Whats the cheapest route to performance convergence, Aero or PU development?

[Just_a_fan]

If you put the regulations to one side and asked where the best bang for buck would come from I would suggest you are in line with Bhall's statement of engine, tyres, aero. Although weight saving probably sneaks in there somewhere.

If you are looking at an open wheeled race car with downforce allowed by the rules, the best way to reduce lap times is to add downforce. It's that simple. At some point, however, you need power to drag that downforce around. Tyres, unless in an open supplier format, are a constant so in this case they're not relevant.

Put another way, if you were building a race car from scratch, you'd see what the engine was giving you and then look to get the maximum downforce that it could reasonably drag around. You'd then look to make the engine more powerful and the downforce more efficient, both within whatever rules you are working.

The reason that Manor is 5 seconds/lap slower than Mercedes but with a similar top speed is because they are much lower on downforce efficiency. They have downforce, lots of it in real terms, but just not as much as Mercedes and what they have is not as efficient. That's where the money is spent in aero - improving the L/D. You can do it a bit by improving downforce (the L) but you can also do it a bit by reducing the drag (the D). Manor and Mercedes likely have similar D but Mercedes has a bigger L.

[FoxHound]

Can you redo your math for PU suppliers only, I think that would actually interesting and relevant. A customer team really isnt included in the OP.

Customer teams make up for 65% of the grid, now that's interesting and relevant.

Besides, PU's are mandated by the FIA to be exact in supply. So differences between customer and manufacturer units will be as low as it has ever been in the history of the sport.
And the actual point of "cheapest route to performance convergence" is saving money.
Manufacturers do not have issue with saving money, customer teams do however.

So if Toyota/BMW/name want to come back and spend, they can. That pony could fart out a full 10 unicorns more than the next horse, they'd still need to sell their engine for 12 million in 2018.
And customer teams remain unaffected due to aero costs remaining exorbitantly high in relation to engines.

[SR71]

Can you redo your math for PU suppliers only, I think that would actually interesting and relevant. A customer team really isnt included in the OP.

Customer teams make up for 65% of the grid, now that's interesting and relevant.

Besides, PU's are mandated by the FIA to be exact in supply. So differences between customer and manufacturer units will be as low as it has ever been in the history of the sport.
And the actual point of "cheapest route to performance convergence" is saving money.
Manufacturers do not have issue with saving money, customer teams do however.

So if Toyota/BMW/name want to come back and spend, they can. That pony could fart out a full 10 unicorns more than the next horse, they'd still need to sell their engine for 12 million in 2018.
And customer teams remain unaffected due to aero costs remaining exorbitantly high in relation to engines.

Customer teams don't develop PUs, you're a smart guy so you know they don't matter in this hypothetical examination.

How would the customer teams do if customer PUs were banned?

Williams, force India, Sauber, probably Toro Rosso would fold immediately.

These teams can afford aero programs but cannot afford a competitive PU development program.

That's says it all really.

[FoxHound]

Customer teams don't develop PUs, you're a smart guy so you know they don't matter in this hypothetical examination.

How would the customer teams do if customer PUs were banned?

Williams, force India, Sauber, probably Toro Rosso would fold immediately.

These teams can afford aero programs but cannot afford a competitive PU development program.

That's says it all really.

So you want a hypothetical answer to a hypothetical question that is so far removed from reality, it's nonsensical. To what end is this hypothesis useful if nothing of value can be ordained from it?

From the outset 66 years ago to today, F1 has always had customers and manufacturers. All participating teams matter, and when you start to pick and chose whom to examine rather than the entity as a whole, you fail.


"Customer PU's banned...."

In this contortion of yours, you seem to already have made your mind up.

[SR71]

Customer teams don't develop PUs, you're a smart guy so you know they don't matter in this hypothetical examination.

How would the customer teams do if customer PUs were banned?

Williams, force India, Sauber, probably Toro Rosso would fold immediately.

These teams can afford aero programs but cannot afford a competitive PU development program.

That's says it all really.

So you want a hypothetical answer to a hypothetical question that is so far removed from reality, it's nonsensical. To what end is this hypothesis useful if nothing of value can be ordained from it?

From the outset 66 years ago to today, F1 has always had customers and manufacturers. All participating teams matter, and when you start to pick and chose whom to examine rather than the entity as a whole, you fail.


"Customer PU's banned...."

In this contortion of yours, you seem to already have made your mind up.

Alright, I can see things from your side of the fence. Williams spends 20 million on PU's so that's what PU development cost right?

Williams should just spend 40 million, they'd be twice as fast. Also someone should tell mercedes they wasted the rumored half billion they spent on PU development as Williams was able to get the same performance for 20 million.

Just kidding, though this is a hypothetical as I was curious what objectively costs more - nobody forced you to participate.

Thanks for your input though.

