Honda Power Unit Hardware & Software

[wuzak]

Still on valving the prechamber .... It should provide better control over timing of ignition/directing jets etc. This would work great; even better if you could control it with a solenoid. Just add a plug and drill some holes for the jets !

http://www.leeimh.com/images/dwg-inserts-reverse.gif

Could a solenoid react fast enough, maybe piezoelectric valve?

And would that count as a second injector?

[ncassi22]

Still on valving the prechamber .... It should provide better control over timing of ignition/directing jets etc. This would work great; even better if you could control it with a solenoid. Just add a plug and drill some holes for the jets !

http://www.leeimh.com/images/dwg-inserts-reverse.gif

Could a solenoid react fast enough, maybe piezoelectric valve?

And would that count as a second injector?

It could still act passively and be beneficial, but actively controlling it would be much better. As to the rules I have no idea. It depends on how they define injector. I'm just spit balling as there is still a week till any action If you want to go really crazy you could do something similar for egr, but I don't know how many holes the combustion chamber can stand! Also almost definitely illegal. Neither adds anything that did not come through the intake or fuel system though.

[glenntws]

Can I ask, why U-shaped crank pins?

In your analysis, did you include piston and rod accelerations, which must be significant at 10,000rpm+?

My first thought is that arrangement would induce a side load on the piston to compensate for the moment caused by having the support (the bearing) offset from the load (combustion forces and piston loads).

The U-shaped crank Pins increase the bearing area which in turn makes the cranktrain safer for higher loads. Also, the oil naturally gets pushed more outwards in higher rpms which Supports the lubrication in the complete bearing area.

Accelerations are not included, but I think they are not a big Problem. What I do think, is that the side load on the piston is fairly high at these rpms, so i think that a lower side support would be necessary (like you find it in many street vehicles).

Then again, it's pretty sure that Honda (like all other manufacutrers) uses steel as a material for the piston. From my experience, I would go for steel at everything higher than 20MPa Peak cylinder pressure. Using steel also dramatically increases the possible side load. I will make some screenshots of the simulations later.

Steel piston problem as far as the public knows is not solved yet for small petrol engines. Oil gets coked up in the rings. See the steel piston thread.

In regards to the cooling of the piston, you are right. But I think I found a solution for that problem. The most important thing to Keep in mind is that you need to increase the amout of oil squirting to the bottom of the piston.

Now think about that: You use a wet-running liner. Below the 2-3 o-rings, which are used to seal the water of from the oil, you could bring in a little space, where 360° around the Piston, you could bore 50 little holes squirting oil directly to the bottom of the Piston. this space could easily be fed with a simple bore down to the main oil line, which is often laying right on top of the crankshaft between the Banks.

I'll Show you some screenshots later of what I mean

[roon]

•Does the turbine benefit more from six evenly (as possible) spaced exhaust pulses, or fewer, larger pulses? Previous seasons showed us tuned length headers, so presumably thae answer at least used to be "six." Perhaps MGUH could 'torque fill' between bigger pulse-gaps.

Maximum number of evenly spaced exhaust pulses sharing an exhaust manifold without loss of blowdown energy is three. That is why the turbines have two entry ports with three cylinders feeding each. Using a 120 deg crank, each bank becomes an even-firing 3 cylinder which simplifies the plumbing - one bank feeds each turbine entry.

Gotcha, thanks. That true of the Merc, definitely. However the Renaults, 2014 at least, had single entry.



My main point is, is the turbine adversley affected by fewer, larger pulses? I'm imaging operating the V6 as a V-twin, firing an entire bank at once. The resulting three-cylinder exhaust pulse might get some heat retention benefits through the shared portion of the exhaust system. A larger pulse should dissipate heat for a longer period of time & conduction through the manifold wall is also time constrained. But I wonder if more constant flow is preferable for turbine operation than more intermittent, hotter flow.

[glenntws]

•Does the turbine benefit more from six evenly (as possible) spaced exhaust pulses, or fewer, larger pulses? Previous seasons showed us tuned length headers, so presumably thae answer at least used to be "six." Perhaps MGUH could 'torque fill' between bigger pulse-gaps.

Maximum number of evenly spaced exhaust pulses sharing an exhaust manifold without loss of blowdown energy is three. That is why the turbines have two entry ports with three cylinders feeding each. Using a 120 deg crank, each bank becomes an even-firing 3 cylinder which simplifies the plumbing - one bank feeds each turbine entry.

Gotcha, thanks. That true of the Merc, definitely. However the Renaults, 2014 at least, had single entry.

https://pbs.twimg.com/media/C40MgsjXAAAdEFT.jpg

My main point is, is the turbine adversley affected by fewer, larger pulses? I'm imaging operating the V6 as a V-twin, firing an entire bank at once. The resulting three-cylinder exhaust pulse might get some heat retention benefits through the shared portion of the exhaust system. A larger pulse should dissipate heat for a longer period of time & conduction through the manifold wall is also time constrained. But I wonder if more constant flow is preferable for turbine operation than more intermittent, hotter flow.

