F1 Engine friction

[alexbarwell]

As a different branch to this, have a look into Baertek (?) Diamond-Like-Coatings as used in some race series and other ceramic coatings, as the DLC is stated to reduce friction and heat transfer, a number of the problems are tackled - reduced friction, reduced temperature, less drag, so higher speeds possible, reduced cooling problems...

[F1_eng]

DLC is interesting technology but it didn't deliver the promise of big reduction in friction.

It was advertised for a while as a friction reducing coating and I was involved with a lot of testing work concerning DLC. Unfortunately we never saw any huge benefits in terms of friction reduction. It is impossible to achieve the same surface finish with DLC covered components even through super-finishing.
The oil film thickness is reduced with the DLC but this is not always desirable.
We did see huge benefits in terms of wear characteristic with the introduction of DLC and most companies these days offer the product as a coating with better wear characteristics rather than friction reduction.

You must be careful with its placement because it is very hard and unless the bonding process is refined, you can end up with very hard bits of DLC around the engine.

We have not investigated using the technology on cylinder bores because of the temperatures involved. By its very nature, the bonding temperatures are much lower for DLC than pure Diamond, this is why DLC is used rather than diamond. The temperatures to bond diamond to a surface would be about 2800K if i can remember, the component would be a big sludge by this time whilst DLC can be bonded at a much lower temperature.

DLC does get used widely in racing engines but its not the amazing friction reducing coating that some people claimed a few years ago.

[xpensive]

From what I can understand, bearing material only matters friction-wise if you are relying on boundary-layer lubrication and not the desired full-film lubrication.

Regarding ball-park numbers for internal-friction I was asking for, it seems F1_Eng has inside knowledge he is reluctant to share.

Anyway, if the engine's mechanical output is some 720 Hp, or 530 kW, at an efficiency of say 20%, I guess that 2000 kW is leaving the engine as heat, basically through exhaust, water-cooling and lubrication-cooling?

Expanding the topic a bit, it would be most interesting to learn more about how the above thermal-power is split up.

[riff_raff]

xpensive,

Recip engine friction losses (FMEP) are due to many factors, and whose magnitude varies with engine speed and load. Small displacement,small stroke, big bore, multi cylinder, high-revving F1 engines make lots of power, but are not thermally efficient. A larger displacement, small bore, long stroke, engine with fewer cylinders would be much more efficient.

At part throttle and high speeds, the friction losses due to intake flow pumping predominate. But at full throttle and high speed, piston ring friction losses likely predominate. Yet at part throttle and low speed, accessory drive losses predominate.

An engine is most efficient when the IMEP-FMEP value is highest. This condition (BMEP) normally occurs at low speed and wide-open-throttle. Friction losses tend to increase with the square of the engine speed, so a slow revving, large displacement engine is the best choice for efficiency. Remember the Mercedes C9 low-revving V8 that won LeMans in '89:

http://en.wikipedia.org/wiki/Sauber_C9

Regards,
Terry

[xpensive]

That is all very interesting Terry, but as the curious engineer I am, I'm even more intrigued to see numbers on the above.

[riff_raff]

xpensive,

Engine efficiencies are nebulous things. There are no hard and fast numbers.

Designing any engine is an exercise in compromise.

But having said that, here's an example: A very good, modern class 8 diesel truck engine has a max BTE (Brake Thermal Efficiency) of about 45%. That means that the remaining 55% of the thermal energy available in the fuel goes out the exhaust, the radiator, or into the accessories. An F1 engine is likely 35% BTE at best, which means that 65% of the fuel's energy is lost.

[xpensive]

Terry, even if the modern 720 Hp F1 engine has an efficiency of 33%, which I seriously doubt, that would still mean that almost 1100 kW is wasted through exhaust and water/oil cooling.

Again, my initial Q was how much power is lost through internal friction at 18k Rpm?

