2014-2020 Formula One 1.6l V6 turbo engine formula

[747heavy]

@ autogyro

A real world EV LiIon battery will provide nominally some 25% of the theoretical energy capacity or 70 – 120 Wh/kg instead of 410 – 450 Wh/kg.
This translates into a Lithium requirement of at least 320 g of Lithium (1.7 kg LCE) per kWh of available capacity.
In addition, Lithium has to be added to this for the electrolyte, irreversible capacity loss and capacity fade.
EV batteries will be 25% oversized to account for capacity fade.
Then allowance has to be made for processing yields of an estimated 70% from the
raw technical grade Lithium Carbonate plus inevitable losses in the use of high control purity Lithium Carbonate in the manufacture of the battery components themselves.
LiMPO4 batteries operate at lower voltage than LiMO2 and therefore induce a further increase.
If one therefore allows 400 g of Lithium (2.1 kg LCE) per battery kWh with a 70% processing yield to produce that, an initial 3 kg of raw technical grade Lithium Carbonate will be required per kWh of final usable battery capacity.
At 3 kg raw technical grade LCE per kWh, current global production of some 100,000 tonnes raw LCE would be sufficient, if available, for some 2 million 16 kWh batteries per year. Even at an optimistic 2 kg LCE per kWh assuming very high purity yields, production would be sufficient for only 3 million 16 kWh
PHEV batteries per year.

full report here

[747heavy]

The prismatic cells have a specific energy/weight ratio of 91 Wh/kg=328 kJ/kg.
We plan to use 4 MJ. This means we need to install 12.21 kg of cells.
At 0.4 kg/cell we need 31 cells.
If we arrange the 31 prismatic cells in series we create 102.3V electric voltage. 31 cells have 12.4 kg battery weight.
If we are allowed to charge/discharge at any rate we would be done now.
A complete cycle of the charge/discharge takes 300s but we have only 40s.
It means we have to split our electric current to eight parallel cells to run within thermal specification.
We are forced to install an array of 31x8 cells which brings the total up to 248 cells.
Our total weight of the battery array would be 99.2 kg.

not sure I can follow your calcs WB, as far as the charging goes.
According to this graph from the A123 website, their best cells charge ~90% in 5 minutes (300 sec), so how can you discharge/charge a full cycle at 300sec. ?

[ringo]

What's happening exactly with the F1 KERS charging almost instantly, in comparison to the 5 minute charges of the other batteries?
I am following the discussion, but i am not sure where it's going.
Is it that the F1 car needs to carry more batteries than meets the capacity becuase of the high charge and discharge rates?

[WhiteBlue]

The prismatic cells have a specific energy/weight ratio of 91 Wh/kg=328 kJ/kg.
We plan to use 4 MJ. This means we need to install 12.21 kg of cells.
At 0.4 kg/cell we need 31 cells.
If we arrange the 31 prismatic cells in series we create 102.3V electric voltage. 31 cells have 12.4 kg battery weight.
If we are allowed to charge/discharge at any rate we would be done now.
A complete cycle of the charge/discharge takes 300s but we have only 40s.
It means we have to split our electric current to eight parallel cells to run within thermal specification.
We are forced to install an array of 31x8 cells which brings the total up to 248 cells.
Our total weight of the battery array would be 99.2 kg.

not sure I can follow your calcs WB, as far as the charging goes.
According to this graph from the A123 website, their best cells charge ~90% in 5 minutes (300 sec), so how can you discharge/charge a full cycle at 300sec. ?

You can figure it out if you know the law of splitting an electric current to multiple parallel pathes with the same resistance. We need to charge and discharge 4 MJ in 80s. That means 8 MJ in 80 s. That is the same as charging 4 MJ only in 40s, or charging 1MJ in 10 s. Our effective power for the charge is 100 kJ/s or 100 kW. 100 kW is the same as 100 kVA.

I thought our cell voltage would be 3.3 V (@ 0.2C) but I later realized that our cell voltage is only 2.75 V. (@10C). The total serial voltage should be generated by a serial arrangement of 111 cells for a total of 305.25 V (and not 102.3V from 31 cells). I decided to do a higher voltage to cut down on cable diameter.

Because we lost cell voltage but gained serial voltage we need to have a different number of parallel cells. Let us use 3. This means our current is 3x10x10.9A=327A. The power is 327Ax305.25V=99.82kW. So with a more detailed computation with more accurate data we get to 111x3=333 cells with 133.2 kg. Not quite so nice as in the first shot when I computed with erroneous cell voltages and too much current.

[xpensive]

This is discussion is most interesting to me, as being a mechanical engineer I don't have a clue of what you are talking about in technical detail, but I grasp the numbers.

Anyway, what I wish to see is a device that can store 1 MJ within 2-3 sec before the corner, as an added means of braking if you wish, hold it for a sec or two, then unleash the MJ through the coming straight. Csan this be done with batteries?

I could easily design a hydraulic device that can repeat that sequence forever, but it would hardly fit in an F1 car.

