Honda Power Unit Hardware & Software

[Craigy]

Where did he say that? Give us the quote and source.

It doesn't make sense to "thump your chest" about something when you are actually struggling with it. He's a real paddock source saying the exact opposite of what you are saying That doesn't eliminate the possibility that someone is struggling with their fuel, but not everyone.

Maybe problems is not the right word but technically challenges are the are his real words "They are more complex, because they are composed of elements with different evaporation points.”
“With sustainable fuels, some components evaporate later than others, and this makes combustion more complex. We are working with hotter combustion chambers, a condition that opens up.

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

[Badger]

Maybe problems is not the right word but technically challenges are the are his real words "They are more complex, because they are composed of elements with different evaporation points.”
“With sustainable fuels, some components evaporate later than others, and this makes combustion more complex. We are working with hotter combustion chambers, a condition that opens up.

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

There was never a "hot limit" as far as I understand, the test was always done at ambient. However there are several regulatory constraints on materials and such that make it hard to greatly expand your compression ratio much above the ambient measurement.

[Craigy]

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

There was never a "hot limit" as far as I understand, the test was always done at ambient. However there are several regulatory constraints on materials and such that make it hard to greatly expand your compression ratio much above the ambient measurement.

I understand why you say this - it's not in the clear, direct wording of the old reg, but in the effect of it.
Previous reg -- https://www.fia.com/sites/default/files ... -04-07.pdf

5.4.6 No cylinder of the engine may have a geometric compression ratio higher than 18.0.

Note that this never specified how it was measured, which is why it was taken to be universally applied, even when hot. Presumably the FIA just looked at stuff along the lines of in-cylinder pressure transducers from team telemetry.
The new regs aren't worded like that at all.

2026 wording (so different that it's essentially a different regulation)

C5.4.3 No cylinder of the engine may have a geometric compression ratio higher than 16.0. The procedure
to measure this value will be detailed by each PU Manufacturer according to the Guidance
Document FIA-F1-DOC-C042 and executed at ambient temperature. This procedure must be
approved by the FIA Technical Department and included in the PU Manufacturer homologation
dossier.

https://www.fia.com/system/files/docume ... 2-10_0.pdf

Hm. Perhaps this belongs in a different thread rather than the Honda dedicated one.

[Badger]

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

There was never a "hot limit" as far as I understand, the test was always done at ambient. However there are several regulatory constraints on materials and such that make it hard to greatly expand your compression ratio much above the ambient measurement.

I understand why you say this - it's not in the clear, direct wording of the old reg, but in the effect of it.
Previous reg -- https://www.fia.com/sites/default/files ... -04-07.pdf

5.4.6 No cylinder of the engine may have a geometric compression ratio higher than 18.0.

Note that this never specified how it was measured, which is why it was taken to be universally applied, even when hot. Presumably the FIA just looked at stuff along the lines of in-cylinder pressure transducers from team telemetry.
The new regs aren't worded like that at all.

2026 wording (so different that it's essentially a different regulation)

C5.4.3 No cylinder of the engine may have a geometric compression ratio higher than 16.0. The procedure
to measure this value will be detailed by each PU Manufacturer according to the Guidance
Document FIA-F1-DOC-C042 and executed at ambient temperature. This procedure must be
approved by the FIA Technical Department and included in the PU Manufacturer homologation
dossier.

https://www.fia.com/system/files/docume ... 2-10_0.pdf

Hm. Perhaps this belongs in a different thread rather than the Honda dedicated one.

The test hasn't changed, it was always done at ambient. The FIA only clarified this in the written rules recently though, presumably because some manufacturers were starting to ask questions about this "trick". But the FIA have never had a way of measuring hot compression. The way I understand it is that it was never an issue when the rule said 18:1 because that was so close to the optimum CR that there was little to no gain to be had by trying to push it much more. When they lowered it to 16:1 it suddenly became an area for improvement.

[gruntguru]

. . . Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

Not necessarily. NOx reduces as AFR goes leaner - and these engines are very lean.

[gruntguru]

. . . . I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune. . . . .

Volumes associated with the injectors and prechambers are far too small to make a meaningful difference.

[diffuser]

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

There was never a "hot limit" as far as I understand, the test was always done at ambient. However there are several regulatory constraints on materials and such that make it hard to greatly expand your compression ratio much above the ambient measurement.

Between 2014 and 2025 knock was the limit. The development of fastburn allowed the flame to out pace the knock and raise compression to unbelievably heights.

[PlatinumZealot]

Maybe problems is not the right word but technically challenges are the are his real words "They are more complex, because they are composed of elements with different evaporation points.”
“With sustainable fuels, some components evaporate later than others, and this makes combustion more complex. We are working with hotter combustion chambers, a condition that opens up.

