Honda Power Unit Hardware & Software

[PlatinumZealot]

Easier said than done. They are making the parts and testing them... Just that we wont see those parts on the track unless they deliver.

The hardest part is solving the problems. If the problems are not solved on the bench theft won't be solved on the track. That box has to be ticked first.

[MrPotatoHead]

Easier said than done. They are making the parts and testing them... Just that we wont see those parts on the track unless they deliver.

The hardest part is solving the problems. If the problems are not solved on the bench theft won't be solved on the track. That box has to be ticked first.

I've been giving them the benefit of the doubt up until now, but if they don't bring a serious update to Canada then their efforts will look like a joke.

Parts I've touched have won half of the races this year. Watching Honda not be there winning as well is frustrating. I want them to be competitive. If I am giving them a hard time now it's because of those frustrations.
But I do think they will get things figured out. It's just taking longer than we all would like.

[ziggy]

I'm amazed they haven't made any major upgrades yet. They have had 4 or 5 months since they knew they had problems on the engine dyno. That is a lifetime in an engine formula without tokens.
In a world of investment casting and CNC machines they "should" be turning around parts in days or weeks not 6 months.

I don't think so. If they had already solved their problems and it were just a matter of producing parts it will not be take so long. But they have to reach a certain define of problem and certain solving of it. I don't think that this is just model and produce job. İf they produce it without more power and reliability, they will be really fool themselfs. They have to be sure they got more power and they have to be sure they are more reliable. This is not a small update related small part of engine. After findind correct solution I think molding process is not so short and after production I think they have to do lots of test.
İn the other hand they did some small updates.

My point is that they should be making Continuous improvements. Not some minor updates after 4 races.
They should be running new parts on the dyno every day.

I wrote about the reasons a few pages back, but got labeled a racist for posting the truth with links. Long story short (my sources are only official articles) they knew of the problems but were too proud to take some "outer" brain to the team. When Hasegawa took over, it was allready too late and pressure too high.

After some research, I think a whole new head lead time should be no more than 4-6 weeks, if you know, what you have to do. If not, this time can extend into 6 months or more. It's been 10 weeks now since the testing, so I think Honda is still in the window and will bring a very strong update on the ICE, electrical side is still big question. But maybe Mario Illien is helping with the ICE and AVL on the electrical side?

[MrPotatoHead]

I don't think so. If they had already solved their problems and it were just a matter of producing parts it will not be take so long. But they have to reach a certain define of problem and certain solving of it. I don't think that this is just model and produce job. İf they produce it without more power and reliability, they will be really fool themselfs. They have to be sure they got more power and they have to be sure they are more reliable. This is not a small update related small part of engine. After findind correct solution I think molding process is not so short and after production I think they have to do lots of test.
İn the other hand they did some small updates.

My point is that they should be making Continuous improvements. Not some minor updates after 4 races.
They should be running new parts on the dyno every day.

I wrote about the reasons a few pages back, but got labeled a racist for posting the truth with links. Long story short (my sources are only official articles) they knew of the problems but were too proud to take some "outer" brain to the team. When Hasegawa took over, it was allready too late and pressure too high.

After some research, I think a whole new head lead time should be no more than 4-6 weeks, if you know, what you have to do. If not, this time can extend into 6 months or more. It's been 10 weeks now since the testing, so I think Honda is still in the window and will bring a very strong update on the ICE, electrical side is still big question. But maybe Mario Illien is helping with the ICE and AVL on the electrical side?

Your comments about Japanese work culture were not especially wrong. Maybe not politically correct, but not wrong. I read them at the time, I probably should have posted then.

I've been to Japan twice in the last 12 months for business. Things are very different there - unless you've seen it you won't really understand.

And yes if you know what you are doing you can turn around new parts in weeks not months. Even with traditional sand cast heads you can "rub" the sand molds to make changes and have fresh castings in days. Add a few more days or weeks if starting from scratch for the CNC work and you are golden.

[nzjrs]

And yes if you know what you are doing you can turn around new parts in weeks not months. Even with traditional sand cast heads you can "rub" the sand molds to make changes and have fresh castings in days. Add a few more days or weeks if starting from scratch for the CNC work and you are golden.

