TERS : Thermal Energy Recovery System

[NL_Fer]

Not allowed.

You have to adopt a monotonically raising torque with increasing torque demand, and cannot shape it according to the actual used gear (which would be mandatory to map the PU, in fact limiting the torque in the lower gears) . These are the main articles in the regulations:

5.5 Power unit torque demand :
5.5.1 The only means by which the driver may control acceleration torque to the driven wheels is via a single chassis mounted foot (accelerator) pedal.
5.5.2 Designs which allow specific points along the accelerator pedal travel range to be identified by the driver or assist him to hold a position are not permitted.
2015 F1 Technical Regulations 24/89 3 December 2014
© 2014 Fédération Internationale de l’Automobile
5.5.3 The accelerator pedal shaping map in the ECU may only be linked to the type of the tyres fitted to the car : one map for use with dry-weather tyres and one map for use with intermediate or wet-weather tyres.
5.5.4 At any given engine speed the driver torque demand map must be monotonically increasing for an increase in accelerator pedal position.
5.5.5 At any given accelerator pedal position and above 4,000rpm, the driver torque demand map must not have a gradient of less than – (minus) 0.045Nm/rpm.
5.6.7 Homologated sensors must be fitted which measure the torque generated at the power unit output shaft and the torques supplied to each driveshaft. These signals must be provided to the ECU.

5.5.5 forbids an intrinsic traction control, in case the driver doesn't alter pedal position (=torque demand) and the engine speed raises.

The torque at PU is always with MGU-K.

Still at Alonso's gp2 moment, ERS was depleted, so torque was dropped at that moment. And that is allowed. So could it not be possible to program the system not use mgu-k at certain moments, as an "efficiency strategy" to save ERS power.

[godlameroso]

The MGU-H is free to be developed correct? It is the one unlimited harvesting system, that means that pouring money into developing the MGU-H would probably give the biggest gains.

[Tommy Cookers]

this below is a great read, including exhaust recovery
look at Advanced Topics and Tech Paper Download (and stuff by Paul Lamar)
Paul Lamar was a racecar wing pioneer incl work for Chapparal and Pontiac

http://www.rotaryeng.net/

@ pgfpro ?
Paul L has been imo a bit optimistic
specifically on what he seems to call dumping loss across exhaust valve
and the extent to which the Mazda Rotary could escape this
(imo no engine can escape it, even with infinitely large/infinitely fast port opening, unless there is high exhaust pressure)
the rotary engine is similar to supercharged aircraft engines ie the CR is rather low so the exhaust energy is unusually high

looking again at the energy balance figs in the Wright TC paper
in lean cruise supercharger power was relatively high and recovery low (implying poor recovery in F1 below full power)
apparently even at this very lean mixture 4.6% of the fuel's energy was lost in dissociation to CO and CH4
(1800 rpm, inherently high temperatures, and maybe the high aromatic content of Avgas causing this ?)
overall efficiency was 32.5% (about 9.5% of this from recovery) with a CR around 7
so if they had made a cruise-only engine with cruise-optimised CR 35% efficiency would be possible ?
(and some of these values seem lower than would correspond to the official BSFC figures and Wright's brochure)
interestingly, Wright seem to have run eg at CR=4 and about 32 bar bmep

Engine Compounding for Power and Efficiency (Pierce&Walsh) SAE quarterly transactions Vol 2 No 2April 1948 looks quite useful
and the Barber-Nichols turbine brochure
PL shows NACA paper saying the (basic not entrainment ?) exhaust jet propulsion effect adds 2% to power at 200 mph eg to F1
ok, we knew this
and there's a paper on Detroit Diesel turbocompounding, worth up to 5% free power and proportionate increase in efficiency

btw - somewhere PL says that car engines (for high capacity with minimal length) often skimp wrt crankshaft fillet size
so have inadequate fatigue life at constant high power/rpm eg for aviation use
that Orenda (car-type 'big-block' V8 engine for aircraft) found this, even (after ?) spending $100000+ on dies for improved crankshafts ?

[pgfpro]

Thanks TC

Great information!!!

[Brian Coat]

TC: A road car crank designer won't stress the fillets based on a 'limited' duty cycle because the 4 stroke engine, making max power at (say) 6000 rpm, will reach 10^7 combustion con rod stress hits in only 55 or so hours ...

