The F1 Technical Developments Thread

Well, thanks, but that?s not what I was looking for. ;)

What I want to know is how the fuel is injected into the engine. By normal standards, fuel delivery at 19000 rpms should be a big problem. I want to know if the injectors still fire "normally" ie at certain intervals, or does the ecu do something cool like just open them up completely in the high rev-range and just regulate the pressure in the rail in order to get the desired air/fuel ratio.

The site is quite interesting though! I came across it while googling, but didn?t take a second look...

I think you're on the right track...

maybe this vid will shed some light:

 
Yeah I love watching that video. :D
Unfortunately it?s impossible to see if the injectors are opening and closing around 100 times /sec, or just pissing constantly. Dammit, I hate when I want to know something and there is no way to find out. I?ve posted in the f1technical forum as well, almost 80 views, no answer. :(
 
Ford, have you tried looking at the other application of high-revving four-stroke engines: high-performance road bikes? There's quite a big Hayabusa modding community, for example.
 
Last edited:
Yeah I love watching that video. :D
Unfortunately it?s impossible to see if the injectors are opening and closing around 100 times /sec, or just pissing constantly. Dammit, I hate when I want to know something and there is no way to find out. I?ve posted in the f1technical forum as well, almost 80 views, no answer. :(

The only way to know, would be to find out the kind of injector and its pulse width limitations. @18,000 the injector needs to open and close 75 times/sec as you mention, that means the injector would need a pulse width of at least 1/75 - 1/100 of a second. Looking at the Delphi direct injection injectors, http://delphi.com/manufacturers/auto/powertrain/gas/injsys/homogenous/ , they have a opening and closing response time of 0.4 msec which is more than adequate in achieving high RPM fuel injection. F1 hasn't yet approved direct injection but I'm merely using this example to show that a regular injector, especially one for F1, can be designed to have a similar response time as the technology does exist and probably has for some time.

I hope this helps a little bit...
 
Ford, have you tried looking at the other application of high-revving four-stroke engines: high-performance road bikes? There's quite a big Hayabusa modding community, for example.

He should look for info about the Yamaha R6, that bike revs to 18000rpm.
And iirc Honda have a i4 250cc bike that revs well above 20000rpms, that bike used carbs though...
 
Last edited:
Let me summarize the article linked below for you: Buried below layers and layers of marketing blurbs and smoke and mirrors, Virgin confirms that the CFD-only approach did not work and they'll start using wind tunnels.

http://www.autosport.com/news/report.php/id/90915
 
For the record....the voice of the commentator for the video that plays on the Autosport website when you open it up sets my teeth on edge. Time to take it down an octave dude.

Not surprised about CFD not working though. I love computers and such, but there is only so many parameters you can really get away with programming in before its excessive. Seems to me that a wind tunnel facility that the smaller teams can time share on might be a cool idea, though they probably wouldn't go for it. Guess there is a reason why most F1 facilities are built with the wind tunnel as the main part of the building.
 
Last edited:
Not surprised about CFD not working though. I love computers and such, but there is only so many parameters you can really get away with programming in before its excessive. Seems to me that a facility that the smaller teams can time share on might be a cool idea, though they probably wouldn't go for it. Guess there is a reason why most F1 facilities are built with the wind tunnel as the main part of the building.

CFD has come a long way in the past 10 years but it will never replace wind-tunnel testing.
 
When asked about the possibility of some windtunnel testing, Virgin Racing president Graeme Lowdon said: "From my point of view, I don't care if it is windtunnel, dowsing with coat-hangers or whatever the technology is ? as long as it fits a commercial profile and works."

There's yer problem! Start giving a damn next time, Mr Lowdon!
 
You got me thinking now!

They wont really have to be like a road car and stick to 14.xx:1 AFR because they dont have a catalyst to worry about and emissions....pah, what emissions. Last on the list surely. But then how do they measure the air flow for the AFR... do they do it from exhaust gas components or do they measure in the air box? Thats surely made more complicated by how they manage to generate a forced-induction effect with pressure waves formed inside the airbox. I was at a presentation given by Mercedes High-Performance engines and they showed CFD of how they time the inlet valve openings with the reflections of these pressure waves set up inside the airbox by the incoming air. They effectively get a supercharging effect without a supercharging device.

