for all the aero parts etc. The graphics arnt amazing but they dont need to be IIRC they are based on an advanced version of the rfactor software..

They spend millions on them, because they can't do proper testing. If you gave them the choice, we all know which one they'd choose.

Saying that, if it gives them even .1 of a second improvement on a lap time it may well have been worth it.

[P-Jay]

obviously they're only as useful as they are accurate - Ferrari spent most of last year and maybe the one before crying that they couldn't get their simulator to mirror the real world so it's little wonder they're not so keen.

Personally I think they're crap, or rather the need for them is - after-all the core reason for banning testing was to force the teams to cut costs, not to level any playing field or for the benefit of making the series 'fairer', but in the never ending fight for that "tenth a lap" advantage the teams put all their efforts into simulators and fluid dynamics to mitigate the lost real world testing so the rule has failed to achieve it's goal.

So for the teams it doesn't save them money, for F1 / FIA it hasn't reduced the price of entry for new / small teams and hasn't produced a better spectacle for fans - in fact an open test day was a very cheap way for fans to see F1 cars in the flesh.

[allegerita]

P-Jay said:

.......Ferrari spent most of last year and maybe the one before crying that they couldn't get their simulator to mirror the real world so it's little wonder they're not so keen.

There is a reason for that.

Ferrari decided to build the sim themselves, with only limited support from knowledgeable people. They should have sticked to what they are good at: making racing cars. Equivalent to the flight sim industry: Boeing, Airbus, Lockheed etc could never build a proper flight sims. They build perfect planes, yes, but sims is an entirely different ballgame.

[uktrailmonster]

Vaud said:

I'm going to take a guess that the good ones are good for:

Letting a driver practice with all of the various wheel stetting based on direction from the pits, in a high demand environment (noise, light, some movement, etc)
Simulating (roughly) tyre wear and pitstops
Providing approximations of fuel load strategies (given you can mimic early throttle off, etc)
Scenario modelling for various downforce settings
Scenario modelling for various throttle maps

Anything else?

In my experience they are crap at simulating tyre wear. The tyre model is usually the least impressive part of an F1 sim. Tyres are notoriously difficult to model accurately due to the complex interactions of many parameters e.g. tyre surface temp, tyre bulk temp, air temp, pressure, track surface, track temp, humidity, compound, etc, etc. The tyre model invariably ends up as a relatively crude bodge, even when millions are spent.

Driver operated sims are most useful for learning new tracks and cockpit controls and procedures. Non-driver sims i.e. automated sim programs are useful for optimising aero downforce levels, ride heights, gear ratios and pitstop strategy. They provide a reasonable starting point for the trackside, but nothing more.

[uktrailmonster]

Graham said:

no use what so ever thats why all the teams spend millions on them

F1 teams spend millions on all kinds of things that give a poor return on investment. Driver operated sims are a great example. Not totally useless, but nowhere near as useful as you might think.

[carinaman]

Would proper testing mean newcomers get more mileage under their belts and therefore are less crash happy when they get onto the F1 grid?

Can the simulators simulate a catastrophic Pirelli tyre failure?

[angrymoby]

uktrailmonster said:

In my personal experience of the Jaguar and Williams F1 car simulators, not that useful in the overall scheme of things. Great for generating PR bullst, but they don't really make a significant contribution to making the cars go faster.

Your post suggests this was a while ago (Jaguar were in F1 from 2000- '04 weren't they?) ...if so, the last 10 years has seen massive software/ hardware advances.

& someone on here has first hand experience of a number of current F1 teams simulators ...i'm sure they'll be along shortly to comment (NDA's allowing) ...especially on how advanced RBR's is!

[uktrailmonster]

angrymoby said:

Your post suggests this was a while ago (Jaguar were in F1 from 2000- '04 weren't they?) ...if so, the last 10 years has seen massive software/ hardware advances.

& someone on here has first hand experience of a number of current F1 teams simulators ...i'm sure they'll be along shortly to comment (NDA's allowing) ...especially on how advanced RBR's is!

Well 2009 was my last experience as an F1 chief engineer and I still have many close friends actively involved. I'm only passing on my personal experience as a direct end user of such systems, believe what you like.

The problem with these threads is that most of the info is second or third hand at best. I'm fortunate enough to have worked in the industry at the sharp end in a relatively senior position for over a decade and thought I would share a little insight. But every time I do I seem to get corrected by people who in all probability have no personal experience of F1 engineering at all. Hey ho

[TITWONK]

Your posts are appreciated matey I for one absolutely love reading what you guys in the industry have to say