[gruntguru]

Compromising on crank fillets is more likely on multi-bank engines due to shorter crankshaft for a given number of cylinders.

[godlameroso]

Does anyone have any good guesses as to what kind of power the MGU-H system is capable of generating? If by Honda's estimates, they were deploying all 4MJ and were not able to sustain it for an entire lap, how much are the other manufacturers deploying with their power unit? It has to be more than 4MJ, otherwise they would not be cruising around the McHondas. I honestly wouldn't be surprised if Mercedes or Ferrari, or even Renault are deploying somewhere in the range of 7-8MJ total, 4 from the ES, and the rest directly from the MGU-H to the MGU-K.

[godlameroso]

Perhaps the only reason that Mercedes split the turbo wasn't to keep the compressor away from the turbine but because they have a massive MGU-H that occupies most of the space within the V. There really aren't many ways to make a more powerful MGU-H, you either use stronger magnets, or you add more of them, but you have to do it in a way that doesn't stall the turbine, and thus cause unnecessary back pressure.

[M greenhouse]

Would the increased mass of the split turbo in the Merc act as a flywheel?
So the mgu-h doesn't have to act for anti-lag and also be able to generate more for the mgu-k.

[godlameroso]

The MGU-H generates electricity by placing a drag on the turbine shaft, thus slowing the turbine down. The MGU-H will recover more energy if the turbine shaft has more angular momentum, this is achieved by increasing the volume of exhaust gases, or increasing the temperature difference between the inlet and outlet of the turbine. How does one maximize the amount of exhaust gases? By being as efficient as possible with combustion, that means having as high a cylinder pressure as possible without detonation. The rest is minimizing the distance of the exhaust plumbing and insulating it before the exhaust to keep it as hot as possible. If the MGU-H drags too much it can cause compressor surge.

[gruntguru]

Would the increased mass of the split turbo in the Merc act as a flywheel?
So the mgu-h doesn't have to act for anti-lag and also be able to generate more for the mgu-k.

Inertia of the rotating assembly needs to be as low as possible for best response (boost needs to be varied rapidly for different speeds and loads). The small amount of energy that might be stored in an unnecessarily heavy mguh would be of no benefit.

[l4mbch0ps]

Inertia of the rotating assembly needs to be as low as possible for best response (boost needs to be varied rapidly for different speeds and loads). The small amount of energy that might be stored in an unnecessarily heavy mguh would be of no benefit.

That's certainly true for a traditional turbo - however with the HERS systems they essentially have boost on demand. Not only can they spool the turbo in excess of the acceleration of the exhaust gasses, but they can just keep the turbo at a speed where they know it will be beneficial based on lap position, throttle position, engine mode, etc.

Obviously everything is a tradeoff - but the coupling of the mgu with the turbo really changes the fundamentals.

[gruntguru]

"boost needs to be varied rapidly for different speeds and loads"

Whether the turbo needs to be accelerated to increase boost or decelerated to reduce boost, the "response" will be better if the rotational inertia is low.

High rotational inertia will affect the response whether the acceleration is provided by a turbine, a MGU or both.

[godlameroso]

The mgu-h can be clutched, so as long as the turbine shaft has little inertia, it'll have good response. But then you have to ask yourself why it's beneficial to have a large compressor, with the inherent inertia that comes with it. Honda tried using a small low inertia compressor, and it did not work for them. Meanwhile the best engines all have very large compressors.

[Per]

Following up on the discussion in the Honda PU topic on MGU-H harvesting energy that is provided directly to the MGU-K, an energy flow which is unlimited.

Would it be technically possible and legal to do it the other way around as well, by any significant amount? The MGU-K is allowed to harvest 2MJ per lap straight into the ES. But there is no limit for energy going from MGU-K to MGU-H. So if you have filled up the 2MJ for that lap, you could use the MGU-K to boost the MGU-H during braking.

Of course this would only make sense if the MGU-H can do something useful with that energy. I don't know a lot about electric motors so my question is: would it be possible to have an MGU-H with two windings with separate control, one 'connected' to the ES and the other to the MGU-K? If so, would it be possible to have one winding on generator mode (using energy coming from MGU-K) and the other on harvesting mode to store this energy into the ES?

If this would work and it is legal, you could potentially recover more kinetic energy than 2 MJ per lap.

I'm sorry if this idea sounds ridiculous, I just don't have enough knowledge about electric motors to know any better.