[tok-tokkie]

Here is an extract from the 'Formula One Engines' thread on the Home page of this board:

* Higher volumetric efficiency. VE is used to describe the amount of fuel/air in the cylinder in relation to regular atmospheric air. If the cylinder is filled with fuel/air at atmospheric pressure, then the engine is said to have 100% volumetric efficiency. Turbo chargers for instance can increase VE to above 100% while normally aspirated engines tipically run anywhere between 80% and 100%. In this region however, a Formula One engine usually can achieve a higher VE than normal road engines because of their highly optimised intake manifolds.
* Unfortunately, from the total fuel energy that is put into the cylinders, averagely less than 1/3 ends up as useable horsepower. Ignition timing, thermal coatings, plug location and chamber design all affect the thermal efficiency (TE). Low compression street engines may have a TE of approximately 0.26, a racing engine may reach approximately 0.34. This seemingly small difference results in a difference of about 30% (0.34 - 0.26 / 0.26) more horsepower than before.
* From all that power generated, part of it is used by the engine to run itself. The left over power is what you would measure on a dynamometer. The difference between what you would measure on the dyno and the workable power in the cylinder is the mechanical efficiency (ME). Mechanical efficiency is affected by rocker friction, bearing friction, piston skirt area, and other moving parts, but it is also dependent on the engine's RPM. The greater the RPM, the more power it takes to turn the engine. This means limiting internal engine friction can generate a large surplus in power output, and where in F1 the stress is on power, on the road it is also on fuel consumption.

These main optimization necessities are what makes Formula One engine design difficult. At the end of the line, an F1 engine revs much higher than road units, hence limiting the lifetime of such a power source. It is especially the mechanical efficiency that causes Formula One engines to be made of different materials. These are necessary to decrease internal friction and the overall weight of the engine, but more importantly, limit the weight of internal parts, e.g. of the valves, which should be as light as possible to allow incredibly fast movement of more than 300 movements up and down a second (this at 18.000 rpm).

[xpensive]

Ooops..missed that one, thanks tok.

If I get this right, 0.34 is the thermal efficiency in the cylinder (TE), why I guess what I'm looking for is the mechanical efficiency (ME), or the relationship between what I would measure on the dyno and the workable power in the cylinder?

[xpensive]

Ok, let's try it this way then; If TE is 34% as suggested by F1T above, crank power is 530 kW and total efficiency is say 25%,
it would mean that ME is 74% and combustion power is 720 kW.

Conclusively, frictional losses is 720 - 530 = 190 kW, is that reasonable, anyone?

[safeaschuck]

Terry, even if the modern 720 Hp F1 engine has an efficiency of 33%, which I seriously doubt, that would still mean that almost 1100 kW is wasted through exhaust and water/oil cooling.

X, do you think that figure (33%) is doubtful because it is to low or too high?

[xpensive]

None of it chuck. I misunderstood, as the F1T ecerpt xplains above, TE is likely to be as high as 0.34, but then you have to multiply that with the ME to get the total efficiency, which is more likely to be in the 0.20 - 0.25 range from what I have learned.

Again, my initial question was really what is a good number for ME?

[Touring23]

This topic reminds me of Spintron testing (acknowledging that Spintrons are employed more towards valvetrain dynamics):
http://circletrack.automotive.com/10879 ... index.html
http://www.diamondracing.net/M5LP-060800-SPIN.pdf
http://161.200.80.88/cu/eng/me/2103471/ ... cation.pdf

Try searching the internet for "engine friction". I don't believe anything useful will come from this approach:

Ok, let's try it this way then; If TE is 34% as suggested by F1T above, crank power is 530 kW and total efficiency is say 25%,
it would mean that ME is 74% and combustion power is 720 kW.

Conclusively, frictional losses is 720 - 530 = 190 kW, is that reasonable, anyone?

[xpensive]

Perhaps you have misunderstood the very purpose of this forum, touring?

[Touring23]

I'm not following you, nor do I mean any offence; I simply don't believe the method you presented has any fundamental basis. You asked a question and I'm doing my best to share what I know. If 25% ME has a supporting basis, please provide a link for my education and I will be humbly grateful. If we're just going to lob engineering factors around without basis, then I bid 15% and let's move on to the next guess.

A better approach would be to track down some Spintron or friction dyno data. Perhaps this information is available for motorcycle engines and can be scaled appropriately.
*edit* Try Figure 5 of this paper for example:
http://www.badgersnowmobile.com/uploads ... r_2009.pdf

A third approach, to lighten things up a bit around here, would be to ask Flavio.
It seems to me that current F1 teams probably employ a Friction Engineer and have a dedicated friction dyno. I wonder if the Shell site would be helpful.