33 replies to this topic

[ZoRG]

Just wondering with F1's budget why they don't use Silicone Nitride valves instead of titanium valves?

[desmo]

It's silicon, not silicone.

[alexbiker]

What is the advantage you perceive to silicon nitride?

Alex.

[perfectelise]

Lower weight according to this article:

http://www.ndt.net/news/fhg_cera.htm

[alexbiker]

F1 Technical Regs:

No parts of the car may be made from metallic materials with a modulus of elasticity greater than 40 GPa

Modulus of elasticity of Silicon nitride: 310 GPa.

Bloody FIA.

Alex

[ZoRG]

It's silicon, not silicone.

Potato - Potaato good thing you knew what I meant!!

Ahh ok. Its a pitty since there is some real world application for this.

[Matt Davis]

F1 Technical Regs:

No parts of the car may be made from metallic materials with a modulus of elasticity greater than 40 GPa

Firstly is Silicon Nitride metallic?

I am studying electronics so I am not an expert on mechanical stuff, but according to data here the modulus of elasticity for Silicon Nitride is 85 .. 105 GPa. This is significantly less than the figure that was quoted by Alex, but still to high to be used if it is metallic, (possibly due to it being a thin film MEMS application).

From the same source here the modulus of elasticity for titanium is 102.6 .. 110 GPa and therefore couldn't be used in a F1 car, but it is. Can anyone explain?

Finally why is single crystal silicon (the same stuff that computer chips are made from) not used for components like valves as it is stronger than steel, harder than iron, lighter than aluminium and is elastic until fracture so doesn't permanently deform.

Matt

[Greg Locock]

Um, Alex, I think you missed something very important out. 40 GPa per gram/cc I think. Otherwise cast iron would be illegal!

http://www.ai.mit.ed...on-nitride.html

says 3.44 g/cc, and 304 GPa

http://www.designins...sider/m1017.htm

says 427 GPa, no density given

http://www.memsnet.o...tridesi3n4film/

gives a whole range of values for E from 104 to 380 GPa and 3.1 g/cc

[ZoRG]

Can anyone shed some light on the "Metalic" part? Ceramics are not a metalic, its a ceramic? am I missing something?

[desmo]

Silicon nitride is- someone correct me if I'm wrong- a ceramic and thus seemingly exempt from the specific modulus limit. In fact my reading of the limit is that ceramic reinforced MMCs whose metallic matrix complies with the limit whilst the ceramic reinforcing fraction- and thus the part itself in toto- might exceed the limit could well be argued as within the limit! It's been out there for quite a while as a potential material for poppets, I assume everyone has had a look at it and either adopted it (heard nothing of this) or dismissed it as in some fashion not up to the task. At the lifts and rpms that F1 engines run, the valves- particularly the exhausts- operate in an incredibly hostile environment and must be subject to considerable mechanical and thermal loads.

Maybe it's too brittle for the application, perhaps it's thermal conductivity or expansion is too low- or high. I've heard rumors of Al matrix MMCs being considered for intake valve duty.

[ZoRG]

I doubt its too brittle, it would actually be interresting to find out why they did not adopt it.

If you ever get your hands on a silicon nitride ball bearing, take one of those balls, get the biggest hammer you can find and see who wins, you will not brake it. It also has very low thermal conduction, and ceramics don't have as much thermal fatigue as metals, and can handle much higher temperatures than Titanium, at 800C titanium becomes quite weak, so in theory silicon nitride should work quite well as a exhaust valve since it can handle upwards of 1000C... It is difficult/expensive to make, and doing a flow polish on a valve might be a nightmare. It will problably also require silicon nitride valve guides and maybe seats.

[VAR1016]

Originally posted by ZoRG

Potato - Potaato good thing you knew what I meant!!

Ahh ok. Its a pitty since there is some real world application for this.

More like potato - teabag....

PdeRL

[desmo]

A brief bit from SAE 2002-01-3298
Advanced Ceramics in Formula 1 Wheel,
Clutch and Gearbox Rolling Bearings
Elmar Bergrath, Christian Klatt and Jens Wemhoener
Cerobear GmbH on Si3N4 as a bearing material for F1 applications:


"...it is not surprising that new materials have
found their way into F1 bearing technology. Today’s fundamental
milestone in advanced Formula 1 racecar rolling
bearing design is silicon nitride ceramics (Si3N4), a
material with only 40% of the specific weight of steel.
Bearing weight, moment of inertia and centrifugal
forces of the rolling elements on the outer race are
significantly reduced by using this new material.


