planetf1.com

It is currently Wed Oct 18, 2017 1:01 pm

All times are UTC




Post new topic Reply to topic
Author Message
PostPosted: Tue Oct 10, 2017 7:31 pm 
Offline
User avatar

Joined: Thu Feb 13, 2003 9:39 pm
Posts: 2925
I originally drafted this post as a response to a discussion on the race thread about the recent problems Ferrari have had with the reliability of their hybrid engines, and the implication about how the simpler engines were better, which was tied to nostalgia about the good old days and history of the sport. However, but the time I had finished researching and writing my post, I realised it was too in depth for the race thread and really needed a thread of its own for discussion.

In 1992 there were 201 retirements in 16 races with a field of approximately 28 cars per race. That's a 45% attrition rate.

In 2004 (a season which at the time people were claiming F1 had bullet proof reliability and was one of the pillars of Ferrari's domination that season) the attrition rate was 93 retirements from 18 races with a field of 20 cars per race. That's an attrition rate of 26%.

In 2017, we have had 67 retirements in 16 races with a field of 20 cars per race. That's an attrition race of 21%.

It's amazing how selective the facts are by the anti hybrid brigade. Fair enough, it's your preferred engine that's fair enough - but claiming "reliability"
as a strike against the technology is nonsense when the glory days harkened back to had car failure rates double of what we see today.

But maybe your argument is that they are "unreliable" compared to modern F1. Well if we look at the facts, the hybrid engines did indeed coincide with an increase in retirements compared to the V8s (still significantly lower than the 'glory days') - however this season the reliability rate is the same as during the V8 era. Plus, the hybrids have to last longer (new V8 every two races, a hybrid must last 5 races) and it should be expected that in the years following the introduction of a new technology will be not be as reliable as an engine formula that has been unchanged for 8 years, and was an evolution of one that went back further still.

2008 -> 22% attrition
2009 -> 24% attrition
2010 -> 22% attrition
2011 -> 16% attrition
2012 -> 31% attrition
2013 -> 12% attrition
2014 -> 36% attrition
2015 -> 34% attrition
2016 -> 32% attrition
2017 -> 21% attrition

The reliability rate in F1 today is so great that if a car gets one retirement it massively affects the data. A 20% attrition rate out of 320 race starts is 64 non finishes.

A 40% attrition rate is 128 non finishes.

Therefore, is a driver happens to have 2 high profile non finishes - out of 64 that's 3%, out of 128 it's 1.5% - and it means it's a) a lot more noticeable -
but also because the cars in general are far more reliable, it means it's unlikely that another car would have a failure.

I mean, there was a 55% chance of making the finish in 1992 compared to 79% this season. If Hamilton was only 55% likely to finish a race that means there is only a 9% chance he'll finish all the remaining races. With a 79% attrition rate is means there is a 39% change he will (based on entire field reliability - Merc's 3% attrition rate means he actually has an 88% chance of finishing all the races - Ferrari's 16% attrition rate means Vettel has a 51% chance of finishing all the remaining races)


Top
 Profile  
 
PostPosted: Tue Oct 10, 2017 7:41 pm 
Offline

Joined: Mon Mar 26, 2012 10:39 am
Posts: 20572
It'd be interesting to see if there's any pattern to the causes of reliability. For instance, in the "good 'ol days" cars were engineered to be practically rebuilt after every session, let alone race. So they ran at pretty much close to their maximum tolerances. I should imagine that stressed components played a big part in breakdowns (this is pure guesswork, BTW). I.e. wear and tear.

Nowadays, components are engineered for a much longer life, so I'm guessing that breakdowns are caused not so much by stressed as faulty items, i.e. manufacturing or design defects, rather than parts giving up the ghost. Certainly I'd wager e.g. Honda's problems were more related to design than wear.

Could be talking complete rubbish, of course, but that's my educated guess


Top
 Profile  
 
PostPosted: Tue Oct 10, 2017 7:50 pm 
Offline

Joined: Wed Dec 19, 2012 1:30 pm
Posts: 23000
It also should be remembered that during 2008-2013 there was not an engine war, no development, the engines were basically spec so reliability should have been really good.

_________________
PF1 Pick 10 Competition

2013: 5th Place
2014: Champion
2015: 3rd Place
2016: 4th Place

2017: Currently 11th

Podiums: 2nd Canada 2015, 3rd Monza 2016, Hungary 2016 and Barcelona 2015


Top
 Profile  
 
PostPosted: Tue Oct 10, 2017 7:53 pm 
Offline

Joined: Tue Oct 04, 2011 1:00 pm
Posts: 4471
Does this analysis only count for in race failures, and if so how much (if at all) does the Honda PU skew the data?

