Honda Power Unit Hardware & Software

[Tommy Cookers]

[quote=Mudflap]
Forgings are typicall stronger as the grain is forced to flow parallel to the faces of the part. The forged part will exhibit anisotropic properties which can be taken advantage of - for example a shaft will be stronger in bending if the grain flow is parallel to the neutral axis.
......An as-forged fillet will always be stronger than an as-cast/machined fillet .....
I know that in V8 days Cosworth used a sump machined from cold rolled 7075 billet that had high directional strength in one plane. Interestingly they experienced failures and had to move to an isotropic alloy........[/quote]

sorry about the crappy quoting - but there is a bit of the forging myth-making

there is no grain in (hot) forging - which is what people have been talking about
because after working the crystals reshape themselves to their original state
so there is no improvement in mechanical properties due to forging (beyond the best innate properties of the material)
there is a guarantee of these best properties in the direction of working (this is wrongly called the 'grain')
this is why the part is made so the direction of working is normal to the loads in use (why you forge a crankshaft shape)
the working means that any inclusions (being less than ductile/plastic) are bridged by the reforming of 'good' metal

you cannot improve strength or stiffness (elastic modulus) by (hot) forging
you can improve 'surface' strength by cold working of course and so have a slightly anisotropic material or part
there is 'stretch hardened' (strengthened really) alloy plate for aviation
and bolts - and shot-peened con-rods of course

there may be some cold- working on (hot) forged parts - some apparently forged parts are actually (cold) stampings

if you can improve elastic modulus (you can't) then patent your process and make trillions
true there's a small gain in E with a bit of lithium in al alloy and a slight reduction in density - together handy for aviation
people often mistake a high yield strength (strain really) for unusually high E

a machined from solid part is probably not as good as a part machined from a part-shaped forging

[63l8qrrfy6]

I worded the post to read as 'forgings are better than castings' rather than 'forgings improve the mechanical properties of an ideal specimen' - at least that was the intention.

Grain flow occurs regardless of whether the part is cold or hot forged - of course as you have noted the difference is that the grain structure is not preservered after recrystalization. If anything, hot forging improves the microstructure (again, with respect to a casting).

I've always refered to the striations visible in a forging section as grain flow lines - if that's not what they are called then I stand corrected.

The obvious issue with residual stresses in forged parts is that they are not always beneficial. It is often difficult to asses the effects on complexly loaded components subjected to multi-axial stresses.

[roon]

People like to make a big deal of the McLaren "size 0" thing, but the reality is all of the engines are extremely compact and packaged very tightly.

I agree, I think this was overemphasized and may have been a hindrance to Honda. Prioritizing chassis design/layout over engine design/layout, as imposed by McLaren. The tight 90* entry into the compressor is a major compromise. Honda has the Merc layout, the only one which offers a straight shot into the compressor inlet, but couldn't take full advantage of it in the McLaren chassis. I suspect STR will be able to provide Honda a corresponding Merc-type fuel tank & safety cell design.

There are many reasons why you would use machined billet parts over cast.
The first of which is that a machined piece is normally stronger than its cast counterpart, especially important in a fully stressed engine component.
Secondly casting is expensive to setup and time consuming - it's more for large production parts than short run parts.

But I'm betting the cylinder head on the Mercedes engines are not machined and are cast along with the engine block.
It's incredibly difficult to machine a billet head or block with coolant passageways as intricate as on an F1 engine.

Everything I engineer on F1 cars is made from billet.

Do the F1 manufacturers not follow what BMW did with their 2.4l V8?

They cast the head and cylinder block as a single unit to get rid of the head gasket weakness... after reading the BMW whitepaper by Theissen, I figured all manufacturers would follow suit...

No not really because of the incredible difficulty of machining the valve seat pockets, seats and combustion chamber that comes with doing it this way.

And BMW only did that for one year during the development of the P85 engine which never raced because of rule changes that made engines have to last a lot longer.
You can read about that and the casting in the white paper you talk of here:

http://sd-2.archive-host.com/membres/up ... BMW_F1.pdf

Interesting that they designed a direct injection system for it.

[MrPotatoHead]

People like to make a big deal of the McLaren "size 0" thing, but the reality is all of the engines are extremely compact and packaged very tightly.

