Honda Power Unit Hardware & Software

[Abarth]

Direct drive is more efficient than geared and 100000rpm shouldn't be a problem for an proper designed electric motor.

It's more about the possibility to stay within the 125k rpm with the MGU-H while spinning the turbo higher.

That enables to reduce turbine and compressor diameter.

[Brian Coat]

re:

"It's more about the possibility to stay within the 125k rpm with the MGU-H while spinning the turbo higher."

What are the attractions of a teeny-weeny >125,000 rpm geared turbo system?

Package and inertia will be better?

Other losses will increase?

[Abarth]

re:

"It's more about the possibility to stay within the 125k rpm with the MGU-H while spinning the turbo higher."

What are the attractions of a teeny-weeny >125,000 rpm geared turbo system?

Package and inertia will be better?

Other losses will increase?

I do not think inertia is important.

Packaging is, obviously.

Operating point is pretty stable too, so it boils down to efficiency of different size/speed ratios.
If gearing is required (i.e. if the turbo shall spin >125k), there is some loss too.

Maybe someone can jump in to quantify any efficiency differences of different sizes of compressors @ same PR and mass flow.

[Nickel]

I do not think inertia is important.

I'm pretty sure inertia is everything for driving the gu-h

[godlameroso]

Think of it this way. You have two ways to make power in an engine, high revs with relatively low displacement, or low revs with relatively huge displacement. Generating power at the MGU-H follows a similar logic. You either have a huge compressor/turbine, that imparts a boatload of torque to the MGU-H, or you have a teeny tiny compressor/turbine that spins at a dizzying pace, now if you apply a gear reduction to the MGU-H torque is increased, but at lower rotational speed. Conversely if your compressor/turbine is large and spins slower in order to achieve the required boost pressure(relative to max MGU-H speed), you could overdrive the MGU-H. It's obvious that using an oversized turbo and relying heavily on the MGU-H to keep the turbo from overboosting, and using the wastegates as a fail safe is the way to go. If the inverse were true, Honda would have been far more successful, however they may be able to make their concept work. We'll see in a few months.

[Abarth]

I do not think inertia is important.

I'm pretty sure inertia is everything for driving the gu-h

Can you elaborate?

[djos]

I do not think inertia is important.

I'm pretty sure inertia is everything for driving the gu-h

Can you elaborate?

Aiui More mass spinning = higher inertia and makes it easier to drive the mgu-h without losing the inertia thus generating more power for longer.

Think fly-wheels, the bigger they are the more load they can handle without stalling.

That's my crude understanding anyway.

[PlatinumZealot]

The advantage to the no Mecedes setup is that the compressor and turbine are right beside each other and have their common shaft and angular velocity. The MGUH part has more freedom to do as they please; gearing assembly ect.

Mercedes now, could possibly have a motor that is designed for high angular velocities which are seen by the turbine and compressor. They may very well have advanced motor technology that permits such high speed operation.

The Mercedes MGUH could have a hollow shaft, through which the turbo's shaft assembly could fit. The MGUH could then be connected to the turbo's shaft using a planetary gear set. The advantage of the planetary gear set is that the gear loadings balance out, so as to not produce a load on the spinning shaft.

The MGUH could be mounted to a structure which holds intermediate supports for the shaft.

It is an engineering challenge with a hollow shaft spining over another.

Bearing assemblies at either end is a challenge. To fit a ball bearing onto a thin wall shaft is going to pose problems. The shaft is going to defelct too much to get the right fit. So you have to bulk up the parts where the bearings go onto. This adds weight.

Your have to have cooling oil supply for your planetary gearsets that are spining at 10000 rpms. And your bearings. More weight and complexity if these are not shared with the existing turbo oil supply i suspect.

There are some other challenges too. It is possible but i think it is too complicated to have the mguh geard when it is in the middle of the turbo and on the same axis as the rule stipulates.

I think it is a solid shaft with the motor designed for high rpms is easier. And more efficient.

I may be wrong but just my view of it.

