Mercedes Power Unit Hardware & Software

[henry]

I also note that at 2:35 in the video it gives an efficiency value of 96%. I assume this is total energy flow efficiency, 1MJ in is 960kJ out.

I think this would set an upper threshold on the total potential energy flow during a race. If we arbitrarily assume the battery starts the race with the upper SOC at 16MJ and ends with the lower SOC at 4MJ then that would allow total flow of around 200MJ or 3.5MJ per lap in a 60 lap race. The ES would make up the 8MJ lost to efficiency.

I’ve assumed they would want to cycle the battery between 80% and 20% to maintain battery life and that the physical battery capacity is around 20MJ.

[Juzh]

Interesting that Andy Cowell mentions the ES can output 200kW in this video (around 3mins in) https://www.youtube.com/watch?v=3C1-KDkGDUA
Obvious comment is obvious, but the ERS K can deploy 120kW, so you'd have to assume the other 80kW is for the H.

I saw that same video and came here to ask exactly where the extra 80 kw seem to be coming from, completely forgot about the mgu-h thanks for pointing it out.

[godlameroso]

It's safe to say that getting those 200kW to translate into power at the wheels is the engineering and technical challenge.

120 can go directly to the wheels via the K, and the other 80 has to be transmitted indirectly.

[izzy]

It's safe to say that getting those 200kW to translate into power at the wheels is the engineering and technical challenge.

120 can go directly to the wheels via the K, and the other 80 has to be transmitted indirectly.

120 is the limit from the battery to the K but they can go unlimited H directly to K. This is why a big fat turbo is a good idea! Tho of course they're burning petrol to do it but somehow it works out more efficient some of the time, like top speed I think

[henry]

The 200kW figure quoted is ES drain. It happens in eboost mode with 120kW to the MGU-K and 80kW to the MGU-H to drive the compressor. The turbine supplements the 80kW to achieve the full power needed by the compressor.

The ES may also be charged at around 190 kW by combined input from the K and H at the end of straights. I think Ferrari may use this but possibly not Mercedes.

For completeness Honda’s extra Harvest route can also cause a similar charge rate.

[roon]

It's safe to say that getting those 200kW to translate into power at the wheels is the engineering and technical challenge.

120 can go directly to the wheels via the K, and the other 80 has to be transmitted indirectly.

120 is the limit from the battery to the K but they can go unlimited H directly to K. This is why a big fat turbo is a good idea! Tho of course they're burning petrol to do it but somehow it works out more efficient some of the time, like top speed I think

You seem to be making a point about energy and compound and extra-harvest modes, but subcritical, henry, and godlameroso are talking about power. The K must never transmit more than 120kw to the wheels. He's saying the remaining 80kw draw (power) of the reported 200kw peak draw from the ES must reach the wheels indirectly if it is not solely powering the compressor.

Somewhat related (if anyone knows or can comment): I've been wondering, can the MGUH cut off or even reverse flow through the turbine when the wastegates are open? Path of least resistance being the WG pipes, the volute path being restrictive, combined with the centrifugal action of a spinning turbine. The blade shape may prevent backflow but it may at least permit to act as a choke or throttle if the pressure differential mismatches turbine speed at WG opening. In which case full exhaust flow during eboost/supercharger mode is flowing through the wastegate pipes, perhaps to aerodynamic effect.

[izzy]

The 200kW figure quoted is ES drain. It happens in eboost mode with 120kW to the MGU-K and 80kW to the MGU-H to drive the compressor. The turbine supplements the 80kW to achieve the full power needed by the compressor.

The ES may also be charged at around 190 kW by combined input from the K and H at the end of straights. I think Ferrari may use this but possibly not Mercedes.

For completeness Honda’s extra Harvest route can also cause a similar charge rate.

Oh yes, from the ES, i needed to watch the video lol &@roon

[henry]

Somewhat related (if anyone knows or can comment): I've been wondering, can the MGUH cut off or even reverse flow through the turbine when the wastegates are open? Path of least resistance being the WG pipes, the volute path being restrictive, combined with the centrifugal action of a spinning turbine. The blade shape may prevent backflow but it may at least permit to act as a choke or throttle if the pressure differential mismatches turbine speed at WG opening. In which case full exhaust flow during eboost/supercharger mode is flowing through the wastegate pipes, perhaps to aerodynamic effect.