Properties of Si3N4 vs. Bearing
Steel


Density: Bearing Steel 0.284 lb/in³ (7,85 g /cm³)
Si3N4 0.116 lb/in³ (3,2 g/cm³)

Max. Hertzian Pressure: Bearing Steel 580,150 psi (4.000 MPa)
Si3N4 725,188 psi (5.000 MPa)

Young's Modulus: Bearing Steel 3.05 E07 psi (210 GPa)
Si3N4 4.64 E07 psi (320 GPa)

Thermal Expansion Coeff.: Bearing Steel 10,8 µm/m K
Si3N4 1,9 µm/m K"

Silicon nitride has an absolute stable chemical structure
comparable to that of inert gases, which results in excellent
resistance to seizure and minimum lubrication requirements.
Rolling friction is 20 to 40% less than steel.
Resistance to Hertzian Pressure of Si3N4, important for
bearing lifetime, is even higher than that of bearing steel."

The paper goes on to say that Si3N4 element bearings are currently
used in F1 for wheel bearings, clutch release bearings,clutch
actuator bearings, input and output shaft bearings, and potentially
for even engine main and camshaft bearings.

So, there's apparently no question of Si3N4's legality as a material
in F1.

[J. Edlund]

"With composite materials we have the ability to increase stiffness but generally we are giving up strength to do that. It's hard to replace really advanced steels and titanium by a composite; I wouldn't foresee that happening in the near future.
I don't foresee the arrival of ceramic materials either, due to the dynamics involved. We need to have qute a bit impact resistance in the valvetrain. All of the components are seeing a very violent life and up to this point metals seem to be the most forgiving", Paul Jette, co-founder of Del West Engineering.

Basicly this can be translated into; ceramics aren't used in the valvetrain because they're too brittle.

However, what's on the way into valvetrains is titanium-aluminide, which is sort of a "low temperature superalloy" with the density of a titanium-aluminum alloy. The disadvantage is the material cost, the amount needed for a valve could cost as much as $500, this is however foreseen to sink very fast.

[ZoRG]

Thanks for the info!!!

[desmo]

Re Jette's quote above: Wouldn't titanium aluminide be considered a classic ceramic reinforced MMC composite? Quite a lot on the web on the material actually.

[J. Edlund]

Titanium-aluminide belongs to the group of "intermetallics".

[desmo]

Thanks. I've heard it referred to both as an intermetallic and an MMC.

[hydra]

Where's the advantage in lighter valves in an F1 anyway? With a pneumatic system valve bounce should be pretty nonexistent shouldn't it? And I doubt that the 200g weight saving is worth all the development time, effort, and cost. What I'm saying is that they'd be better off pursuing gains elsewhere.

[ZoRG]

Its, not about bounce, well actually it is ... the lighter the valves, the steeper the profile can be(Although I think its at its limits anyways) but it also allows for less pressure in the pneumatic system, so less friction to drive them.

The reason I asked about Silicon Nitride, was that it does not conduct heat like metals, and is better for ionization since ceramics cools so fast, with ceramic valves you might even get some ionization on the inlet valves, which means faster burn, and less chance of detonation.

[desmo]

When it comes to reciprocating mass, all else being equal, lighter is always better esp. in F1. Lighter valves reduce the stresses on the valvetrain, permitting more accurate motion control and/or less friction. The reduction of fMEPs is the most attractive means of increasing engine output in F1 as the benefits are a square or even cube function of engine speed.

The relative lack of thermal conductivity of Silicon Nitride might actually be a hindrance to their application as a valve material in F1. I'm told F1 engines run pretty close to the limit as far as having enough thermal energy available in the intake tracts to accomplish the gas/vapor exchange within the scant time allowed at 18krpm, and the majority of the thermal energy used to accomplish this is drawn from the intake valves.

And detonation is simply not an issue in F1 in the rpm ranges used. If it were physically possible, CRs would be even higher than they presently are.

[ZoRG]

I don't know if you get what I am saying... yes the thermal conductivity is low, however it also cools very fast, which is one problem of titanium, it conducts heat much faster, but it also has a problem of getting rid of it, although I am sure they use a heat barrier coating. Now the ceramics valve's face will get just as hot, actually maybe a little hotter, but the heat won't spread into the stem, and it will cool itself faster, the heat has to go somewhere, it will go into the new fresh mixture, making for much better ionization.

Have a look at this:

[Manson]

Originally posted by hydra
Where's the advantage in lighter valves in an F1 anyway? With a pneumatic system valve bounce should be pretty nonexistent shouldn't it? And I doubt that the 200g weight saving is worth all the development time, effort, and cost. What I'm saying is that they'd be better off pursuing gains elsewhere.

