Mercedes Power Unit Hardware & Software

[AJI]

...
This would allow a gain in power by bypassing the friction generating parts of the drive train eg. gearbox etc....

How do you bypass the gearbox? Am I missing something?

Edit: I just saw your updated post

[gruntguru]

In summary; within the rules, the ICE can continuously drive the MGU-H, which in turn can continuously drive the MGU-K (even when not harvesting), since the rules allow the MGU-H to provide the MGU-K with an unlimited amount of energy.

This would allow a gain in power by bypassing the friction generating parts of the drive train eg. gearbox etc.

The electrical losses of such a system would be much higher than the frictional losses in the drive train. Besides, an electrical machine in the final drive area would violate the rules, which allow only for an MGUK directly coupled to the engine.

[gruntguru]

I just saw this.

5.2.3 The MGU-K must be solely and permanently mechanically linked to the powertrain before the
main clutch. This mechanical link must be of fixed speed ratio to the engine crankshaft.

It makes bypassing the gearbox impossible. However the loophole for continuous driving of the MGU-H is still possible.

There is no need for continuous electrical driving of the MGUH by the K. The crankshaft power robbed by the K to do this is far more valuable than the waste heat that drives the turbine.

[GrandAxe]

I just saw this.

5.2.3 The MGU-K must be solely and permanently mechanically linked to the powertrain before the
main clutch. This mechanical link must be of fixed speed ratio to the engine crankshaft.

It makes bypassing the gearbox impossible. However the loophole for continuous driving of the MGU-H is still possible.

There is no need for continuous electrical driving of the MGUH by the K. The crankshaft power robbed by the K to do this is far more valuable than the waste heat that drives the turbine.

The loophole is that only mechanical connections to the MGU-H are regulated. Electrical connections are not, therefore the MGU-H can be directly driven by an alternator, not just the turbine. With that, the engine can continue running at as optimal speeds as possible, even under braking.

The electrical losses of such a system would be much higher than the frictional losses in the drive train. Besides, an electrical machine in the final drive area would violate the rules, which allow only for an MGUK directly coupled to the engine.

This part I got wrong, because I hadn't seen the relevant regulation; but I corrected the notion in my very next post.

[henry]

I posted on the equivalent Ferrari to this, that it should be possible to drive the MGU-H from the engine through an alternator (since there are no limits to electrical connections to the MGU-H, the rules only cover mechanical ones).

In summary; within the rules, the ICE can continuously drive the MGU-H, which in turn can continuously drive the MGU-K (even when not harvesting), since the rules allow the MGU-H to provide the MGU-K with an unlimited amount of energy.

This would allow a gain in power by bypassing the friction generating parts of the drive train eg. gearbox etc.

The only (somewhat contentious) limiting factor according to the discussion on that thread is the 120kw limit the MGU-K is allowed.

In the energy flow diagram, Appendix 3, the energy flows for ancillarys are shown. My reading of this is that:

Engine ancillaries are governed by the MGU-K restrictions.

Other ancillaries can only receive energy from the H or K or ES, they can’t send.

Would an alternator connected to the MGU-H not be constrained by this?

[dren]

There's this:

5.2.1 The use of any device, other than the engine described in 5.1 above, and one MGU-K, to propel the car, is not permitted.

But, engine ancillaries are allowed to drive or be driven by the MGUK, limited by 120kw. They are unlimited to be driven by the ICE.

5.13 Engine ancillaries :
5.13.1 All coolant pumps, oil pumps, scavenge pumps, oil/air separators, hydraulic pumps and fuel pumps delivering more than 10bar must be mechanically driven directly from the engine and/or MGU-K with a fixed speed ratio.

So, in theory you could drive the MGUK via an engine ancillary but it'd source its power from the ICE, so makes little sense to do it.

[restless]

But the idea of constantly sending energy to MGU-H will work only if ICE is very fuel efficient - so no need of saving fuel right?
Then of course they can have different modes - one with driving MGU-H and another with saving fuel on mind

[Zynerji]

Wasn't there speculation that RBR was using their alternator as a MGU back when they were frying them during races in the V8 era?

[dans79]

But the idea of constantly sending energy to MGU-H will work only if ICE is very fuel efficient - so no need of saving fuel right?

IMO, what you want to be doing is using the MGU-H to constantly drive the MGU-K. ALL the heat, flow, and noise coming out the exhaust pipe, is wasted energy. In an ideal world you would use the MGU-H to recover all of it and feed it to the MGU-K.

