Honda Power Unit Hardware & Software

[Tim.Wright]

They just go for the tightest achievable tolerance (not necessarily repeatable as long as the amount of scrapped parts is acceptable). If it can be measured on a CMM then it goes on the drawing.

That would be a massive waste of time. 1 micron parts don't pop out every 5 minutes.

[PlatinumZealot]

The F1 parts have a smaller budget and shorter time limits. Formula 1 is entertainment. You only need to be beating your competition and no more. It warrant thats your tolerances (and hence machine time!) need not be as extrememe as that of a jet engine... There was nothing to be gained from being more than 2 seconds faster than your rivals... Nothing to be gained from having an engine last 10,000km... When the rules only need to to last 3000km. So this goes for the conventional engine as honda knew it. Conventional F1 tolerances for the conventional engine parts. Only thing is the long shaft turbo charger is nothing convential at all and the old tolerances that thy have been using just werent cutting it. They need better design tools and better machining tools AND knowledge base to make parts to those tighter tolerances and closer clearances.
It could be possible that Mercedes are using jet engine level tolerances as well... We just don't know.

this is going to be sort of an off topic rant, mods feel free to remove it if you feel it is out of place

Your logic contradicts few simple design rules, there is absolutely everything to be gained when you're 2 seconds faster than your rivals, and same for your engine lasting 10k instead of required 3k, because that means you can gain even more by shaving off of the components of your 10k engine to make it into a 3k engine while maintaining reliability, and that allows you to extend your lead to 2+ seconds - and if you have any strategical thinking at all, you will sand bag and show that you're only 0,5 seconds faster

Intentionally designing and manufacturing parts with loose tolerances (when you can do better) sacrificing performance in any sport is a neglect, there is no other way around it, and I would like to think that no one is actually doing that, I can't even think of any part on an F1 car that can be sort of neglected, everything counts for the lap time, and not just that, if you're a designer working on the bodywork, and you see people on the "wheel nut" team or the "head rest cushion" team slacking and saying something like - naah, it doesn't matter, it will never happen, let's go and grab some coffee after the extra long lunch break, well, then look at what happened in last few races where valuable points were lost to things that might have been improved upon (just as example, not blaming anything on anyone) - and this mentality is a cancer that spreads and destroys not just racing teams, but whole companies and even countries.

This all being said, I recognize that aero related engineering has its own challenges, but saying something like - oh we're far superior and we will come to fix your childish sand box level of a project called hybrid turbocharged 1,6l petrol engine that produces nearly 1000hp while being fuel efficient to extreme, then that person doesn't seem to know challenges involved in this design when compared to axial flow turbine engine, where performance is secondary to reliability, performance is still important, but you don't sacrifice reliability to it - this is what surprised me in that translated statement, the message between the lines was not - we will come to aid our fellow Honda engineers, but more like - ok, we will come and fix things because our work is much more challenging and this child's play project of yours

Have you ever sent out a part of CNC machining yet and asked for tigher tolerance?

[loner]

It hopes to be in a position to introduce 'spec 4' in one of the early races after the summer break, but will not do so until it is confident in the update.

http://www.autosport.com/news/report.ph ... -introduce

Is spec 4 the final spec 3 or is the real deal?

pretty obvious its the big upgrade coz final spec 3 don't need to wait its just some software and small changes.
Hasegawa is just confirming the spec 4 in October

[jz11]

Have you ever sent out a part of CNC machining yet and asked for tigher tolerance?

If you don't provide a context for your question, I will not answer it, because you will probably try to pull me into some nonsensical, out of this world way off topic discussion.

What I'm getting at is that the press release or whatever that translation was from, was simply wrong and you can find tight tolerances in F1 engines where it matters, and I'm also sure that they have people with equipment able to machine and measure things to make sure they are within the design parameters, and btw aero tolerances are not THAT far off the conventional world, of course, if your world doesn't end at laying bricks.

Yes, aircraft designs in average may contain higher tolerances, but more than likely that has no bearing on the vibration problem whatsoever, and doesn't guarantee, that someone with experience of putting a lot of zeros after a comma will be able to see where the problem lies with Hondas F1 engine, put proper amounts of zeroes in dimensions, provide them with the magical machining cutter and solve the problem, that is just silly. More than likely they simply want input from the designers involved in axial flow engines to get a fresh perspective on the problem they are solving, and maybe access to some specialized testing equipment and practices.

[PlatinumZealot]

Asking for tigher tolerances they charge you much more money and longer leadtimes because:

1) They have to use a better tool (or even just a new tool) sometimes and sometime additional tools have to used.

