2026 Drama: Alleged engine loophole

[AR3-GP]

What would stop Mercedes from rushing all of the allocation into use before August like they did in 2017 with the oil burning TD?

All of the allocation into use?

Yes the PU allocation. They can put 4 PUs in the pool before August. Then what? Will the FIA make Mercedes teams take grid penalties to replace the PUs that no longer pass the test?

[chrisc90]

Well that pu would be in breach of regulation so effectively pointless.

No different to a rear wing for example.

[AR3-GP]

Well that pu would be in breach of regulation so effectively pointless.

No different to a rear wing for example.

We've seen how it was handled before. In 2017, any PUs introduced to the pool before the reduction of the oil burning limit was grandfathered. Mercedes and Ferrari agreed not to rush new PUs into the pool before Spa, and then Mercedes did it anyway.

[dialtone]

Well that pu would be in breach of regulation so effectively pointless.

No different to a rear wing for example.

We've seen how it was handled before. In 2017, any PUs introduced to the pool before the reduction of the oil burning limit was grandfathered. Mercedes and Ferrari agreed not to rush new PUs into the pool before Spa, and then Mercedes did it anyway.

Yeah I get your concerns, and I also understand that Mercedes has cheated, but have you stopped for a second to consider that if FIA chooses to act, then Mercedes would be punished? What is a team supposed to do, respect the rules? No chance they'll be punished, then FIA would need to justify punishing a cheating team, nobody is going to understand that.

[Rodak]

130°C sounds like the maximum temprature that the water cooling systems can handle. There was a max pressure and water would boil abobe 130°C IIRC.

130° sounds like a number engineered for Merc to pass the test lol

Yeah. The maximum coolant pressure is 3.75 bar. At that pressure water boils @ 141°C; I don't know what coolant they are using though, although I guess it doesn't matter.

[PlatinumZealot]

130°C sounds like the maximum temprature that the water cooling systems can handle. There was a max pressure and water would boil abobe 130°C IIRC.

130° sounds like a number engineered for Merc to pass the test lol

That sounds about right for an engine even on the hot side.
I think they can pre-heat the block seembly and head assembly in an oven or an induction heater or whatever and conduct the measurements.

[LM10]

Maybe because FIA initially declared the engine LEGAL? Meaning they are partially to blame?

Oh, really?

So Toto went to the FIA - holding a cup of Espresso in his hands - and said "Hey... *sip* so as you know our track record without a significant PU advantage is not THAT good... *sip* what if I told you that we've thought of something extremely clever and built our engine specifically in a way to have a CR of 16:1 at ambient temperature, but shooting up to 18:1 as soon as it gets hot like my Espresso? *sip* I know it was not your intention with this rule, but will you be OK with that?" *sip*

And the FIA said YES? This makes sense!

LOL,

https://uz.kursiv.media/en/2025-12-25/f ... 026-rules/

FIA back in December gave the all clear, after in October adding the wording in the rules ''at ambient temp''
Mercedes said it had communicated with FIA all along through this development so yes, they were aware,

Grow up man.

Sure, buddy, sure. Mercedes definitely provided the FIA with all the necessary details instead of gambling.

[venkyhere]

In addition, the strict prohibition of variable geometry systems under Article C5.7 applies here. If an engine increases its compression ratio from 16:1 to 18:1 solely through material expansion, thermal physics effectively acts as a dynamic actuator. Such a change in combustion characteristics of more than 12% is no longer a negligible tolerance, but a functional geometry change that aims to circumvent the static limits of the homologation dossier.

Sorry for being pedantic, the 16:1 changing to 18:1 isn't a 12% change in dimensions of something.
(16+x)/(1-x) = 18.
Solving, we get x=(2/19)=0.105
which works out to (0.105/16)*100 = 0.656%

If we are talking about conrod+piston that is in contact with something upwards of 2000+ celcius, I think 0.6% is not out of the ballpark of 'tolerance'.
Of course the regulations would have restricted the usage of metals for the rod/piston/gudgeonpin/bearing inserts etc etc to disallow as much expansion as possible ; and we have theories ranging from 'exotic 3D printing' to 'some speciall milling technique' to enable unequal expansions to the full-hollywood 'secret chamber' going around to explain the 16:1 rising to 18:1. That said, the reason I posted is to highlight 'how little' of change in dimension is needed to increase the compression ratio , it's not a giant number like 12%, more like 1/20th of that.

While that’s an interesting take, it unfortunately relies on a fundamental misunderstanding of how compression ratio (CR) is defined and how engine tolerances actually work.
​1. The Mathematical Error:
Your equation \frac{16+x}{1-x} = 18 doesn't actually represent the physical volume change in a cylinder. The compression ratio is defined as CR = \frac{V_s + V_c}{V_c} (where V_s is swept volume and V_c is clearance volume). To move from a 16:1 ratio to 18:1, the clearance volume (V_c) would need to shrink from 1 unit to roughly 0.882 units. That is a 11.8% reduction in combustion chamber volume—not 0.6\%. Your math significantly underestimates the physical change required.