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

Wait wait.. I think you along with some people are seeeriously underestimated the engineering time involved to do that sort of complexity (just to cheat?!) imagine the sensors you will need to verify that your cheating magic material is workimg evenly accross all cylinders! Lol

Also consider this... What would have stopped the teams from doing this in previous regulations? Nothing. They would have even more compression ratio to gain in the previous regs because the ratios get bigger as space diminishes.

Heat engines are... Hot.
The temperature of these metals in a running engine is very hot.. Lets say 120 degress Celcius for arguments sake.

This is happening to ALL engines. The rules practically state measurements at room temperature. Lets say all thermal expensions of standard materials in standard engines accross manufactuers are equal at operating temps. And so is any geometric compression ratio change. Let's say that compression ratio gain is positive. It is 0.1 points in a non cheating engine (too generous perhaps!) now let's compare a cheating engine.

The piston crown is say 5mm thick. The squished volume is 1.6 liters divided by 6 divided by 16 = 16.667 cubic centimeters at 16 to 1 compression ratio at room temp.

To get an 17 to 1 compression ratio the squished volume needs to be (assume the cylinder volume as the engine gets hotter which it will with crankshaft thermal, and piston bore thermal expansion increases by 0.5%. And displacement is now 1,608 cc) = 1608/6 cylinder/ 17 = 15.767 cc.

You need a reduction of 0.9cc to get 17 to 1 compression ratio.. Or using F1 piston bore of 80mm you get a change in height of 0.018cm or 0.18 millimeters.

In other words.. If the crown is made of some fancy expanding material the thermal expansion needs to be 0.18mm/5mm = 0.036 or 36,000 micro strains. OVER room temp.

Aluminum is 24 micro-strains per degrees celcius (while solid and same crystal structure). So you need 36000/ 24 = a whopping 15,000 degrees celsius!!

Ok we need a cheating material with somthing 15,000 / 100 degrees C of higher thermal expansion. That would bd 150 times aluminum or 150* 24 microstrain per degrees celcius. = 3600.

So guys.. Perspective!! Imagine the engineering resource to do this cheating outside of going to moving parts.

[Hoffman900]

The only way they can practically enforce this is geometric compression ratio is calculated when there is 0mm of piston to head clearance. The gap can never be any closer for obvious reasons. So in reality, all engines will have to undershoot this at ambient.

[dialtone]

Not much to gain much past 18:1 CRs, the additional frictional forces reduce engine life and wipe out the extra power gain. This wasn’t happening before because there was no big advantage.

It’s happening now because the rule was written badly before (they just changed the 18 to 16 for the new rules and then added the temperature measurement later), some teams were looking for marginal gains and going back to 18 was a low hanging fruit and the rule was conveniently written if FIA was corrupt enough, and I mean… they knew the answer to that last question.

Some teams thought the challenge was to get as close as possible to 16:1 in working conditions, others thought start at 16 and see if you get to 18. Amateur rule making.

[Hoffman900]

. . . . I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune. . . . .

Volumes associated with the injectors and prechambers are far too small to make a meaningful difference.

This. There is a masters thesis written in conjunction with Ilmor Brixworth that gets into the design of them. This just isn’t a practicle theory at all, let alone any change in geometry would wreak havoc on the entire injection process.

[diffuser]

In other words he didn't say what you attributed to him. But he very clearly said that they could do higher compression, and that the 16:1 CR was "too low".

On the topic of compression ratios for the 2026 formula.

The measured value at room temperature of 16:1 or just under that value, for legality. The Historic limit of 18:1 at any time. If I read this correctly, there's not really a hot running limit at all in 2026. You could engineer for 20:1 or more in a hot running config if you wanted to.

People keep talking about this as if the manufacturers have achieved the hot running/room temperature differences using only the bottom end (crank/rods/pistons) - presumably some form of elastic behaviour based on temperature. I think it'd make a lot of sense for it to also be in the injector, where you could (for example) control temperature of materials at will more easily.

A system to up the compression ratio by (for example) disabling a cooling loop that causes a material to expand and lower the volume in the injectors would be pretty straightforward to implement and relatively easy to tune.

With 70kg per race and 3000MJ/h flow, 2026 is a fuel efficiency formula for the ICE. Nobody's measuring the horrific NOx emissions practically guaranteed from high compression ratios, so the development direction is pretty clear in that regard.

If nothing else, ignore the discussed 18:1 limit. It was in the previous formula and doesn't exist any more. The sky is the limit now.

Wait wait.. I think you along with some people are seeeriously underestimated the engineering time involved to do that sort of complexity (just to cheat?!) imagine the sensors you will need to verify that your cheating magic material is workimg evenly accross all cylinders! Lol

Also consider this... What would have stopped the teams from doing this in previous regulations? Nothing. They would have even more compression ratio to gain in the previous regs because the ratios get bigger as space diminishes.

