2025/2026 Hybrid Powerunit speculation

[gruntguru]

I get your point. It is really only the ICE that converts fuel energy to mechanical - and that includes the turbo-compounding. Current efficiency of 50%+ includes a very minor loss converting turbine energy to electrical.

Does anyone have any thoughts on how the ICE might change and what efficiency might be possible, now that the only source of work is the crankshaft?

My initial thoughts:
1. Later EVO to increase in-cylinder expansion and minimise EGT.
2. Reduce or eliminate EIVC (Miller) to maximise pumping work done on piston during intake stroke. (This will be the sole means of returning recovered exhaust energy to the PU)
3. Larger turbine and housing to reduce backpressure. Minimise wastegating.

The way to totally avoid wastegating is achieving a steady state as in some old diesel engines, regulated by fuel flow (a.k.a. very lean burning).

Yes. Unfortunately unlikely, since the rules limit boost. The only way to run high boost over a wide range of RPM while ensuring boost doesn't creep above the rules limit, is a wastegate.

[wuzak]

There have been some reports that the best ICE will have about 420kW, which is 50.4% TE.

A while back, Tombazis suggested that the PU could have as much as 1,100hp peak power, which would mean ICE power of 470kW and TE of 56%!

Wow!

Just to clarify, Tombazis said that the 2026 PU could have up to 100hp more than the current PU.

Most think that the current PU has a peak power of ~1,000hp, which is why I said 1,100hp.

The current PUs may not actually have 1,000hp though.

Mark Hughes had an article where he suggested that the best would be about 450kW. That would be 54% TE.

But 400 - 420kW sounds like a more reasonable guess to me.

The current PUs have been frozen since 2022, so no development has gone into them.

That may be why it is possible for the 2026 ICE to be more efficient.

[JRodrigues]

Some news from Audi's camp regarding their engine:

[vorticism]

Great to hear from the PU-Chief:

5 seconds to get the turbine up to speed... ultimately it is a small displacement engine attached to a very large TC. I guess the long spooling time means they can’t use the K to speed up the engine while it’s declutched on the grid. He also implied no, or at least limited, anti-lag. The regs specify 1 spark per cycle or stroke, depending on definitions, but either way this limits how much late ignition there can be. He implied the timing of the fuel they can inject per stroke/cycle is prescribed, but I wasn’t able to find that wording.

Regarding the “creative” ways to keep the turbo spinning during reduced power demand that he alluded to, the main one that comes to mind for me (I think someone already suggested it earlier) is increasing the engine speed while the throttles are open in order to cold blow the turbine. Use higher RPM during low or zero PU torque demand. The reciprocating components should be lighter than they currently are, owing to the reduced power output. Fictional numbers but imagine f.e 11k RPM for peak ICE power output but still have another +4k RPM usable for engine braking and low-demand scenarios. Highest engine notes achieved during braking and mid-corner. High RPM spikes during braking.

Would there not also be scope for using the BOVs to help maintain turbine shaft speed by unloading the compressor?

[saviour stivala]

It is not correct to say that the 2026 formula 1 turbo will take five seconds to get up-to-speed, that is a misunderstanding, as the much more powerful electric motor is expected to provide enough torque at low speed to offset this issue. The 2026 MGU-K will not be declutchable, it will be permanently integrated unlike the current set-up.

[wuzak]

The turbo lag will mostly happen after slow corners.

The MGUK will be deployed to overcome this.

Also, the ICE will probably be run at higher power in certain corners, with the MGUK recovering to bring the output back to the required amount.

That way the turbo will be kept up to speed.

They will probably also use lower gears in corners, where they can, to keep the rpm up and maximise energy recovery.

Back in 2013/2014, the Renault engine boss at the time said that the turbo lag without the MGUH was ~10s.

[Tommy Cookers]

... The 2026 MGU-K will not be declutchable, it will be permanently integrated unlike the current set-up.

what ??
how isn't the 2026 MGU-K connected to the ICE and to the planet Earth in exactly the same way as the 2025 MGU-K ?

[diffuser]

... The 2026 MGU-K will not be declutchable, it will be permanently integrated unlike the current set-up.

what ??
how isn't the 2026 MGU-K connected to the ICE and to the planet Earth in exactly the same way as the 2025 MGU-K ?

The way I understand it, is that they moved it from the rear of the ICE to the front. Aside from that it hasn't changed. There is still a transmission between it and the crankshaft.

I don't understand why you'd need a clutch since you can probably do everything you'd need to do with the directions of the current through the coils/magnets of the MGU-K ?

[karana]

...The 2026 MGU-K will not be declutchable, it will be permanently integrated unlike the current set-up.

2026:

C5.18.2 Under normal operating conditions all MGU-K rotating parts must be permanently mechanically
linked to the ICE with a fixed speed ratio to the crankshaft. The MGU-K and its drive axis must be
parallel to the crankshaft axis.

C5.18.3 All mechanical power to and from the MGU-K must pass through a single shaft to the MGU-K
transmission. The connection to the ICE crankshaft must be ahead of XPU=100.

C5.18.4 An in-line, passive, dissipative energy torque limitation device may be incorporated in this link
which temporarily allows the speed ratio to change for the sole purpose of protecting the
components from dynamic torque overshoots. This device may only act above 520Nm when
referred to crankshaft speed.

2025:

5.3.3 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. An
in-line, passive, dissipative energy torque limitation device may be incorporated in this link
which temporarily allows the speed ratio to change for the sole purpose of protecting the
components from dynamic torque overshoots. This device may only act above 220Nm when
referred to crankshaft speed.

Basically no difference.