[bhall II]

By referencing 2006, we are comparing an all out war between Honda, Mercedes, Toyota, BMW, Ferrari, Renault and Cosworth. Engines were disposable items, unlike today and engines are tasked with reving to 12k, unlike the good ol days. A 12,000rpm limit is attainable in a (tuned) Toda racing Honda S2000, 5 years ago... innards changed for around £5k of our English sterling.
Obviously, the cost lies in the Hybrid part of these engines.

It's still the same deal. Manufacturers are still pouring truckloads of money into combustion chamber design...

It has now been confirmed that Ferrari, at least, has been using this technology since its Canada upgrade engine of 2015. The Scuderia is supplied by Mahle Powertrains and in the latter’s latest annual report, it quotes Fred Turk, head of Mahle Motorsports: “The Mahle Motorsports engineers have found the right solution for Scuderia Ferrari. More than five years of development preceded that crucial phone call to Maranello with the proposal: ‘We’ve found an interesting new solution for you.’ Mahle Jet Ignition is the name of the innovation that gives the engines from Maranello a boost. Within a few weeks, in spring 2015, we adapted our solution to the Formula 1 requirements, allowing Ferrari to compete in Canada with this solution for the first time.”

All development is iterative. The difference between aero development and engine development is that you can print wind tunnel models...





They're not even actual size...



Remember Illien's prototype cylinder fiasco with Renault last year? That was basically 1/6 of this, and it had to run...

[FoxHound]

There is actually a clue in what you posted, Ben.

Mario Illien could replicate the combustion process on a 0.27 litre 1 cylinder engine to that of a 1,6 litre V6.
It would be pointless, time consuming and more expensive to go full size when what he was trying to demonstrate could be attained with a 85%% size reduction.
It was that much better than Renault's concept, for Renault themselves to eventually bring it on board instead of using their own.

Ilmor have the facilities to design, test, evaluate and produce engines then, clearly. It is their mainstay after all.
But I took a quick look at the Ilmor books from 2010 to 2014...

Their turnover in 2014 was £14 million with a gross profit of about £4.5 million.
https://companycheck.co.uk/company/0547 ... l-accounts

And their business is varied, as listed by their website.
Design Racing Defence Development Automotive Aerospace Precision Manufacture Electronics Marine.
http://www.ilmor.co.uk/home


An organisation that can knock up a prototype mule, with an overall budget 15 times less than the average F1 team, and exponentially less to spend on a lil prototype...man that's a pretty convincing case that Aero is by far and away, more expensive.

[bhall II]

Britain's The Independent reports that Mercedes increased its Formula 1 budget by 17 percent last year to a record £324.6 million [$517m].

The report cites company documents for Mercedes' F1 engine manufacturing division as showing that in 2013 it spent £133.9m [$213m] to develop its new 1.6-liter turbo engines. That is nearly double Mercedes' engine budget from three years earlier and of the current engine budget of the Caterham team, according to The Independent. Mercedes F1 engines are also sold to customers as well as supplying the works team.

Beyond the engine costs, Mercedes spent £190.7m [$304m] just on running the F1 team, the report claims.

According to The Independent, Red Bull’s company accounts reveal that their budget has increased by A$36m, capping off a A$117.34m rise over the past five years during which the team dominated Formula 1.

“The team’s income hit £197m last year and it comes from three main sources: sponsorship, prizemoney and payments from their owner, the energy drinks company Red Bull, run by billionaire Dietrich Mateschitz,” The Independent reports.

“Their biggest cost was research and development spending, which came to £83m and was up by 10 per cent.”

That's not all aero, either; it's transmission, suspension, cooling, electronics, etc. If you want to say chassis spending, all-inclusive, can be comparable to engine spending, then I can agree. Just aero? No.

There's a very good reason why engine work is so expensive: it's the most important aspect of motor racing.

If you were to give me $100 million to spend on a racing car - and knowing full well that I'm an aero nut - I would spend as much as I responsibly could on the engine. Then I would spend as much as I responsibly could on suspension/tire development. Only after that would I spend as much as I possibly could on aero. If I did it any other way, I'd have no chance to win.

What was the first component in F1 to be heavily restricted on cost grounds? The engine (homologation, 2006). What was the second? Tires (standardization, effectively 2007). Third? Aero (wind tunnel/CFD restrictions, 2014). The order that of succession was no coincidence.

[FoxHound]

Before we go any further, help a man out...

Red Bull have a 470 million dollar budget. They spend 20 million odd on engines this year, but for accuracy, they actually received their engines free of charge from Renault in previous seasons.

The quote you sourced says spend on research and development was £83m($100m), which leaves 370 million to go on what exactly?

[bhall II]

The quote you sourced says spend on research and development was £83m($100m), which leaves 370 million to go on what exactly?

Richard paid for the Companies House filings from RBR and RBT around the time of the Resource Restriction Agreement debate. They spell it all out exactly, but I can't find them here anymore.

Even so, F1 cars don't just hang themselves up on an atrium wall...