[gruntguru]

There is a very rigid limit to downsizing of the turbomachines. The flow requirement will dictate the diameter of the compressor inducer (inlet) and the turbine exducer (outlet). You can't go any smaller than that.

Now consider that the compressor must accelerate the air to impart energy (dynamic pressure). True centrifugal compression does this as the air moves away from the compressor axis where the tangential speed is higher. Air entering near the outer edge of the inducer must be accelerated to match the high tangential speed of the impeller at that radius. This is more akin to axial compression than centrifugal. This air is accelerated further as it moves outwards to the tips of the impeller - a centrifugal process. Achieving high pressure ratios efficiently becomes increasingly difficult as the tip diameter approaches the inducer diameter. So for a given pressure ratio there will be an optimum tip diameter and reducing this value and spinning the turbo faster will involve some efficiency tradeoff.

[NL_Fer]

I can imagin that at 12000rpm at 2.5 bar, more as 4 times as much airflow is needed, than my 1.6 diesel, i drive daily. That turbo must be really huge and i can hardly image such a big turbo running faster than 125000 rpm.

Non of current 4 pu's are geared now p, are they?

[GoranF1]

ceribus_peribus-reddit user


It all starts with the "size 0" body that the designers wanted at the back end of the car. The extra-narrow pinch at the back would have certain aero advantages: better downforce directly over the rear axle, better rear grip, less drag, and so on.
Problem: this means less space for the engine. In particular, there wasn't a good spot to put the compressor for the turbo.
Solution: they'll fit the compressor right into the "V" of the engine, instead of ahead or behind.
Another problem: that's a pretty small space to begin with. A normal compressor wouldn't fit there. You could squeeze in a small one, though.
Another solution: since it'll be smaller, the compressor will have to cycle faster to deliver the same power. About 130,000rpm instead of the more typical 110-120krpm. Most other teams aren't sure they could even make a (normal sized) compressor cycle faster than 120k without blowing up -- if they could get more power they'd already be running them faster. But Honda are confident that they can make a smaller one go that fast.
Season starts. To everyone's surprise, the small compressor does go that fast, without blowing up.
But, it has extreme, violent vibrations at max speed. Connections get shaken loose. When they are attached more securely, they get snapped and broken from the thrashing. Honda redesigns the connections.
Next problem is the compressor gets too hot. Unexpectedly hot. It's frying the nearby connections and electronics with heat. Honda tries to fix this, add cooling or insulate the rest somehow.
They don't find a way to completely fix the thermal issues (it's not exactly easy trying to route cooling airflow through the middle of your engine), and ultimately the only solution is to turn down the compressor so it doesn't get as hot.
Problem: with the compressor turned down so far, the ICE is losing a lot of power. Like 60-80bhp at least. At the lower power level, there's less heat created exhaust power available for the MGU-H to collect, meaning a further loss of power -- the hybrid batteries don't get fully charged, and so on. The losses just cascade and multiply through the entire power plant.
Honda spends the rest of the season making incremental efficiency improvements to the ICE (making the most of what they get from the turned-down compressor) and riding the edge of how high they can run the compressor before ruining the works. They squeeze out many tens of additional BHP but it's nowhere near enough to make up for the shortfall.
Most sessions cost them an engine but they do manage to get a couple of them through entire races, always at the back of the midfield because of the compromised power levels.

[Andres125sx]

What credibility does that have?

Because that´s different to the mgu-h problem commonly commented, it actually is a ICE problem wich also affect mgu-h. Does make sense and look plausible tough

[Mr. Fahrenheit]

Thanks for sharing that GoranF1 and thank you ceribus_peribus for explaining the issues so succinctly.

Perhaps romanticising a little about Honda's potential but if, IF, IF they tame that compressor, it sounds like they suddenly have a very potent power unit.

[proteus]

How is it looking for brazilian GP? Will they have deployment problems again, or can we expect pace similar to Austin?

[GoranF1]

In theory what would(besides aero) be the advantage of small TC and more RPMs whit it?