[sosic2121]

But the idea of constantly sending energy to MGU-H will work only if ICE is very fuel efficient - so no need of saving fuel right?

IMO, what you want to be doing is using the MGU-H to constantly drive the MGU-K. ALL the heat, flow, and noise coming out the exhaust pipe, is wasted energy. In an ideal world you would use the MGU-H to recover all of it and feed it to the MGU-K.

IMO no, not really. What you suggest is most energy efficient, but it’s not the quickest.
First, after a corner you use e-charger mode(for how long depends on your energy surplus).
After that you deploy full 120kW and H contributes as much as it can, without compromising crankshaft power(let's say 60 kW).
Then at the end of the straight H starts to charge ES(possibly in higher power mode than it was in turbo-compound mode), so you could do it again on the next straight.
IMO this is the quickest way.

[GrandAxe]

I posted on the equivalent Ferrari to this, that it should be possible to drive the MGU-H from the engine through an alternator (since there are no limits to electrical connections to the MGU-H, the rules only cover mechanical ones).

In summary; within the rules, the ICE can continuously drive the MGU-H, which in turn can continuously drive the MGU-K (even when not harvesting), since the rules allow the MGU-H to provide the MGU-K with an unlimited amount of energy.

This would allow a gain in power by bypassing the friction generating parts of the drive train eg. gearbox etc.

The only (somewhat contentious) limiting factor according to the discussion on that thread is the 120kw limit the MGU-K is allowed.

In the energy flow diagram, Appendix 3, the energy flows for ancillarys are shown. My reading of this is that:

Engine ancillaries are governed by the MGU-K restrictions.

Other ancillaries can only receive energy from the H or K or ES, they can’t send.

Would an alternator connected to the MGU-H not be constrained by this?

That's a valid point indeed. However, the MGU-K is allowed a 120kw input and a 120kw output; if input and output can be employed simultaneously, then a total 240kw can pass in and out of the MGU-K continuously (in ideal settings). This should allow 120kw to be piped through the MGU-K at a continuous rate (again, in ideal settings) from the engine to the MGU-H via the unlimited path between MGU-K and MGU-H. Of course, in less than ideal settings (e.g. optimisations for overall efficiency or other peculiarities), energy flows might be interrupted.

[PlatinumZealot]

Can't be simulataneously continuous. It won't be 120kW anymore! It is done intermittently.

[roon]

Twin stators, 120kW per. 120 in, 120 out, simultaneous, continuous. Might even split the ES to feed each side. Same for the H. No compromises on flow paths.

[subcritical71]

Twin stators, 120kW per. 120 in, 120 out, simultaneous, continuous. Might even split the ES to feed each side. Same for the H. No compromises on flow paths.

I may be missing something, but if you are doing this simultaneously, wouldn’t net output be 0 kW? Why do that at all and incur the losses (FIA set this at 0.95) to get net 0?

[Laserguru]

MGU-H can send as much power as it wants to the MGU-K as long as it goes through the MGU control unit (which has a maximum storage of 5KJ).

Assume Magneti Marelli's MGUH produces 90kW.

When thinking of a twin battery, as Ferrari seems to do according to FIA, you could imagine the MGU control unit as a secondary battery. So one big battery of 4MJ and a small one of 5KJ which is not called a battery (loophole).

MGUK energy to battery is max 2MJ per lap.
Battery energy to MGUK is max 4MJ per lap.
MGUH energy to MGUK is unlimited but must go through the MGU control unit.
MGUH energy to battery is unlimeted in both directions.

Maximum 120kW (160bhp) energy deployment limit between the MGU-K and the internal combustion engine. Full throttle for 4MJ/120kW=33.333s and your battery is depleted.

At any time the battery is fully charged with 4MJ you can harvest/store/deploy 5kJ from the MGUH through the MGU control unit ‘battery’ to the MGUK.
A propulsion boost of arbitrary 40bhp or 30kW is possible during 5kJ/30kW=167ms. It takes the MGUH 5KJ/90KW=56ms to harvest the excess energy.
This can be done at any circuit where you can harvest more than 2MJ per lap by MGUK and 2MJ by MGUH. Excess of 2MJ MGUH can be used in bits of 5kJ.

Imagine there are many corners and harvesting is plenty, every heavy breaking corner after lift and coast (battery 4MJ) you may ‘store’ MGUH energy in the MGU control unit for a whopping 167ms lasting 40bhp, deployed when accelerating out of a corner onto the straight. What does 40bhp for 167ms bring? A few kmh extra top speed? Faster acceleration out of the corners? Would be only a software update.