[Tommy Cookers]

2026:

C5.18.4 An in-line, passive, dissipative energy torque limitation device may be incorporated in this link
which temporarily allows the speed ratio to change for the sole purpose of protecting the
components from dynamic torque overshoots. This device may only act above 520Nm when
referred to crankshaft speed.

2025:

5.3.3 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. An
in-line, passive, dissipative energy torque limitation device may be incorporated in this link
which temporarily allows the speed ratio to change for the sole purpose of protecting the
components from dynamic torque overshoots. This device may only act above 220Nm when
referred to crankshaft speed.

Basically no difference.

the 2014/2025 ? rules permit MGU-K torque only up to 200 Nm (crankshaft equivalent) ....
ie below 5775 rpm 120 kW MGU-K power cannot be maintained (5775 is 50% of (10500 +10%))

the 2026 rules apparently permit MGU-K torque only up to 500 Nm equivalent ....
ie below 5775 rpm 300 kW cannot be maintained
ie below 6737 rpm 350 kW cannot be maintained

[vorticism]

Regarding some preceding comments: Notice I wrote “the engine while it is declutched on the grid.” Ideally the clutch between a running engine & its transmission is open when the car is stationary, in lieu of, say, having a 1000000:1 gear ratio for first gear. What I was getting at was, if it was permitted to use the 500 hp MGUK to increase the crankshaft speed of the 500 hp ICE beyond what the ICE itself can provide, then this should help spool up the turbine more quickly. But as the video alluded to, it takes 5 seconds to spool--does this mean the K can’t be used during idle? There are limits, discussed earlier in this topic, about what the MGUK can do independent of pedal position. It's too bad they can't use two accelerator pedals! How much fun would that be. Fiddle-throttle has a nice ring to it.

As for relying on the MGUK & ES alone to compensate for lag, you’d be leaving performance on the table if there is a way to overcome it by another means.

[Farnborough]

Regarding some preceding comments: Notice I wrote “the engine while it is declutched on the grid.” Ideally the clutch between a running engine & its transmission is open when the car is stationary, in lieu of, say, having a 1000000:1 gear ratio for first gear. What I was getting at was, if it was permitted to use the 500 hp MGUK to increase the crankshaft speed of the 500 hp ICE beyond what the ICE itself can provide, then this should help spool up the turbine more quickly. But as the video alluded to, it takes 5 seconds to spool--does this mean the K can’t be used during idle? There are limits, discussed earlier in this topic, about what the MGUK can do independent of pedal position. It's too bad they can't use two accelerator pedals! How much fun would that be. Fiddle-throttle has a nice ring to it.

As for relying on the MGUK & ES alone to compensate for lag, you’d be leaving performance on the table if there is a way to overcome it by another means.

I'd see it more logically the other way. Available energy to spin exhaust turbine is greatest with the ice at absolute full load, not being spun unloaded by a E contribution to crankshaft.
If the ICE were to be run against resistance with throttle wide open (against a "brake") as in a dyno facility, then it should spool the turbo into peak.
Could be start procedure, arrive at grid slot with 50% battery e.g. select first, clutch open, max regeneration selected, rev nuts off it at max regeneration for 5secs prior to lights out, away at full boost/battery IF strategy and load correctly applied.

Note to the pedantic, this is thinking out loud, and not after reading nuance of deployment regs

[Farnborough]

Maybe room for a "dual action" clutch operating lever ? Under sole control of driver.

Pulled in to max, clutch open, regeneration (crank brake) maximum .... as the lever is brought out to close clutch and move car .... the regeneration function is "tapered" down to minimum .... max torque, and boost (according to throttle pedal %) is transferred from regeneration to transmission drive to wheels.

[saviour stivala]

... The 2026 MGU-K will not be declutchable, it will be permanently integrated unlike the current set-up.

what ??
how isn't the 2026 MGU-K connected to the ICE and to the planet Earth in exactly the same way as the 2025 MGU-K ?

In 2026 the 350kw MGU-K will no longer be mounted under the exhaust system under the right hand bank, as in the current (2025) engine. Instead, it will be integrated into the chassis (safety cell) for safety reasons, as part of the ERS, next to the battery under the fuel cell.

[saviour stivala]

Regarding some preceding comments: Notice I wrote “the engine while it is declutched on the grid.” Ideally the clutch between a running engine & its transmission is open when the car is stationary, in lieu of, say, having a 1000000:1 gear ratio for first gear. What I was getting at was, if it was permitted to use the 500 hp MGUK to increase the crankshaft speed of the 500 hp ICE beyond what the ICE itself can provide, then this should help spool up the turbine more quickly. But as the video alluded to, it takes 5 seconds to spool--does this mean the K can’t be used during idle? There are limits, discussed earlier in this topic, about what the MGUK can do independent of pedal position. It's too bad they can't use two accelerator pedals! How much fun would that be. Fiddle-throttle has a nice ring to it.

As for relying on the MGUK & ES alone to compensate for lag, you’d be leaving performance on the table if there is a way to overcome it by another means.

I'd see it more logically the other way. Available energy to spin exhaust turbine is greatest with the ice at absolute full load, not being spun unloaded by a E contribution to crankshaft.
If the ICE were to be run against resistance with throttle wide open (against a "brake") as in a dyno facility, then it should spool the turbo into peak.
Could be start procedure, arrive at grid slot with 50% battery e.g. select first, clutch open, max regeneration selected, rev nuts off it at max regeneration for 5secs prior to lights out, away at full boost/battery IF strategy and load correctly applied.

Note to the pedantic, this is thinking out loud, and not after reading nuance of deployment regs

The MGU-K being part of the power unit's functionality for acceleration, it can be used in deploy mode on race starts.