2) The parts takes longer to make. Might have to slow down the machine feed rate. Might have to have the part temperature controlled during the process.

3) Labour costs go up. More checks are needed. Hey, maybe they guy has to take the peice to another workstation to lap or polish or whatever.

4) Sometimes a different material or surface treatment is required to make the part to a tigher tolerance. (more stable) or a different material in the mould etc etc.

5) you may need an entirely new machine. Part may have to be sent out to use such machine.

And this is just about geometrical tolerances. We are not even talking about mechanical tolerances yet.

The cost and time can go up immensely just by moving the tolerance by one decimal place.

On side note. People always associate tolerances with tight fits and gaps and holes etc. Not always the case. It could a mass, having to do with balancing.. Surface hardness etc material compostions etc etc.

[godlameroso]

These reports aren't new news, Honda doesn't allow details out while the information is relevant to what they're doing. For example they showed their 2015/16 power unit in any amount of detail only after they abandoned the concept. We only just got a partial glimpse at the 2017 power unit after spec 2. So if they were drawing on experience from the jet division, it probably happened several months ago. Probably around the Bahrain test, which may have been the "outside" consultants all along.

[loner]

These reports aren't new news, Honda doesn't allow details out while the information is relevant to what they're doing. For example they showed their 2015/16 power unit in any amount of detail only after they abandoned the concept. We only just got a partial glimpse at the 2017 power unit after spec 2. So if they were drawing on experience from the jet division, it probably happened several months ago. Probably around the Bahrain test, which may have been the "outside" consultants all along.

yep Schmidt the reporter said it few weeks ago they will gain decent power in 2017 and will attack Mercedes in 2018
point is.. Honda plan already known months ago.. just not to the average Joe.

[63l8qrrfy6]

They just go for the tightest achievable tolerance (not necessarily repeatable as long as the amount of scrapped parts is acceptable). If it can be measured on a CMM then it goes on the drawing.

That would be a massive waste of time. 1 micron parts don't pop out every 5 minutes.

This is in the context of a few important dimensions on a critical component. I'm not implying that anyone specifies 0.1 mm profile tolerances on the outside surface of a cam cover just because.

@PZ Really now, when has anyone cared about costs in F1? Do you seriously believe that F1 engine manufacturers are tighter than manufacturers of production aero engines? If your argument is constructed around the idea that tolerances are slacker because they care about prices then well.. you are wrong.

[wuzak]

Will it be better to have a common casing for the compressor, mgu and turbine?

No, I don't think so.

For a start it would make it harder to assemble. And it may also mean that when you change an failed MGUH you are also changing the TC.

[j.yank]

Is it possible the current engine to have TJI but because of the vibrations in the turbo area they had to run the ICE severely detuned? The turbine, MGU-H and compressor are on common shaft and vibrations in MGU-H means that the compressor is running on lower speeds which means less pressure. If they fix the vibrations in MGU-H could we suddenly see that the current engine has TJI? This also could explain why Wazari said that he doubts that his version of the ICE will be implemented by the end of the season. Probably they do not need it anymore?

[FW17]

Will it be better to have a common casing for the compressor, mgu and turbine?

No, I don't think so.

For a start it would make it harder to assemble. And it may also mean that when you change an failed MGUH you are also changing the TC.

With the single casing the number of bearing will be significantly reduced from 4 to 2

Vibration on the shaft is better controlled as the 3 components can be tuned as one

Changing component should not be a problem as you are opening a casing to change a part with in.

[adrianjordan]

Will it be better to have a common casing for the compressor, mgu and turbine?

No, I don't think so.

For a start it would make it harder to assemble. And it may also mean that when you change an failed MGUH you are also changing the TC.

With the single casing the number of bearing will be significantly reduced from 4 to 2

Vibration on the shaft is better controlled as the 3 components can be tuned as one

Changing component should not be a problem as you are opening a casing to change a part with in.

Would it possibly increase the chances of compound failures though? So when one goes, the others in the same housing could be damaged by debris etc from the failed unit?

[63l8qrrfy6]

Having one common housing would mean that it has to be horizontally split through the middle of the bearing bores which is a big no no for high speed turbo machinery.

Secondly the turbine housing end has to be inconel - that would dictate that the whole housing is inconel which is denser than steel and almost 3 times denser than aluminium.

The vibration benefits are questionable to say the least.

[JonoNic]

Honda have seemed to fix their PU issues

[Sasha]

People have very short memories on here.

TJI was found to not work correctly in Dec 2016 so they had to run it detuned(richer) at the beginning of the season.

Vibrations wasn't found until first in car test(Feb).

Two separate problems.