Heat engines are... Hot.
The temperature of these metals in a running engine is very hot.. Lets say 120 degress Celcius for arguments sake.

This is happening to ALL engines. The rules practically state measurements at room temperature. Lets say all thermal expensions of standard materials in standard engines accross manufactuers are equal at operating temps. And so is any geometric compression ratio change. Let's say that compression ratio gain is positive. It is 0.1 points in a non cheating engine (too generous perhaps!) now let's compare a cheating engine.

The piston crown is say 5mm thick. The squished volume is 1.6 liters divided by 6 divided by 16 = 16.667 cubic centimeters at 16 to 1 compression ratio at room temp.

To get an 17 to 1 compression ratio the squished volume needs to be (assume the cylinder volume as the engine gets hotter which it will with crankshaft thermal, and piston bore thermal expansion increases by 0.5%. And displacement is now 1,608 cc) = 1608/6 cylinder/ 17 = 15.767 cc.

You need a reduction of 0.9cc to get 17 to 1 compression ratio.. Or using F1 piston bore of 80mm you get a change in height of 0.018cm or 0.18 millimeters.

In other words.. If the crown is made of some fancy expanding material the thermal expansion needs to be 0.18mm/5mm = 0.036 or 36,000 micro strains. OVER room temp.

Aluminum is 24 micro-strains per degrees celcius (while solid and same crystal structure). So you need 36000/ 24 = a whopping 15,000 degrees celsius!!

Ok we need a cheating material with somthing 15,000 / 100 degrees C of higher thermal expansion. That would bd 150 times aluminum or 150* 24 microstrain per degrees celcius. = 3600.

So guys.. Perspective!! Imagine the engineering resource to do this cheating outside of going to moving parts.

Connecting rods are manufactured from tightly controlled titanium alloys or a small number of high-strength steel alloys. Combined with the significantly lower thermal exposure of the bottom end relative to the combustion chamber, thermal expansion of the rod is negligible and has no meaningful impact on effective rod length. Additionally, peak rod stretch occurs under inertial tensile loading after TDC (during overlap), rather than at firing TDC, and therefore does not contribute to an increase in effective compression ratio.

Increased effective compression ratio comes primarily from piston crown growth and cylinder head expansion under high thermal load. Even there, material choices are tightly restricted, and the cylinder head must be machined from a single billet of the specified alloy.

My assumption is that the most viable way to increase compression ratio beyond ~16.7:1—typically reached at normal operating temperature from a nominal 16.0:1 at ambient—would be through combustion chamber geometry. By locally thinning regions of the aluminum combustion chamber, those areas would experience higher thermal expansion, provided structural integrity is maintained. In principle, this could allow effective compression ratios beyond 16.7:1. How feasible that is in practice is unclear.

[Hoffman900]

If you change combustion chamber geometry, you’ve now deformed and compromised the valve seats / effect the valve height, it has to work around the TJI unit and in-situ pressure transducer, all of that still has to support 250bar + of cylinder pressure, AND it can’t compromise cylinder sealing.

To even get 16:1 compression requires a very compact chamber to begin with. There just isn’t any room for this and imo is just not a viable theory.

[diffuser]

Sorry, I don't remember who mentioned that it would be hard to measure real time CR.....There was a you tube video with Pat Symonds from 2025 that said that the all F1 ICE used a Pressure sensor to detect knock. That those sensors were very expensive and they wanted to do away with them.

Meaning if they wanted to measure CR real time, they could very easily.

[Hoffman900]

Sorry, I don't remember who mentioned that it would be hard to measure real time CR.....There was a you tube video with Pat Symonds from 2025 that said that the all F1 ICE used a Pressure sensor to detect knock. That those sensors were very expensive and they wanted to do away with them.

Meaning if they wanted to measure CR real time, they could very easily.

Not entirely, because those are measuring firing engines. Peak combustion pressure is partly a function of geometric compression ratio, but also other things, especially in a Miller Cycle engine. In an Otto Cycle engine you can compare it against an idealized P-V loop, but even then, it depends on other things as well.

We even use these sensors in amatuer racing now (on the dyno). You’re not competitive anymore if you’re not developing with them, and yes they are very expensive, and F1 teams go through them like candy running them in the cars, on every cylinder. They burn up, go out of calibration, etc.

Most sanctioning bodies use these to measure compression ratio (even NASCAR and IMSA): https://katechengines.com/i-30497781-ka ... ester.html . If you fail, then you get torn down and they actually measure them, but again, measuring parts is done at ambient, not hot.

As I said, the only way to fairly do this is assume geometric compression ratio is when the piston to head clearance is zero. Everything else is variable to some degree