Agree, my bad, sorry. I typed the reply hurriedly, and made the mistake of using '16+x' instead of using '16-x' in the numerator. Even then, it would have been x=2/17=0.118 (lines up with your calculation).
If we assume that the bore is unchanged, and if this change were to come from rod-length alone (I mentioned in another reply that this was a back of the envelope crude calculation anyway), it has to come from a length that "represented" 15x1.414 (minimum rod length, to be pessimistic) into representing 15x1.414 + 2/17.
That would mean an increase in length of 0.55%, whatever be the length of the conrod.

When the clearance volume reduces by 11.8%, it doesn't mean that some physical object is changing by 11.8%. That 'small space' is created by 'large objects', so a small % change in a large object is enough to effect a large % change in a small empty space.

​2. Mechanical Reality vs. 'Tolerance':
Even if we look at your 0.6\% figure, in high-end motorsport like F1, that is an astronomical distance. On a 100 mm connecting rod, 0.6\% is 0.6 mm. Engine builders obsess over 'Piston-to-Head Clearance' (squish) measured in fractions of a millimeter. An unplanned expansion of 0.6 mm would cause the piston to mechanically strike the cylinder head at high RPM, resulting in immediate engine failure. It’s not 'within the ballpark of tolerance'; it’s a catastrophic collision.

Typical engines experience conrod length expansion by 0.1 or 0.2% at operating temperature, and yes of course engine builders obsess and account for such a change when calculating clearance. The point I intended to make, was that an F1 engine made with exotic materials and clever manufacturing, could well design their rod expansion with a recipe to expand more than usual (from 0.1, to say 0.55, percentage), and that these % are decimal value %s and not a big number like 12%. Calculating the thermal expansion change % in terms of the clearance volume will yield a big number, whereas the actual expansion needs to be only in terms of decimal number %, to cause the 'empty space' to reduce by a large % number - that was the only point I was trying to make (in the eagerness of which, made the math error in the last post).

​

[Waz]

14 cars instead of 22 isn't good reasoning. As Vasseur said, it's a lot easier to reduce compression than increase it, and the Brabham fan car is only the most famous precedent for giving no time to replace a car that was deemed illegal after one race.

If you make mockery of the rules like this as a governing body you will soon have a circus going on.

That reasoning is literally scaremongering. Mercedes will definitely have an engine spec they can revert to that is fully compliant from race 1. It will just be several months behind in development.

And they could surely build enough for Melbourne. A lot of components would be the same and are already manufactured.

[nico5]

130°C sounds like the maximum temprature that the water cooling systems can handle. There was a max pressure and water would boil abobe 130°C IIRC.

130° sounds like a number engineered for Merc to pass the test lol

That sounds about right for an engine even on the hot side.
I think they can pre-heat the block seembly and head assembly in an oven or an induction heater or whatever and conduct the measurements.

As hollus mentioned, that's right for the coolant. Rod beam temperatures are likely more towards 200 than 130.

[Juzh]

The fact that these questions did not arise in 2025 and prior when the limit was 18:1 may suggest that there's nothing to the rumors nor the FIA directives/rule changes. Was it not advantageous to exceed 18:1 in 2025? What did the test entail last year? Was it not tested?

law of diminishing returns :
how much more can be increased beyond 18:1 until knocking starts to happen ? 19 ? I don't think so. Probably some 18.5 (max). That's nothing like going from 16 to 18, which can be easily accommodated knock-free, since they have already done 18 previously.

Tombazis said previous 18:1 wasn't a real limit because no could really hit that number, that's why it wasn't a problem. so anything even higher than 18:1 is in my opinion completely out of the question. Probably even 18:1 is pushing it.

[PlatinumZealot]

130° sounds like a number engineered for Merc to pass the test lol

That sounds about right for an engine even on the hot side.
I think they can pre-heat the block seembly and head assembly in an oven or an induction heater or whatever and conduct the measurements.

As hollus mentioned, that's right for the coolant. Rod beam temperatures are likely more towards 200 than 130.

I'm sorry but he should be corrected.
There is F1 telemetry screens showing F1 coolant temperatures and oil temperatures.

The bulk temperature of the block is not to be confused with it's surface temperature.

[AR3-GP]

Tombazis said previous 18:1 wasn't a real limit because no could really hit that number, that's why it wasn't a problem. so anything even higher than 18:1 is in my opinion completely out of the question. Probably even 18:1 is pushing it.

Pushing it? Honda hit 18:1 in 2021. Hodgkinson said they had the combustion speed for 18:1

[nico5]

That sounds about right for an engine even on the hot side.
I think they can pre-heat the block seembly and head assembly in an oven or an induction heater or whatever and conduct the measurements.

As hollus mentioned, that's right for the coolant. Rod beam temperatures are likely more towards 200 than 130.

I'm sorry but he should be corrected.
There is F1 telemetry screens showing F1 coolant temperatures and oil temperatures.

The bulk temperature of the block is not to be confused with it's surface temperature.

Be that as it may, can we agree that what is cooled should be hotter than what is cooling by very simple thermodynamical principles?

[Juzh]

Tombazis said previous 18:1 wasn't a real limit because no could really hit that number, that's why it wasn't a problem. so anything even higher than 18:1 is in my opinion completely out of the question. Probably even 18:1 is pushing it.

Pushing it? Honda hit 18:1 in 2021. Hodgkinson said they had the combustion speed for 18:1

I'm just going by what tombazis said a week ago in this interview.