[autogyro]

The problem of fast charging in such a short time, is why I suggested spinning up a flywheel electricaly, with it coupled to a compressor, so as to maintain compressor (turbo) rpm for max boost on the following strait and to then trickle charge the battery packs from the charged flywheel at a slower rate.
I added the possibility of re-directing the airflow to and from the compressor when the ic is off load under braking, perhaps to increase DF.

WBs multiple charge paths in this system, would also allow a split between charging a fairly light flywheel and a smaller number of cells.

There are also other battery chemicals in development potentialy better and more in balance with the environment than Lithium.

[autogyro]

The next stage would be to place the compressor stage of the turbocharger in the gearbox casing in place of the clutch and couple it to a new generation stepped gearbox (ESERU), which would halve the weight of the geartrain reduce its size and do away with the need for a seperate geared rear MGU.

The ESERU would also give the 7 stepped ratios required by the F1 regulations plus a range of seperate and combined ratios for both harvesting energy and applying it.

[747heavy]

@X
good & valid question IMHO, as far as I can see it at the moment, the way WB is going you would only use a tiny fraction of the availible battery capacity, to achieve the quick charge times required.

2.75V looks very low as a cell voltage to me.





http://www.fmadirect.com/support_docs/item_1229.pdf


http://www.iis-b.fraunhofer.de/de/profi ... wenger.pdf

[WhiteBlue]

@X
good & valid question IMHO, as far as I can see it at the moment, the way WB is going you would only use a tiny fraction of the availible battery capacity, to achieve the quick charge times required.

2.75V looks very low as a cell voltage to me.

I have used the data from the quoted 2007 presentation. On page 17 you see the cell voltage of the 11Ah prismatic cell at different charge multipliers. I have used the brown 10C line which gives you 10x10.9A and 2.75V @10 Cell Capacity/Ah.

In Jumbo's second graph "A123 Multi C test" you get roughly the same data if you use the 10.6C line.

With these data I have simply optimized my cell array for 100 kVA. I'm quite prepared to use higher cell voltage if there is a newer product documentation that can be applied consistently. I'm not so happy with the heigh weight either but I tried to honestly use consistent manufacturer data without making any assumptions about improvement.

It is clear from this computation that your battery size is not determined by the energy capacity but by the size that supports sufficient charging/discharging rate. The nanophosphate cells with 109A max current are pretty special in that regard.

I have quickly checked for the physical size. You are looking at 60L of volume which is not such a problem as the battery weight because you can arrange the cells at very different shapes. LxWxT is 102mm x 71mm x 25mm.

The PDF document from Jumbo's post is adamant that maintaining battery temperature below 60°C (140°F) is basolutely crucial to the life cycle of the batteries. You can take peak discharge/Charge to 33C - which we do in my application - if you maintain a good thermal regime. These batteries require good fluid cooling under the conditions we are thinking to use them. We can expect approximately 500 laps operating time without too much degradation. The batteries would be limited to 2500 km use. You would need 2 batteries per engine, which I think is acceptable.

[xpensive]

As hydraulic power is pressure times flow, 1 MJ in 2 seconds would mean 500 kW, or 25 liters per second at 200 Bar.

As this has been discussed before, I recall that Newey designed something of the kind for McLaren, didn't he?

[WhiteBlue]

McLaren 2009 KERS used custom A123 cells with 20kW/kg

There are obviously custom systems being used in F1 which use different data than those published for general purpose use. I guess we will probably learn a bit more about the latest specs when the 2011 KERS will come out in February. At the moment I just have to assume that my weight estimates for a 2013 4MJ system is very conservative and exceeded by custom systems.

[747heavy]

borrowed from the similar autosport thread:

Actually all IC engines achieve _optimal efficiency_ at piston speeds around 10-15m/s. Eg in current turbo IC engines peak efficiencies occur at 2-3000rpm. Remember that with limited fuel flow efficiency is king. This is due to combined effects of ring and piston friction and port flow pressure losses (proportional to the square of piston speed). The Ø88mm bore implies 66mm stroke and 42m/s peak piston speed at 12000rpm (same as found in current F1 engines, S2000, and high speed Audi V8's - this represents the absolute limit for IC engines, and is certainly a long way from being an efficient operating point - but speed is they only way to increase power when you can't boost or increase displacement.

Turbos don't care what speed the engine operates at - and so are not really part of the optimisation - though boosting may tip the case slightly towards higher piston speeds.

Operating the turbo at 6000rpm would halve the piston speed and thereby help a lot, but ultimately in a fuel flow limited class with a desire for maximum efficiency at max power designers would be targetting a stroke of probably about 50mm (bore Ø100 or more) to give optimal piston speeds in the 6-7000rpm range (low speed also reduces the fuel pressure pumping losses by reducing pressure required for atomisation)

So if the FIA wants glamorous high speed engines in a fuel flow limited engine with no boost limit they will have to give the designers the freedom to use a larger bore, as otherwise the only sensible option available to the designer is to use higher boost to achieve their power goals at lower speeds.