_________________
{Insert clever sig line here}


Top
 Profile  
 
PostPosted: Tue Oct 10, 2017 9:08 pm 
Offline
User avatar

Joined: Thu Feb 13, 2003 9:39 pm
Posts: 2925
RaggedMan wrote:
Does this analysis only count for in race failures, and if so how much (if at all) does the Honda PU skew the data?

Well, because I originally was only researching to reply to a post, I didn't do it as in depth as it could have been as I wanted ball park figures.

I just counted the number of retirements per year and did it as a percentage of race starts.

The Honda PU wouldn't affect it too much because they only account for 10% of race starts. But looking at it, they look to be in line with the average anyway (probably as a result of the fact they usually replaced engines when starting at the back anyway)


Top
 Profile  
 
PostPosted: Tue Oct 10, 2017 9:09 pm 
Offline
User avatar

Joined: Sat Aug 09, 2008 9:22 pm
Posts: 7670
I don't know how it fits in to your data, but my theory of reliability on modern cars v 'back then' is mostly about connectors.

My personal and family cars, and cars I worked on seemed to have about a 50% chance of it being a connector it there was a problem, not it is very unusual.

My race cars were usually 'stricken' by electrical or cooling system 'disconnects' but I have no idea how this converts over to F1.


Top
 Profile  
 
PostPosted: Thu Oct 12, 2017 12:56 pm 
Offline

Joined: Tue Oct 04, 2011 1:00 pm
Posts: 4471
moby wrote:
I don't know how it fits in to your data, but my theory of reliability on modern cars v 'back then' is mostly about connectors.

My personal and family cars, and cars I worked on seemed to have about a 50% chance of it being a connector it there was a problem, not it is very unusual.

My race cars were usually 'stricken' by electrical or cooling system 'disconnects' but I have no idea how this converts over to F1.

Drove a lot a Jaguars then?
(only half joking I had an '86 XJ6 project car)

_________________
{Insert clever sig line here}


Top
 Profile  
 
PostPosted: Thu Oct 12, 2017 1:06 pm 
Offline

Joined: Tue Oct 04, 2011 1:00 pm
Posts: 4471
Alienturnedhuman wrote:
RaggedMan wrote:
Does this analysis only count for in race failures, and if so how much (if at all) does the Honda PU skew the data?

Well, because I originally was only researching to reply to a post, I didn't do it as in depth as it could have been as I wanted ball park figures.

I just counted the number of retirements per year and did it as a percentage of race starts.

The Honda PU wouldn't affect it too much because they only account for 10% of race starts. But looking at it, they look to be in line with the average anyway (probably as a result of the fact they usually replaced engines when starting at the back anyway)

Okay. From the title I was thinking that this was more PU focused but now realize that you're talking retirements in general.

_________________
{Insert clever sig line here}


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 4:39 pm 
Offline

Joined: Wed Jan 17, 2007 3:29 pm
Posts: 905
RPM killed those 2004 engines. 100%. Loved hearing their sound. But rpm is far harder on connecting rods, cylinder sleeves, and pistons than torque. These current ICEs we're always more reliable on paper even in new regulations and development.

I love all F1 tech though so I've been pretty indifferent in my preferences for the most part.

_________________
[ Kimi Raikkonen ]
2007 Formula 1 World Drivers Champion


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 5:32 pm 
Offline
User avatar

Joined: Sat Aug 09, 2008 9:22 pm
Posts: 7670
Honda Quick wrote:
RPM killed those 2004 engines. 100%. Loved hearing their sound. But rpm is far harder on connecting rods, cylinder sleeves, and pistons than torque. These current ICEs we're always more reliable on paper even in new regulations and development.

I love all F1 tech though so I've been pretty indifferent in my preferences for the most part.


Just as a related aside, many people do not realise how short the stroke is on the 'new' cars. It is not much more than a model air plane engine. Less than 40mm 8O


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 5:51 pm 
Offline
User avatar

Joined: Fri Apr 01, 2016 12:13 pm
Posts: 1019
Thank you for the insightful information Alien.

Our perception on reliability is jaundiced by the penalties that are usually attached to any reliability issues. So what we see is reliability issues affecting the start of the race (drivers suffering huge position drops) rather than during the race where a DNF is barely noticed.

In racing there is a well worn adage where if a car finishes a race, it is too reliable and thus too heavy. So mass must be removed. The opposite is that if the car breaks, something is too light, and weight must be added. In racing, finding this balance is where a lot of money and resources are spent. Even where the entire car may be underweight, being able to relocate mass can yeild noticeable performance gains.

