Mercedes Power Unit Hardware & Software

[ferkan]

Should be put in Hardware W07 thread:

[F1NAC]

So they're adding black, similar like in 2010? Nice explanation by AC about PU

[hurril]

I was referring to the MGUH and it's capacity to harvest at high rotational speeds.

I'm glad you bring this up, we all should know that there's only 3 real ways to increase electromagnetic induction. The regulations also create a few limitations on those 3 ways. For instance, the total system voltage cannot exceed 1,000 volts, I don't think anyone is using 1,000v yet, so there's scope for development there. We're also limited to 125,000 rpm for the speed of the conductor. We're not limited however by the geometry of the magnetic field, or the geometry of the conductor. So do you design a stubby but very wide MGU-H, or a long skinny one, or somewhere in between? Being that it's unlimited if your MGU-H is powerful and efficient enough at generating electricity, you could deploy indefinitely.

This is interesting. Those different geometry extremes - can you say something about which way their properties would affect the performance of the MGU-h?

[godlameroso]

One way to strengthen the magnetic field would be to stack strong permament magnets in a china doll type arrangment. Where you have a cylinder of magnets in a coil, and then a smaller cylinder of magnets isolated but inside the first cylinder. The more layers, the more the magnetic field is boosted

Kind of like these, but as permanent magnets instead of electromagnets.

[gruntguru]

The MGUK and MGUH are both capable of converting over 95% of the applied mechanical power into electricity. The only way to improve the performance of these components is to improve that number. Ask yourself:
"How much improvement might be possible?" (Answer. <5%)
"How much difference will that make?" (Answers <8hp (K) <5hp (H))

[godlameroso]

So putting more mechanical power into the MGU-H generates more electricity. Given the constraints in place how do you dump more heat energy to the turbine? Increase exhaust pressure? You can only burn so much fuel, so you have to get clever in how it's burned. How big of a compressor and turbine, and how much cylinder pressure do you need to dump 140-150kW to the MGU-H @ > 10,500rpm ?

[gruntguru]

There are a lot of calculations in the V6 thread. In a nutshell - more MGUH power requires higher exhaust back pressure, higher exhaust temperature or more exhaust flow.

Higher back pressure will reduce crankshaft power. Up to a point, the increase in turbine power will be greater than the loss in crankshaft power. At some point however, the law of diminishing returns kicks in and you lose more than you gain.

Higher exhaust temperature will also reduce crankshaft power (less energy transferred to the piston) or reduce massflow.

Higher mass flow requires a leaner AFR (fuel massflow is constant at 100 kg/hr), higher boost and more compressor power. Once again it is diminishing returns, increase in compressor power requirement approaches the increase in turbine power. Higher turbine and compressor efficiencies improve the situation - massflow can be increased further before diminishing returns cut in.

[OO7]

I guess that's why Renault mentioned that combustion design is where the large gains are hidden.

[hurril]

The MGUK and MGUH are both capable of converting over 95% of the applied mechanical power into electricity. The only way to improve the performance of these components is to improve that number. Ask yourself:
"How much improvement might be possible?" (Answer. <5%)
"How much difference will that make?" (Answers <8hp (K) <5hp (H))

Only 5%, sure, but those five percents are part of a feed-foward loop so some made up "area under the graph" can still be significant.

[GoranF1]

https://twitter.com/techF1LES/status/697777148845752320

[Bomber_Pilot]

Andy Cowell talking about the split turbo:

http://www.espn.co.uk/f1/story/_/id/147 ... -get-right

[ringo]

The MGUK and MGUH are both capable of converting over 95% of the applied mechanical power into electricity. The only way to improve the performance of these components is to improve that number. Ask yourself:
"How much improvement might be possible?" (Answer. <5%)
"How much difference will that make?" (Answers <8hp (K) <5hp (H))

Only 5%, sure, but those five percents are part of a feed-foward loop so some made up "area under the graph" can still be significant.

Also it's the motor characteristics that need to match as well as possible with your turbine and engine. The geometry will have some influence on the characteristics. The best efficiency point for two very different motors can be the same peak but the motor characteristic over a range of speeds may tell different stories.

[godlameroso]

The MGUK and MGUH are both capable of converting over 95% of the applied mechanical power into electricity. The only way to improve the performance of these components is to improve that number. Ask yourself:
"How much improvement might be possible?" (Answer. <5%)
"How much difference will that make?" (Answers <8hp (K) <5hp (H))

Only 5%, sure, but those five percents are part of a feed-foward loop so some made up "area under the graph" can still be significant.

Also it's the motor characteristics that need to match as well as possible with your turbine and engine. The geometry will have some influence on the characteristics. The best efficiency point for two very different motors can be the same peak but the motor characteristic over a range of speeds may tell different stories.

Most industrial 150kW generators are huge bulky devices, and they only work at much lower RPM ~1,800rpm, the F1 generators are allowed to spin up to 125,000. Getting the generator to be 95% efficient over a wide range of speed is easier than getting the turbine to be that efficient over a wide range of speed. Particularly since the generator is also a motor.

The point is, making a good MGU-H is a huge engineering exercise because you're downsizing something huge and clunky with a narrow efficiency window, into something tiny, and light, that requires a huge operating window.

[OO7]

The 2014 compressor is very different - a bit shallower? but wider?

2014
http://www.pitpass.com/images/engines/8 ... edes02.jpg

2015
http://cdn-2.motorsport.com/images/mgl/ ... hybrid.jpg

Judging by the following images, it seems to me that it is actually the 2014 PU106A studio images which featured a faux compressor covering:
2014 PU106A Studio Picture -


2014 PU106A -



2015 PU106B -

[Webber2011]

https://twitter.com/andrewbensonf1/stat ... 6847038464

Thanks for that
For those of us like me, who aren't as technically minded as some of you, I found that well written and easy to understand.

Cheers.