A continous exhaust flow instead of pulses is definetly preferred. In regards to TJI, a very efficient turbine design is the most important thing, because the exhaust gas pressure is already lower then in a conventional SI engine. The argument that the exhaust manifold also isn't vers strong. If you get 3 exhaust pulses at one time to the main pipe, the backpressure also increases. Also, the travelling distance in the single pipes to the main pipe will still be the same, so the heat-transfer would only be reduced in the throat area into the turbine.

I don't think that a 3-way V-Twin is used. Simply put, it's not only bad in regards to vibrations but also regarding to power delivery. A formula 1 car isn't like a motorcycle, where These little torque spikes could create a short controlled minislide. With the high downforce of the car, everything you want is a very evenly distributed power delivery.

The probably biggest indicator for a 1-4-3-6-2-5 firing order (even firing over the single banks) is the picture of the exhaust System of the RA616H. In case of multiple firings at the same time, the main throat would be way bigger than it is now (a bit bigger then the single cylinder pipes to compensate for exhaust pulse overlapping).

[godlameroso]

Firing order by itself is more important for the engine vibrations(and partly power delivery), Porsche had horrible vibration problems with their v4 engine in the 919, this was solved by altering the firing order. Toyota's failure at both Spa and Lemans last year was due to resonant frequencies in their v6 engine, they too had to alter their firing order.

I still think it's two even firing I3's like GG said, the clever bit is the timing between banks, where the second bank fires very close in time with the first bank.

Of course there's no way to know for sure, that's just what my ears tell me.

As far as how efficient these gases are at driving the turbine at low speeds, I'd say that these engines rely on the MGU-H a bit more than exhaust mass flow at low rpm. I would say that the engine itself doesn't produce enough exhaust gases to deliver meaningful boost until ~7,500rpm area. Without MGU-H aiding boost and MGU-K torque fill these engines simply wouldn't work.

[wuzak]

I still think it's two even firing I3's like GG said, the clever bit is the timing between banks, where the second bank fires very close in time with the first bank.

How would that work?

Given that the first bank has even firing impulse and the bank angle is 90°? Surely the timing between banks is, therefore, fixed?

[glenntws]

I still think it's two even firing I3's like GG said, the clever bit is the timing between banks, where the second bank fires very close in time with the first bank.

How would that work?

Given that the first bank has even firing impulse and the bank angle is 90°? Surely the timing between banks is, therefore, fixed?

It means that you have a firing timing of 90° - 150° - 90° -150° - 90° - 150°. Because both banks on their own fire evenly but because of the 30° crank offset (which results out of the 90°-V angle and the 3-crankpin rule) you get 90° - 150° insted of always 120°.

[wuzak]

I still think it's two even firing I3's like GG said, the clever bit is the timing between banks, where the second bank fires very close in time with the first bank.

How would that work?

Given that the first bank has even firing impulse and the bank angle is 90°? Surely the timing between banks is, therefore, fixed?

It means that you have a firing timing of 90° - 150° - 90° -150° - 90° - 150°. Because both banks on their own fire evenly but because of the 30° crank offset (which results out of the 90°-V angle and the 3-crankpin rule) you get 90° - 150° insted of always 120°.

I don't think that is what he was suggesting.

And it is hardly clever, being determined by geometry.

[godlameroso]

I was, but there's always something I fail to take into consideration. Luckily I have you guys to 'cunningham's law' me.

That is unless Honda is making an Ultra High tech and downsized version of the Buick Grand National engine. But that engine was a beast, so that would be fine by me.

[63l8qrrfy6]

Not sure what are the objections to fork and blade rods. In hydrodynamic terms you effectively have three bearing shells sharing the journal instead of two, ie four leakage paths instead of three. Since the two inner faces are low relative velocity (forked rod rocking relative to blade rod) and the axial tolerance between fork and blade much easier to control, the clearances could be closed up and perhaps even sealed completely with an o ring, piston ring or labyrinth seal. The result would be better hydrodynamic performance than side-by-side rods. From there you can reduce the length and or diameter of the crankpin, with a reduction in friction.

First off, the discussion started when someone proposed using forked rods for shortening the crank pin. This leads to a reduction in projected area regardless of whether you use conventional rods or forked rods.

Secondly it will be very hard to achieve a good bore stiffness for the forked rod big end bores.
On a conventional rod the beam widens close to the big end in order to provide stiffness and prevent bore distortion (elongation) at high tensile loads associated with high engine speeds. Additionally, as the 2 big ends on the forked rod are off-centre they will tend to tilt under load, leading to increased edge loading.

