Actual 2014 gearing from online videos

[hollus]

As the teams are locked into their current gear ratios for a while, I looked for some on board videos and extracted many RPM-Km/h pairs. And then I plotted them here and fitted straight lines (constrained to pass through the origin) to the data. I tried to collect data only in acceleration phases where there was little risk of wheel spin, and after removing a few outliers, I am surprised at how good the data we get fed is. There is quite some jitter, but most of it is due to the fact that the refresh rates for RPM and Km/h seem not to be synchronized. For the lower gears I had to accept cornering data, but it still works quite good.

Please note that this does not reflect actual limits of RPM usage. The data are a mix of drivers pushing in quali, pushing in the race, pit stops, drivers in traffic and cruising around.
Also note that for many teams 1st and 8th gears are missing for now. 8th gear is the orange circles (notably missing for Mercedes, unused AFAIK), 1st gear the squashed salmon rectangles.

Whenever I compared the two drivers from the same team, the numbers matched reasonably well, so I think it is safe to assume the same set of gear ratios for both drivers in each team.

There is a video with RPM for Ricciardo, but I skipped it as it was with wet tires, which are larger than dry tires.

Data for Mercedes. Mostly Hamilton, the lower gears include Rosberg in slicks in a damp track.


Data for Williams. Mostly from Bottas, second gear from Massa in slicks in the wet.


Data for McLaren. Mostly from Button. The hovering spot in eight is button maxing up at the end of the main straight due to drag! (His data are from the race, most others are from quali, including the top speed in 8th from Magnussen).


Data for Force India. From Hulkenberg.


Data for Toro Rosso. From Kvyat.


Data for Red Bull. From Vettel in P3 and Ricciardo in the race. Vettel seems to have somewhat shorter 3rd, 4th and 5h gears than Ricciardo, but with possible issues with tire pressures and wear from practice to race, I decided to leave all the data in. Data points to the right of the linear fit can often correspond to wheelspin.


And finally a comparison of calculated speeds for each gear between 8000 and 13000RPM, where they are likely to spend 99% of the time. I hope to get all 8 gears from all 11 teams in eventually (and it better be soon before they start to change them). Note that I made space for whatever the monstrous 8th gear from Mercedes is going to be. But I doubt they'll reach the 400+ Km/h implied in this data.
I am starting to consider whether Mercedes is saving weight with 1st and 8th gears made of plastic (kidding), because they really don't seem to need them, at least until Spa. And notice how incredibly compressed all gears are in the Williams.

*Note, look at the colors, don't be fooled by the missing 1st and 8th gears calculations.
**Note2, the data fro Lotus and Marussia are from Blanchimont's calculations.
####### Edit: Updated and with the correct colors for Torro Rosso in page 2 ######




There are quite some teams missing and also many gears missing. If anyone can point me to appropriate videos, I'll be more than happy to update the data, but I fear this will have to wait until Malaysia. I'd be even more happy to update this is someone sends me the already extracted RPM-Km/h pairs. A text or word or excel file should be enough.

[Richard]

Awesome work, than you.

Could you show a table comparing the ratios, ie numeric? You have the engine rpm and can divide the km/h by the wheel circumference to get wheel rpm.

[piast9]

Excellent job! Congratulations!

It is actually interesting to see that two teams are noticeably short geared in comparison to other teams. Let's see how does that affect the relative performance differences on low and high speed tracks. And it will also be interesting what change the teams will make during the season if they choose to do so.

I really would like to see that chart completed for other teams so I keep fingers crossed for good onboards and your patience, Hollus!

[hollus]

How long is the circumference of 2014 slicks?

[Blanchimont]

There is a video with RPM for Ricciardo, but I skipped it as it was with wet tires, which are larger than dry tires.

The wet weather tyres are 670mm in diameter compared to the 660mm of the dry weather ones. The error of the radius of 0,5/330 = 0,15% shouldn't matter for these calculations.

I also had a look into some videos and found out that the 8th gear of the Toro Rosso is ~437km/h, the longest 8th gear so far. It could mean that the Toro Rosso is designed for low drag and maybe the car performs quite well in Italy and Belgium? We'll see.

