Thread by @NaturalParadigm

“Adding power makes you faster on the straights. Subtracting weight makes you faster everywhere.”
—Colin Chapman

The best answer to this version of the “10% problem” is 4) Worst Car/Less Weight. Let's see why I think so... (a 🧵)

#F1Tech #F1

https://twitter.com/NaturalParadigm/status/1514344420522483716

First of all, what exactly is being asked here? 🤔 The fastest car on the grid is slowed (by increasing weight or decreasing power), and the slowest car on the grid is made faster by the same variables (less weight/more power). The race is ONLY between these four modified cars.

Since there's no limit to how much time a bad driver can cost you[1], we suggest using the best qualifying lap from each team as a metric for “pure” car performance. In Melbourne, this gives the following results:

LEC - 1:17.868
VET - 1:21.149 (+3.281s)

formula1.com/en/results.htm…

This gap will vary from circuit-to-circuit, but performing a similar calculation for Bahrain and Jeddah (+2.219s and +2.292s, respectively) suggests that–if anything–this gap is actually larger than average, giving an advantage to the modified “fast” cars. We continue...

Now that we know the gap, how much time is actually gained or lost through weight and/or power changes? 🤔

Based on the best information I can find from multiple sources [2-5], a representative value for the weight effect is ~0.03s/kg-lap, or 0.3s per lap per 10kg.

#F1Tech #F1

Here, Sir @LewisHamilton confirms his understanding that the weight effect of a modern F1 car to be ~0.03s/kg-lap:

The horsepower effect will vary between circuits (think Monza versus Monaco 😲), but over the course of a season, a reasonable value for this parameter seems to be somewhere in the range of 0.0160-0.0175 s/hp-lap. [4,6,7]

📸: @autosport

Laptime simulations provided by Juan[8] are also available, but are used only for qualitative analysis here. Based on a 795kg car with 900whp, his simulations suggest the following laptime derivatives:

Weight Effect: 0.0266s/kg-lap
Power Effect: 0.0154s/hp-lap [9]

#F1Tech #F1

So, what does a 10% change in power (or weight) buy in terms of laptime? Again assuming a 795kg car with 1000bhp and weight/power effects of 0.030s/kg-lap and 0.0175s/hp-lap respectively, we have:

Power: (100hp)*0.0175 = 1.750s/lap
Weight: (79.5kg)*0.030 = 2.385s/lap

The largest power effect value was chosen to make the calculation more conservative, but it's already clear that weight has a greater impact on laptime than power. Indeed, Juan's simulations show this same trend: the 110% weight car was slower than the 90% power car. ⤵️

#F1Tech

As intimated by the quote of Colin Chapman, removing weight pays dividends in every aspect of performance: acceleration, braking, and turning; power (mainly[10]) saves time on the straights. It's the reason F1 teams chase weight targets SO aggressively...

https://twitter.com/LuisFeF1/status/1514301541930389504?t=zyJoUYy_UUAHtLHjdiJRuQ&s=19

Relative Δ:
Best Car/Less Power: +1.750s
Best Car/More Weight: +2.385s
Worst Car/More Power: -1.750s
Worst Car/Less Weight: -2.385s

Δ to “Fastest” Car:
Best Car/Less Power: +1.750s
Best Car/More Weight: +2.385s
Worst Car/More Power: +1.531s
Worst Car/Less Weight: +0.896s

Even with these fairly conservative assumptions, the worst car with 10% less weight handily bests the other combinations; in fact, it's now actually within one second of the unaltered fastest car. Using the field gap seen in either Bahrain or Jeddah, it 𝙬𝙤𝙪𝙡𝙙 be faster. 🤯

But hang on Bryson, didn't @ScarbsTech clearly say the best car with less power would be the fastest? What gives?

https://twitter.com/ScarbsTech/status/1514532515566043139?t=UoNWT31W-RqCgnpgWpwFUQ&s=19

Scarbs did indeed say this, and his reasons for doing so are implied by the “22.5%” value he gave us. Leclerc's pole time was 1:17.868, while Latifi's time from Q1 was 1:21.372 (+3.504s). Assuming a 795kg car weight, the weight effect is:

(3.504s)/(0.225*795kg) = 0.0196s/kg

For an F1 car, this value of weight effect is significantly lower than any number I have been able to find thus far while researching this problem. That's not to say it's “wrong” in an objective sense (far from it), but its origin is totally opaque to me.

f1technical.net/features/21637

Both Craig and I agree that, for a typical car/circuit combination, the weight effect is greater than the power effect (except perhaps at tracks like Monza). But even if we share the same calculations, there simply is no way to arrive at the same conclusion from different inputs.

This was a fun problem to research! It spawned some great conversations and taught me a lot about vehicle dynamics[11]. I have made every effort to research the parameters of this problem carefully and check my calculations. My answer is definitely 4), but I'm open to discussion!

Notes/References:

[1] See Nikita Mazepin.
[2] Telemetry Analysis:

https://twitter.com/Formula1Trends/status/1514402867846602759?t=P8AtSHJor1u6r25aTIh0KA&s=19

[3] GurneyFlap.com: gurneyflap.com/Resources/arai…
[4] f1technical.net/features/21637
[5] Ex-Mercedes F1 engineer:

https://twitter.com/gabednconfused/status/1514409622161047557?t=P8AtSHJor1u6r25aTIh0KA&s=19

[6] reddit.com/r/formula1/com…
[7] bbc.com/sport/formula1…

[8] Fastest-Lap is very powerful and has been used to make some particularly cool simulations, definitely worth a look: ⤵️

#F1Tech @fastestlap_f1

[9] “900whp” means 900 wheel horsepower. Since all other reported figures are based on brake horsepower, this power effect value is is believed to be under-represented by an unknown (but presumably small) amount.

[10] The important caveat here is that more power allows you to run more downforce than you otherwise would be able to, which improves cornering speeds and could make enough of a difference to offset the purely mechanical effects.

[11] Researching this project introduced me to some people I would really like to meet. @onthechicane, could you put me in touch with Jahee from Wavey Dynamics? Seems like my kind of people... 👊

waveydynamics.com/post/lap-time-…