My understanding is that the wastegates act as pressure relief to reduce the exhaust manifold pressure. Not all of the exhausts gases go down the wastegate pipes. Some, carrying the kinetic energy of the blowdown at the opening of the exhaust valve do not go down the wastegate pipes. That kinetic energy continues to act on the turbine and provides some of the power required to drive the compressor.

If it takes 120kW to drive the compressor and the blowdown provides 40kW we get to the 80kW that the MGU-H needs to provide. These are ballpark figures not facts.

This is one more multi dimensional trade off that needs to be made in the design of these power units which is made not on the basis of absolute power numbers but lap time at the mix of circuits on the calendar. Mercedes seem to be very skilful at these trade offs not only in their PU, where they enjoyed an absolute rather than marginal advantage in the early years, but also in chassis, aero and operations.

[pb6797]

Puffs of smoke from the two Mercedes in the pitlane again.

[roon]

My understanding is that the wastegates act as pressure relief to reduce the exhaust manifold pressure. Not all of the exhausts gases go down the wastegate pipes.

Perhaps they could. If turbine speed exceeds the pressure gradient/differential across the turbine. The turbine will be driven to an artificially high speed by the MGUH while the wastegates are open. At a certain angular velocity within a low pressure gradient the turbine blades should begin to stall or act as a pump.

[henry]

My understanding is that the wastegates act as pressure relief to reduce the exhaust manifold pressure. Not all of the exhausts gases go down the wastegate pipes.

Perhaps they could. If turbine speed exceeds the pressure gradient/differential across the turbine. The turbine will be driven to an artificially high speed by the MGUH while the wastegates are open. At a certain angular velocity within a low pressure gradient the turbine blades should begin to stall or act as a pump.

Maybe. But the turbine and compressor have to be matched to operate on the same speed range so I don’t think there will be much opportunity to overspeed the turbine. And they need to use the turbine to drive the compressor alongside the MGU-H otherwise the ES drain will be even more than 200kW. So stalling doesn’t seem like a good idea.

It seems to me that over time the area of the wastegate pipes has reduced. The Honda pipes look particularly small. So maybe they’re encouraging more flow through the turbine and looking for the balance between extra power in eboost versus overall turbine recovery and hence MGU-H output.

[roon]

...so I don’t think there will be much opportunity to overspeed the turbine.

Its not as though there will be a choice, however. When wastegates open, the MGUH will still be driving the compressor.

[henry]

...so I don’t think there will be much opportunity to overspeed the turbine.

Its not as though there will be a choice, however. When wastegates open, the MGUH will still be driving the compressor.

When the wastegates open the compressor speed, and hence output, will be maintained. The pressure differential across the turbine will fall and it will make less power from gas expansion than it did before the wastegates open. Maybe this will stall the flow. However it will continue to make power from the kinetic energy in the exhaust pulses.

It seems unlikely that the MGU-H is driving the turbine as a pump since this would take even more power and hence energy drain on the ES.

Having said that I have no expertise in turbo-machinery, and the papers I’ve read on the subject don’t consider this use case, which I guess is unique. So there may be effects that come in to play that change the effect of wastegate operation from what I currently understand.

[godlameroso]

Roon makes an interesting point, and possibly a reason why an oversized compressor is desirable.

Both IMAP, and EMAP can be controlled via the various pressure charging and relief systems. So the bias between exhaust back pressure, or intake pressure can also be varied. Playing with this bias can serve for more outright engine power, or increased MGU-H harvesting.

[henry]

Roon makes an interesting point, and possibly a reason why an oversized compressor is desirable.

Both IMAP, and EMAP can be controlled via the various pressure charging and relief systems. So the bias between exhaust back pressure, or intake pressure can also be varied. Playing with this bias can serve for more outright engine power, or increased MGU-H harvesting.

If I understand it correctly @Roon’s point is that it might be beneficial to ensure all the exhaust gases go through the wastegate pipes() since this high energy mass flow could be used to improve downforce, and hence traction in the early acceleration zone. He offers a mechanism that might implement this. I don’t understand the mechanism but that’s no reason to suppose it would not work. An alternative would be to treat the wastegate as a diverter rather than pressure relief. I have seen very few clear images of wastegate installation and those I have seen look like pressure relief. But again that’s no reason for it not to be happening.

What is your characterisation of an oversize compressor and why would that be beneficial in @Roon’s scenario?