Not let's say you want to add some crowning to alleviate the high edge loading - what you end up with is 2 skinny bearings with even less area.

Also, 'sealing' the bearing is not necessarily helpful in high speed engines - you should actually be looking at encouraging flow out of the bearing in order to remove heat quicker - this is typically achieved by increasing the crush relief height and adding grooves in the side of the rod for crank guided rods.

[ncassi22]

Still on valving the prechamber .... It should provide better control over timing of ignition/directing jets etc. This would work great; even better if you could control it with a solenoid. Just add a plug and drill some holes for the jets !

http://www.leeimh.com/images/dwg-inserts-reverse.gif

Could a solenoid react fast enough, maybe piezoelectric valve?

And would that count as a second injector?

So I checked the rules. The actually don't limit the maximum amount of valves per cylinder, however they limit the type and diameter.

5.1.8 Engines must have two inlet and two exhaust valves per cylinder (NOTE THEY DON'T SAY ONLY). Only reciprocating poppet valves with axial displacement are permitted (SO BALL VALVING IS OUT). The sealing interface between the moving valve component and the stationary engine component must be circular.

5.3.3 Valve stem diameter must not be less than 4.95mm (SO IF YOU'RE CLEVER YOU COULD STILL MAKE IT WORK).

I'll get back to you on whether it'll be deemed an injector or not, however 5.9.2 of the tech regs state: Variable valve timing and variable valve lift profile systems are not permitted (SO NO ACTIVE VALVING EITHER , IT'LL HAVE TO BE PASSIVE )

[godlameroso]

Could a solenoid react fast enough, maybe piezoelectric valve?

And would that count as a second injector?

So I checked the rules. The actually don't limit the maximum amount of valves per cylinder, however they limit the type and diameter.

5.1.8 Engines must have two inlet and two exhaust valves per cylinder (NOTE THEY DON'T SAY ONLY). Only reciprocating poppet valves with axial displacement are permitted (SO BALL VALVING IS OUT). The sealing interface between the moving valve component and the stationary engine component must be circular.

5.3.3 Valve stem diameter must not be less than 4.95mm (SO IF YOU'RE CLEVER YOU COULD STILL MAKE IT WORK).

I'll get back to you on whether it'll be deemed an injector or not, however 5.9.2 of the tech regs state: Variable valve timing and variable valve lift profile systems are not permitted (SO NO ACTIVE VALVING EITHER , IT'LL HAVE TO BE PASSIVE )

Must, means no more and no less but exactly 2.

[ncassi22]

And would that count as a second injector?

So I checked the rules. The actually don't limit the maximum amount of valves per cylinder, however they limit the type and diameter.

5.1.8 Engines must have two inlet and two exhaust valves per cylinder (NOTE THEY DON'T SAY ONLY). Only reciprocating poppet valves with axial displacement are permitted (SO BALL VALVING IS OUT). The sealing interface between the moving valve component and the stationary engine component must be circular.

5.3.3 Valve stem diameter must not be less than 4.95mm (SO IF YOU'RE CLEVER YOU COULD STILL MAKE IT WORK).

I'll get back to you on whether it'll be deemed an injector or not, however 5.9.2 of the tech regs state: Variable valve timing and variable valve lift profile systems are not permitted (SO NO ACTIVE VALVING EITHER , IT'LL HAVE TO BE PASSIVE )

Must, means no more and no less but exactly 2.

Must is actually defined as 'being obliged' to or ' to be compelled by physical necessity to'. So you are not allowed to have less than the four; as they 'must' be there. However the wording does not limit it to just 4. It's the spirit of the law vs. the letter of it. To limit it to 4 they need to reword it to say 'must only'.

E.g. If you're stuck in a desert for two weeks you 'must' drink water to survive. I.e. your survival is subject to drinking water, but it doesn't say you can't have a burger and fries with it. If the object was to survive solely on water (i.e. say if eating the burger and fries would kill you), you'd say: To survive in the desert for two weeks you ' must only' drink water.

[Tommy Cookers]

Firing order by itself is more important for the engine vibrations(and partly power delivery), Porsche had horrible vibration problems with their v4 engine in the 919, this was solved by altering the firing order. Toyota's failure .....
........I still think it's two even firing I3's like GG said, the clever bit is the timing between banks, where the second bank fires very close in time with the first bank. .....

firing order has a big effect on torsional vibration (ie whether the crank breaks or not) and almost none on vibration exported to the chassis

the phasing of the pulse trains from the banks will depend also on bank-bank difference in system length (if any)
differences 2014 on seemed big enough (eg 20 cm) to cancel the train phase difference @typical rpm ie turbo can see a big I3 - as Renault 1977-79 ?

(to another poster) - how could any 1.6 12000 rpm 90 deg V6 give more vibration than the 2.4 20000 rpm 90 deg V8 ?