I agree, I think this was overemphasized and may have been a hindrance to Honda. Prioritizing chassis design/layout over engine design/layout, as imposed by McLaren. The tight 90* entry into the compressor is a major compromise. Honda has the Merc layout, the only one which offers a straight shot into the compressor inlet, but couldn't take full advantage of it in the McLaren chassis. I suspect STR will be able to provide Honda a corresponding Merc-type fuel tank & safety cell design.

STR are way behind on their design process because of the large amount of work they have to do to make the Honda engine fit.
As a result they will more than likely have compromises in the design.
They are way behind on production. Parts that were being made for RBR months ago aren't even in the pipeline for STR yet.

Do the F1 manufacturers not follow what BMW did with their 2.4l V8?

They cast the head and cylinder block as a single unit to get rid of the head gasket weakness... after reading the BMW whitepaper by Theissen, I figured all manufacturers would follow suit...

No not really because of the incredible difficulty of machining the valve seat pockets, seats and combustion chamber that comes with doing it this way.

And BMW only did that for one year during the development of the P85 engine which never raced because of rule changes that made engines have to last a lot longer.
You can read about that and the casting in the white paper you talk of here:

http://sd-2.archive-host.com/membres/up ... BMW_F1.pdf

Interesting that they designed a direct injection system for it.

Indeed. Until fuel pressure was capped.

[ncassi22]

Honda's recently filed patent for a magneto-rheological torsion damper.

http://www.freepatentsonline.com/y2017/0363171.html

[Snorked]

Motorsport Technology 2017-2018

https://www.amazon.co.jp/dp/4779634903/ ... oAbBE0GKQS

I am keenly editing and writing work, but I will inform you that the cover has been completed.



The cover of the 2017 edition, which is the fifth book (5th year) as "technology of motorsports", is Honda's F1 power unit, RA 617H.

Since I wanted to post alongside two past works (RA 615 H in 2015 and RA 616 H in 2016), I took it at the same angle (as a professional photographer by example). It is clear from the side that the change from 16 to 17 years is much larger than the change from 2015 to 16 years.

It summarizes the change point and its background. Also, not only the parts that can be seen from the outside, but also the contents of the contents (combustion concept)

I covered not only the engine's story but also the energy unit's energy management. In 2017, the full-open rate has been rising compared to the previous year, but as the full-open duration becomes longer, the time to deploy (energy release) also becomes longer, and the energy becomes insufficient.

It is the center of the story about how we coped with it.

http://serakota.blog.so-net.ne.jp/2017-12-23

Looks to have telemetry from Spanish GP:

[godlameroso]

Sounds like they improved deployment but not enough to cover the higher amount of full throttle these new cars are capable of.

[PlatinumZealot]

Spanish GP was early days when they just got the thing working to a pre-season level so I don't expect the best energy management there. Japan would have been lovely telemetry.

[dren]

I'd love to see that telemetry blown up. Were you guys able to view any of it at a decent size?

[Joseki]

Spain telemetry is from the alien lap Alonso did in Q3?

[Jolle]

Looking at the pictures, it looks to me that they ditched the charger in-between the cylinder banks and lowered the turbo and but the charger in front of the ICE, even more then Mercedes have. And it that in intercooler worked into the intake ontop?

[ncassi22]

A few months ago pre-chambers were discusses here. Recently I ran across ideas for co-axial injector needles and wondered whether it could be used in pre-chamber design for F1. This is a conceptual idea. Would be interesting to know what you guys think?

[hurril]

An interesting picture! How would the richer mixture enter the area with the spark plug?

[Snorked]

I'd love to see that telemetry blown up. Were you guys able to view any of it at a decent size?

As with the previous magazines, Amazon put up full resolution previews after release date (usually 15-20 pages) - you have to click on the magazine cover to load the viewer - I can't check if it's available because I'm on my phone, but we should get the full pages soon.

[ncassi22]

An interesting picture! How would the richer mixture enter the area with the spark plug?

Apologies, I should have annotated the thing. There's a small port that's exposed when the outer pintle pushes forward. It'a in the sketch ,but easy to miss. I see no reason why, instead of having the chamber integrated with the injector, you can't branch it off into a hydraulic passage with a one way valve... so as soon as the injector opens a remote pre-chamber gets fed. There's another one I have that uses one driver that reverses direction to essentially do the same thing. (I have no idea if that's possible)