Every nut, bolt, part is subjected to a comprehensive weight and stress analysis, in an effort to understand what stresses each part is subjected to, and hot to make it strong enough, but as light as possible. Computers have become a huge part of this process, and by running (for example) a Finite Element Structural and Fatigue Analysis do engineers come closer to finding that magic sweet spot.

Image
https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTvHjOToYE6P8W5CqiFU3vshFJVxUSZ9jobXuixh0JZAHX5tuR5

We did not have these kinds of tools a couple of decades ago, and the ability to make such detailed analysis has contributed heavily to performance and reliability gains.

But this is racing, and when you are locked in a battle worth many millions of dollars and the difference in performance between teams is so minuscule, each part is stressed to the absolute maximum. In today's racing, reliability is determined by each team's willingness to push the boundaries.

Here's a little true story that may assist in understanding. NASCAR has three tiers, and occasionally drivers in the top tier would race in the 2nd tier for money or just to learn the track. And they would usually win. That is because the regular second tier drivers are running for a championship, and thus their cars are not as stressed. But the ringers could run at the limit because they were in it to either win or blow up.

_________________
Only dogs, mothers, and quality undergarments give unconditional support.


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 6:19 pm 
Offline
User avatar

Joined: Sat Aug 09, 2008 9:22 pm
Posts: 7670
Blinky McSquinty wrote:
Thank you for the insightful information Alien.

Our perception on reliability is jaundiced by the penalties that are usually attached to any reliability issues. So what we see is reliability issues affecting the start of the race (drivers suffering huge position drops) rather than during the race where a DNF is barely noticed.

In racing there is a well worn adage where if a car finishes a race, it is too reliable and thus too heavy. So mass must be removed. The opposite is that if the car breaks, something is too light, and weight must be added. In racing, finding this balance is where a lot of money and resources are spent. Even where the entire car may be underweight, being able to relocate mass can yeild noticeable performance gains.

Every nut, bolt, part is subjected to a comprehensive weight and stress analysis, in an effort to understand what stresses each part is subjected to, and hot to make it strong enough, but as light as possible. Computers have become a huge part of this process, and by running (for example) a Finite Element Structural and Fatigue Analysis do engineers come closer to finding that magic sweet spot.

Image
https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTvHjOToYE6P8W5CqiFU3vshFJVxUSZ9jobXuixh0JZAHX5tuR5

We did not have these kinds of tools a couple of decades ago, and the ability to make such detailed analysis has contributed heavily to performance and reliability gains.

But this is racing, and when you are locked in a battle worth many millions of dollars and the difference in performance between teams is so minuscule, each part is stressed to the absolute maximum. In today's racing, reliability is determined by each team's willingness to push the boundaries.

Here's a little true story that may assist in understanding. NASCAR has three tiers, and occasionally drivers in the top tier would race in the 2nd tier for money or just to learn the track. And they would usually win. That is because the regular second tier drivers are running for a championship, and thus their cars are not as stressed. But the ringers could run at the limit because they were in it to either win or blow up.


I spent lots of evenings and week ends on the drill stand or when I could bum time on the Bridgeport :]

Holes are the best thing to put in anything if you race


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 7:00 pm 
Offline
User avatar

Joined: Wed Feb 18, 2015 5:38 pm
Posts: 1504
Location: Miami, Florida
LOL Moby!

The technical terminology is "Swiss Cheesing".

_________________
HAMILTON :: VETTEL :: ROSBERG :: RAIKKONEN :: VERSTAPPEN :: SAINZ :: MASSA :: BOTTAS :: NASR
ALONSO :: BUTTON :: PEREZ :: RICCIARDO :: GROSJEAN :: KVYAT :: HULKENBERG :: MALDONADO
THE REST… THERE ARE FAR BETTER DRIVERS THAT SHOULD BE IN FORMULA 1


Top
 Profile  
 
PostPosted: Fri Oct 13, 2017 8:26 pm 
Offline
User avatar

Joined: Sat Aug 09, 2008 9:22 pm
Posts: 7670
F1 MERCENARY wrote:
LOL Moby!

The technical terminology is "Swiss Cheesing".


Yip. I was always short of cash, so if it wanted light, I had to make it so :D

Being more 'classical' I always called it well ventilated though


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic

All times are UTC


Who is online

Users browsing this forum: Black_Flag_11, Fiki, kerneys, mcdo, mds, sandman1347, Siao7 and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group