For Lotus i got(speeds @ 15000rpm):
3 187,0
4 225,0
5 280,0
6 327,0
7 368,0
8 424,0

Marussia:
2 158,0
3 200,0
4 240,0
5 281,0
6 323,0
7 363,0

Red Bull:
3 192,0
4 239,0
5 290,0
6 345,0
7 382,0
8 422,0

[Tim.Wright]

There is something strange in those regression lines. You said they are constrained to pass through zero, and their equation does indees suggest that, but looking at the lines... they really dont look like they are passing through 0,0.

Scratch that... I see the horizontal axis isn't starting from zero

[Blanchimont]

There is something strange in those regression lines. You said they are constrained to pass through zero, and their equation does indees suggest that, but looking at the lines... they really dont look like they are passing through 0,0.

The x axis doesn't start at 0 rpm!

[Tim.Wright]

That would be why then...

[tok-tokkie]

In the Williams thread tomazy has done a similar thing. He posted a link on Saturday to a video of the Williams where it clearly goes from 1st to 8th gear but the display never illuminates the 8. The video gear display is not always correct when showing 7th gear. Surprisingly that does not seem to cause problems with the data you have collected.

[hollus]

Thanks for the heads-up, Tok-tokkie. the initial post is now updated. 8th gear is not a problem. If you follow the RPM, it is obvious when they shift.
I found his link to a link to 25 minutes of Ricciardo in the race, which crucially include pit stops, safety car and the parade lap. Interestingly Ricciardo seems to have somewhat longer middle gears than Vettel. I decided to include both sets of data points, but Vettel is systematically to the right of the fitted curve in 3rd, 4th and 5th gears. This can also be wheel spin from the limited data I got from a video suggested by Blanchimont, since 6th and 7th gear match well. The Ricciardo video also included 1st gear, but that is used only out of the pit stop and all measurements included a lot of wheel spin, so I left it out. By the time he hits the rev limited, he is already in second gear.

By the way, I noticed that wheel spin can be detected in those graphs, it often results in more revs that expected for the car speed (and in me discarding the data point if I had many without the risk of wheel spin). But how is this possible? Is the speed we get from a ground sensor and not calculated from engine RPM? Or maybe the associated changes are too fast for the telemetry software to pick it up? Different smoothing functions for RPM and Km/h?

The update also includes calculated gearing for Lotus and Marussia after Blanchimmont's data.

And for Richard, here the engine RPM to wheel RPM ratios I have so far, assuming 660mm diameter and a perfect circumference:

Gear	Merc.   Will.   McLaren F.India T.Rosso R.Bull  Lotus   Marussia
1	13.597		19.288					
2	11.509	11.238	12.579	11.584	12.429	13.508		11.815
3	9.577	9.600	9.347	9.519	9.305	9.913	10.231	9.333
4	7.879	8.021	7.732	7.904	7.713	7.825	8.294	7.776
5	6.299	6.824	6.344	6.293	6.486	6.460	6.664	6.642
6	5.105	5.702	5.454	5.122	5.471	5.524	5.707	5.778
7	4.299	4.953	4.839	4.399	4.723	4.886	5.072	5.141
8		4.401	4.376		4.278	4.418	4.402

Edit: Updated in page 2

[chip engineer]

Thanks for the heads-up, Tok-tokkie. the initial post is now updated. 8th gear is not a problem. If you follow the RPM, it is obvious when they shift.
I found his link to a link to 25 minutes of Ricciardo in the race, which crucially include pit stops, safety car and the parade lap. Interestingly Ricciardo seems to have somewhat longer middle gears than Vettel. I decided to include both sets of data points, but Vettel is systematically to the right of the fitted curve in 3rd, 4th and 5th gears. This can also be wheel spin from the limited data I got from a video suggested by Blanchimont, since 6th and 7th gear match well. The Ricciardo video also included 1st gear, but that is used only out of the pit stop and all measurements included a lot of wheel spin, so I left it out. By the time he hits the rev limited, he is already in second gear.

By the way, I noticed that wheel spin can be detected in those graphs, it often results in more revs that expected for the car speed (and in me discarding the data point if I had many without the risk of wheel spin). But how is this possible? Is the speed we get from a ground sensor and not calculated from engine RPM? Or maybe the associated changes are too fast for the telemetry software to pick it up? Different smoothing functions for RPM and Km/h?