[mrluke]

Put another way, if you were building a race car from scratch, you'd see what the engine was giving you and then look to get the maximum downforce that it could reasonably drag around. You'd then look to make the engine more powerful and the downforce more efficient, both within whatever rules you are working.

To paraphrase if you were building a race car from scratch you would start with the engine.

Outside of the confines of regulation it is definitely easier to add power than to add downforce. Look at drag cars and even tractor pulls, anybody can make an awful lot of power for not that much money. There is cost in making it reliable and efficient sure, but now you are talking about refining.

http://www.chevrolet.com/performance/cr ... s/ls9.html

$25k for a brand new 650bhp engine in a reliable tune that will do 100k miles.

Whats does $25k of aero time get you? about 25 hours? Call it 5 working days in a tunnel without any costs for alternate parts, wages for your aero R&D ....

If you had a £100k (sorry im in uk $ hurts my head) to spend on building a race car, how would you spend it?

[FoxHound]

Cmon....370 million dollars?

How about, as I suspect, that Red Bull have their own R&D expense as listed. And the rest is made up of supplier costs, who have had RB R&D costs palmed off onto them as creative accounting?

Not suggesting subterfuge in any way here, I just find it implausible for an F1 team known for it's high end aero etc to spend 100 million on R&D, with 370million mostly unaccounted for.

[bhall II]

I dunno what to tell you, man.

F1 teams do like their toys, though. From last year...

And Michael Schmidt, the highly-respected correspondent for Germany's Auto Motor und Sport, claims: "Red Bull has ordered its own 'VTT' facility'."

He said the only other companies with access to a 'virtual test track' - allowing full car testing under a roof - are Mercedes and Honda.

Schmidt added: "When you ask what a team needs with this most extreme form of a dynamometer, there is really only one answer -- to become an engine manufacturer."

Those are apparently 40 million (euros/dollars/beatshehelloutofme), and Ferrari allegedly bought one, too. (All to be filed as rumors, of course.)

[FoxHound]

But a VTT sim is as relevant to aero as it is to engines.

I mean, possibly even more so when you consider that aero upgrades and packages occur 20 times in a season these days.
Whereas engines remain constant until the next token package arrives, which could be 6 races into a season.

Be that as it may, I see where you are getting here.

Simulation tech dont come cheap, especially on the finer scale. I'll quote some or other geezer on Mercedes $160 million sim in Sindelfingen.

Here's a copy of the Mercedes GP sim.

This one is based in Sindelfingen, Germany. But Brackely has a copy of this same simulator.

With its 360° screen, fast electric power system and a twelve-metre long rail for transverse movements, the dynamic simulator is the most advanced in the automobile industry. As an equally advanced feature, part of the energy required to drive the simulator is obtained by means of energy recuperation when braking, and fed into the power network of the Sindelfingen plant.

I also stumbled across an interesting link which shows the scale of an F1 operation. 40 man hours to paint a single front wing....
http://www.idgconnect.com/blog-abstract ... or-sensors

Then we get into the Qualcomm/Mercedes collaboration which gets Mercedes data near instantaneously, from car to pit, and probably back to the sims in Brackley.

The real challenge that we had was around tyre data
One of the key things here is that you have a big surface there and you want to see the temperature on all four tyres.
That requires effectively a video of telemetric data, rather than single channels from single sensors. The bulk of data you then need is suddenly in a whole different order of magnitude.
We didn’t have the infrastructure to bring that off the car quickly. So what we developed with Qualcomm was a WiFi solution that allowed us to actually gather that data back into our system before the car had actually stopped, whereas previously we had to download it afterwards.
And it also might have taken us twenty minutes to half an hour because it was so bulky, effectively video-scaled data.”

"I think we’ve got a little more breathing space on the current system, but we are already looking at, and we talked about this a number of times now with the Mercedes guys, moving from the 5 gigahertz 802.11ac system to the 60 gig 802.11ad system.This system was designed for use in buildings. It’s short ranged but it has a very, very wide bandwidth.
So we are now looking if we can put this system in an environment it wasn’t designed for, which allows us to rapidly understand how we can improve that technology so it can be used in an environment where there is the possibility of smog, moisture and rain, things that affect a 802.11ad because of its very high frequency.
So we can do things like compression on the car. This way we can reduce the amount of data that you need to physically dump off. So you can collect more and dump it off quicker. One of the interesting things we discovered was that our system was connecting early, At certain circuits, we actually connected with the car while it was going down the start-finish straight.
So we realized that the system, with some architectural changes, might actually be able to gather data in a very ad hoc burst mode, as the car goes through certain sections. So that’s an area we looked at."

http://www.motorsport.com/f1/news/analy ... 16-676175/

Mostly to your contribution Ben, I reckon we should add simulation to the "cheapest route to performance convergence"....

[bhall II]

VTT is different. It actually runs the car.



I imagine it uses aero data for the simulation. But, aero itself is just about the only aspect of performance it doesn't test.