200 g savings in the valvetrain is huge! I use stainless steel as Ti is a custom order and out of my price range currently.

[ZoRG]

I found some affordable ones, let me know if your ineterrested.

[sblick]

A few weeks ago I read a story about a guy here in the states who has been developing carbon fiber valves for F1. I am not sure of the process, or what the matrix may be if there is one, but I think a quick internet search (I used Carbon-Carbon Valves) may provide details.

[ZoRG]

ahh yes, the ultimate material ... I don't think even the f1 guys will wan't to spend that much on valves, they are already trying to cut costs.

[McGuire]

Personally, I believe we will see non-metallic pistons before we get to valves, and probably in some form of two-piece composite construction.

[turby]

I don't think so, McGuire, the regulations do not allow carbon or aramid fiber reinforced materials for the piston. May be a piston made of glass fibre ??;)

[McGuire]

Originally posted by turby
I don't think so, McGuire, the regulations do not allow carbon or aramid fiber reinforced materials for the piston. May be a piston made of glass fibre ??;)

I was speaking in general, not of F1 in particular.

[J. Edlund]

Carbon/carbon pistons are already in use, not in F1 however. My guess is that they are used mostly in high performance two stroke engines where the heat is a larger problem than for 4 strokes.
http://nctn.hq.nasa....n54/complic.htm

Toyota has titanium and TiB reinforced titanium valves for production engines. The amount of TiB is 5%. The problem when using ceramics are otherwise, as already mentioned, that the stength of a part is decreased, there is however a gain in stiffness.

Titanium can't also withstand any higher exhaust temperatures in the long run. Therefore it's common for turbocharged engines to use nimonic or inconel instead.

The lower the mass of the valvetrain is, the softer springs, or lower pneumatic pressure can be used. This will decrease the friction loss from the valvetrain. The valvetrain is resposible for about 20% of the friction loss in an engine. This will increase power and decrease fuel consumption, it will also make the engine more quiet.

In F1 it also seems like the exhaust valves can be coated with a TBC, that part that is exposed in the exhaust port that is, not the parts in the chamber. The exhaust manifolds are also coated, but not on the outside.

[ZoRG]

I did some research about carbon/carbon pistons, I even got into contact with the company that made the test pistons for the US Army, thats all that they were, prototypes, to this day I have not yet been able to find anyone to actually use these, also the company said that the way its made means its not strong enough.

I found a manufacturer that makes titanium valves that can handle 800C with little strength loss, couple this with a heat barrier coating, and some decent valve guides/seats (better than beryllium copper) and you should be fine on turbo, inconel is very heavy. And indy cars run Ti exhaust valves with Turbo's.

[J. Edlund]

Originally posted by ZoRG
I did some research about carbon/carbon pistons, I even got into contact with the company that made the test pistons for the US Army, thats all that they were, prototypes, to this day I have not yet been able to find anyone to actually use these, also the company said that the way its made means its not strong enough.

I found a manufacturer that makes titanium valves that can handle 800C with little strength loss, couple this with a heat barrier coating, and some decent valve guides/seats (better than beryllium copper) and you should be fine on turbo, inconel is very heavy. And indy cars run Ti exhaust valves with Turbo's.

If you mean CART they run on methanol which reduce the exhaust temperature. The problem isn't also only the strength loss but that titanium takes up nitrogen which harden them and then they must be replaced. So titanium can't be used in the long run at high temperatures. The question is which has the highest specific strength at those temperatures, titanium or a metal like inconel or nimonic?

At 800 degC inconel or nimonic alloys can have a strength of about 100-200 MPa, that gives specific values of about 12.5-25 MPa. Then we must also concider fatigue, creep and oxidation, at these temperatures the Ni-based superalloys have excellent properties as long sulphur isn't involved (there are also better but more expensive materials availible than nimonic or inconel). The question is how titanium is at those temperatures, I haven't been able to find much data about that except that it start to oxidize at 500 degC.

It will also be interresting to see how it goes for valves in titanium aluminide, they can resist higher temperatures, they were even mentioned as a possible turbine material in low temperature diesel turbochargers. However at higher temperatures the advantages were back at the superalloys.

[ZoRG]

I will look into the nitriding... why would it be a problem if they harden though?

The valves I buy have a strength of 380-430MPa @ 800C and 110-140MPa @ 900C, the "race" version have 190-200MPa @ 900C, but these don't last long, I assume for the same reason you mentioned above.

[ZoRG]

These are the valves:







They will get a ceramic coating on the back of the exhaust valves, and on the valve faces.