The update also includes calculated gearing for Lotus and Marussia after Blanchimmont's data.

And for Richard, here the engine RPM to wheel RPM ratios I have so far, assuming 660mm diameter and a perfect circumference:

Gear	Merc.   Will.   McLaren F.India T.Rosso R.Bull  Lotus   Marussia
1	13.597		19.288		12.429			
2	11.509	11.238	12.579	11.584	9.305	13.508		11.815
3	9.577	9.600	9.347	9.519	7.713	9.913	10.231	9.333
4	7.879	8.021	7.732	7.904	6.486	7.825	8.294	7.776
5	6.299	6.824	6.344	6.293	5.471	6.460	6.664	6.642
6	5.105	5.702	5.454	5.122	4.723	5.524	5.707	5.778
7	4.299	4.953	4.839	4.399	4.278	4.886	5.072	5.141
8		4.401	4.376			4.418	4.402

Here is what the relative ratios are between gears:

Gear	Merc.   Will.   McLaren F.India T.Rosso R.Bull  Lotus   Marussia							
1								
2	1.181	0.000	1.533	0.000	1.336	0.000		0.000
3	1.202	1.171	1.346	1.217	1.206	1.363	0.000	1.266
4	1.216	1.197	1.209	1.204	1.189	1.267	1.234	1.200
5	1.251	1.175	1.219	1.256	1.186	1.211	1.245	1.171
6	1.234	1.197	1.163	1.229	1.158	1.169	1.168	1.150
7	1.187	1.151	1.127	1.164	1.104	1.131	1.125	1.124
8		1.125	1.106			1.106	1.152

Williams seems to be using the closest spacing through the range.

[tomazy]

Great work =D> I tried something similar in another thread as you allready know, but with much less data and visual interpretation. Nice to see that the videos in my post helpt a bit.

One thing that it is iteresting to me also, is the difirence in RPM teams are using. If we just look at Williams and Force India for exemple, they use the same engine, but Williams changes its gear at 11500RPM when FI does it at 13500.

[Lt_Boards]

Hi All,

I'm a new member and first time poster but have been enjoying readin the F1Tech forums for some time.

In regards to top speed, Will this year's cars ever hit 15k RPM?

Given the instantaneous fuel flow limit, wouldn't most teams not exceed approximately 12.5k rpm like in Melbourne?

This would give a top speed of approximately 353kph for the FW36, which is in the 350kph ball park I would expect them to hit around Monza (based on my readings of other topics on this forum).

Thoughts?

[SamH123]

The plots are interesting, but do they have any interesting implications?

Can anyone summarise any interesting points from the data? (I am a bit clueless )

[hollus]

It just tells you many small things.

Williams, McLaren and Red Bull are not counting on going over 350Km/h any time soon, because they can only do that in 8th by running well over 12000 RPM, and I haven't seen anybody happy to stay over 12000 for more than a split second, except Force India, which goes to 13000 in 4th and 5th gear. Force India chose very spaced middle gears, so they have to, and they'll have to for a few races. This also tells you that the Mercedes engine can routinely take 13000, but likely they are taking a hit in cooling or in efficiency to do that, and we'll see if in reliability also.

Williams with its impossibly compressed gears, shifts at the lowest revs, and in general if they choose to (and they seem to do so), they can run all the time between 10000 and 11500 RPM, as low as possible while using (almost) the full fuel allocation. Maybe this is also the reason for their bulletproof reliability. But they might be compromised out of Loews and in starts and pit stops, because 1st gear (which I haven't included yet) is left to deal with everything under 90Km/h all by itself. This 1st gear problem might also apply to Mercedes.

Edit: The Red Bull / Toro Rosso was wrong, as the gears in the Toro Rosso are displayed in the wrong colors in the combined graph. In reality they are 2nd through 8th.

Mercedes, with gears very far from each other starting at 180Km/h in 4th, will be forced to either go to 12000 and above before upshifting, or to accept that the shift will leave them below 10000RPM, and hence accept not to use all the fuel they could otherwise. This suggests that their fuel consumption is indeed very high, since they are optimizing for not using it all anyways.

And lots of stuff other people might pick up.

But all in all, with the wide power bands, it might not make such big a difference.