You are also wrong about the miller cycle. Until recently I worked for a high-end turbo engine consultancy and development firm that specialised in supercar engine developement, and miller cycle in a Gasoline turbo engine can give +5-10% efficiency (=+5-10% power) gains. That is a race winning advantage that cannot be ignored.

With regard to stratified charge and lean burn - that makes zero sense as it is less efficient in a high power high efficiency application. You must remember that most IC engine technologies (like lean burn) are focused on improving part load efficiency by allowing de-throttling, hence the drive towards down-sizing engines, turbocharging, down-speeding, VVT (for optimised internal EGR), higher compression ratios (that have biggest impact when throttling). This is all simply not relevant to wide open throttle race engines.

Reading some other comments: The key points with DI are that it increases volumetric efficiency (gaseous fuel does not displace air going through inlet port) and overall engine efficiency (evaporation of fuel in air rather than on hot inlet valves reduces charge temperature in the cylinder allowing higher compression ratio, and also increases volumetric efficiency slightly). The biggest problem is that it is difficult to integrate the injector and spark plug in a central position without costing valve area (This is a big problem given desire for efficiency) - and this may motivate a spray guided solution with the injector outside the inlet valves

Turbos will be on exhaust port side (not in some other fanciful position) with the shortest possible individual exhaust runners of equal length meeting at scroll in order to get the highest efficiency (preserving gas velocity from exhaust port) and lowest back pressure.

On reflection I won't be surprised if the FIA also institutes a boost pressure limit on top of the fuel flow limit in order to force higher engine speeds, but regardless this should also be done with bores >88mm being allowed - I really can't see the point of limiting the bore.

@WhiteBlue

I (think I)understand what you did to achieve the required engergy absorbtion rate in your accupack, nevertheless I´m sure your battery is not fully charged at this point.
As we see from the A123 data, they need ~5min to achieve 90% charge/capacity.

This is allready quite impressive for an accu/battery (12C), but means you wont charge to full capacity in 10 sec, not even close, independent of how many cells you connect in series and/or parallel, therefore only utilizing a fraction of the available capacity. IMHO

@Xpensive
I´m not 100% sure, I fully understand your hydraulic system.
How would you store the pressure?
As I see it you would need a "springing medium" like air/nitrogene/gas to compress against.
Can´t see how a incompressible fluid could store energy.
Would it in this case, for weight reasons, not be better to use only a pneumatic systems?

Which I guess brings us towards autogyros idea.
I can see a option, to just store excessive boost pressure from the turbo and use it later on when rpm/turbo speed is low to improve performance.
Similar to the systems used (in the past) in the WRC.

[ringo]

I wouldn't use a centrifugal compressor for gyro's idea. I would use 2 small pistons to compress the air. Pitson compressors can gain much higher pressure ratios.
During normal operation, the compressor valves are open, or maybe some way of operating in a vacumm, and the pistions turn freely. Or maybe a clutch disconnecting the compressor from the drive train.
Under braking, the clutch is engaged, and the decelerating wheels turns the compressor and air is compressed and stored in a tank. It's almost like a jake brake.

When the energy is required on the sratight, the air is released into the compressor turning the pistons. It would be like an air car hybrid.

[youtube]http://www.youtube.com/watch?v=uVIwropRMME[/youtube]

The compressor could be designed with discharge rate in mind instead of range, for a power suitable for racing. I would have to work out what mass of air would be required.

[WhiteBlue]

I wouldn't use a centrifugal compressor for gyro's idea. I would use 2 small pistons to compress the air. Pitson compressors can gain much higher pressure ratios.

Ok, now I start to understand what could be done with a compressor. The point is we would not be talking 3 bar like a turbo compressor but 200-400 bar like a heavy duty industrial or diving compressor.

Perhaps one cold actually use the ICE in the breaking phase to compress air and store it in a bottle for later use for boosting. The exhaust gas can then be totally used 100% for some time to drive a turbo compounder while the intake pressure would be fed from the air bottle.

[WhiteBlue]

@WhiteBlue
I (think I)understand what you did to achieve the required engergy absorbtion rate in your accupack, nevertheless I´m sure your battery is not fully charged at this point. As we see from the A123 data, they need ~5min to achieve 90% charge/capacity.

This is allready quite impressive for an accu/battery (12C), but means you wont charge to full capacity in 10 sec, not even close, independent of how many cells you connect in series and/or parallel, therefore only utilizing a fraction of the available capacity. IMHO

You need to understand that the parallel concept avoid discharging the battery between 10% and 90%. By coupling three series of cells in parallel you deplete the combined cells only by a third of 80%. It means you will only take a third of the time to charge them back to full 90% capacity. Effectively you only skim the top of the capacity and carry the rest around. It is a big waste but the only way you can go if your charge/discharge rate isn't high enough.

Principally adding serial cells builds voltage and adding parallel cells builds current. To build power you multiply current and voltage. This why the weight increases so rapidly. The only way to improve is having chemical systems that run higher voltage at higher or equal current. The current chemistry is temperature restricted at 60°C. If they find a reaction that tolerates higher temps at equal thermal efficiency and heat conduction they can create a bigger heat flow and from the driving temperature difference and also bigger charging power.