I guess the main things they can change is injector nozzle design and location of the injector with respect to the ignition point if they use DI (though with such shallow strokes maybe they dont?) That and fuel pump design.


Ford.... I know people who may well know the answer to this. Here at work there are a number of guys who did work for the Jaguar F1 team when they were around, on engine development. I remember they showed me what happens when a valve breaks at 13000rpm :| . There is also a guy who works for Lotus hanging around in the test labs and while Lotus dont design their own F1 engines he may know what goes on.

CFD has come a long way in the past 10 years but it will never replace wind-tunnel testing.

interesting little fact.... the big teams can solve CFD with 100's of millions of control volumes in a little over 2 hours. Like full car simulations... I cant remember the specs of the cluster they mentioned but it must need its own power station to run!
 
Last edited:
interesting little fact.... the big teams can solve CFD with 100's of millions of control volumes in a little over 2 hours. Like full car simulations... I cant remember the specs of the cluster they mentioned but it must need its own power station to run!

Yeah it's pretty amazing, I heard Ford developed a system that models by using a discrete number of air particles and solving CFD equations for each one.
 
interesting little fact.... the big teams can solve CFD with 100's of millions of control volumes in a little over 2 hours. Like full car simulations... I cant remember the specs of the cluster they mentioned but it must need its own power station to run!

Problem is that there's CFD and there's CFD. If they're solving unstructured Euler with a coarse boundary mesh then you'll get answers but they're not going to be accurate representations of the whole flow.
 
Problem is that there's CFD and there's CFD. If they're solving unstructured Euler with a coarse boundary mesh then you'll get answers but they're not going to be accurate representations of the whole flow.

Thats a second hand story from my friends who went to an ANSYS conference.... It blew their minds and CFD is their day job, so it must of been awesome-sauce. I dont think they'd be doing stuff like that unless they wanted a really rough idea about whats going on. Still, it doesnt beat a wind tunnel no matter how many months you spend building a CFD model. You still need to validate it and if the data's not out there for you to use (ie you can often validate your CFD work with other peoples CFD and physical experiments from papers as its classed as the techniques involved can be perceived best practice) you have to go get it yourself.


I have to wonder at what point does doing bag loads of CFD actually become less economical than actually doing a physical test in a wind tunnel?

I'd love to work for an F1 team doing that, but apparently its not terribly well paid. The reason being you do it because you love F1..... so you'll do it for less than going rate. Thats what this Ex-McLaren aerodynamicist told us anyway.
 
Last edited:
I'd love to work for an F1 team doing that, but apparently its not terribly well paid. The reason being you do it because you love F1..... so you'll do it for less than going rate.

And i can't even see anything wrong with this, as long as it pays the bills. At the end of the day, i'd rather do a job i love than a job that pays better.
 
I'd love to work for an F1 team doing that, but apparently its not terribly well paid. The reason being you do it because you love F1..... so you'll do it for less than going rate. Thats what this Ex-McLaren aerodynamicist told us anyway.

Is the pay poor, or is it just the amount of hours worked?

I read Steve Matchett's books and he details the annual salary he got as a race mechanic which on it's own he thought was a good figure, but when he started factoring in the massive amount of hours he put in, it worked out to be a very small amount :p. It's a lifestyle, you probably wouldn't do it if you didn't have a love of your job, employer/team and industry.
 
I'd love to work for an F1 team doing that, but apparently its not terribly well paid. The reason being you do it because you love F1..... so you'll do it for less than going rate. Thats what this Ex-McLaren aerodynamicist told us anyway.

+ i can imagine it does wonders on a CV, if it says you've worked for an F1 team, you'll have some advantage over other candidates...(unless they've worked for NASA :lol:)
 
I like the way this discussion is proceeding.

Anyone know of more books similar to Steve Matchett and his experiences ? Mechanics, Engineers, Technical Directors, Team Managers, etc ???
 
Might anyone happen to know if there is any sizable source of technical information on the Lotus 56? For once, google doesn't provide much.
 
Top