That they encountered so many integration level problems when they put the car together this year makes me think they lacked some test facilities. One would hope they have remedied that and now place even more emphasis on dyno running that more approximates real life.

Given that you sound experienced here, whats the normal ratio of validation/test time to design time. If you do a minor design iteration in 5 days, would you test for 5 or 50 race distances before you had enough data to trust and iterate again?

[MrPotatoHead]

And yes if you know what you are doing you can turn around new parts in weeks not months. Even with traditional sand cast heads you can "rub" the sand molds to make changes and have fresh castings in days. Add a few more days or weeks if starting from scratch for the CNC work and you are golden.

That they encountered so many integration level problems when they put the car together this year makes me think they lacked some test facilities. One would hope they have remedied that and now place even more emphasis on dyno running that more approximates real life.

Given that you sound experienced here, whats the normal ratio of validation/test time to design time. If you do a minor design iteration in 5 days, would you test for 5 or 50 race distances before you had enough data to trust and iterate again?

Given the season so far I think lasting 2 race distances would be cause for celebration.

[Sasha]

One of the persons who designed the new CC/heads/pistons stated it would take about 4 months.(stated in late Feb)
July? after Summer Break?

And that timeframe is only if there is no major problems.

Also the vibrations are like a Godzilla running in the building!

[MrPotatoHead]

One of the persons who designed the new CC/heads/pistons stated it would take about 4 months.(stated in late Feb)
July? after Summer Break?

And that timeframe is only if there is no major problems.

Also the vibrations are like a Godzilla running in the building!

If the design work is done it should not take that long.
They should be aiming for half of that.

[ziggy]

One of the persons who designed the new CC/heads/pistons stated it would take about 4 months.(stated in late Feb)
July? after Summer Break?

And that timeframe is only if there is no major problems.

Also the vibrations are like a Godzilla running in the building!

If the design work is done it should not take that long.
They should be aiming for half of that.

Depends on the situation. If you have to develop a new material for lets say valves or turbo, it can take that long or longer for sure. That's the difference between in house developing and buying a ready developed product from a partner or supllyer. Time.

Second big problem is packaging. Changing the head can lead to numerous other changes, basically to a "new" PU.

[etusch]

I don't agree people who thinks that out help will make things seriously faster for Honda except tji part. I accept that understanding tji, it can make things faster, but for solving Honda's problems, adviser or engineers coming from another place must understand Honda's concept and find out where did they it wrong. Then finding a solution. İt is same road but with more knowledge and more experience.

[mwillems]

One of the persons who designed the new CC/heads/pistons stated it would take about 4 months.(stated in late Feb)
July? after Summer Break?

And that timeframe is only if there is no major problems.

Also the vibrations are like a Godzilla running in the building!

If the design work is done it should not take that long.
They should be aiming for half of that.

They didn't fully understand the engine they placed their focus upon to bring to the track. If they did not fully understand that engine or it's problems, then clearly they are going to know a lot less about their secondary concepts, as it appears that these concepts only made it to single cylinder testing before choosing the concept to move forward with. Either way you look at it, there is a level of naivete. Either now, or back then. Personally, it seems to me that they are doing the work now that they should have done prior to the seasons start, and ensuring that the design is fully developed and tested before bringing it to the track.

They will always look like a joke, right up until the moment that they don't. It sounds like a stupid thing to say, but the moment Mclaren and Honda starting knocking on 6th place more than just out of luck, then the half empty glass will be half full, in a heartbeat. Perceptions change fast in this sport.

I also think that it will be easier for people to accept an engine to appear later that delivers the goods, than for an engine to appear a little earlier that still needs some work to look respectable, even if they start performing at the same level at the same time. Otherwise it is another engine that doesn't work and a further perceived setback. I'd imagine that would also affect Alonso too, that if the update came and broke down or didn't quite perform, even if two races later it was up to scratch, the final nail might in the coffin for his contract renewal.

This is my speculation but they have to draw a line under the seasons start, and moving forward, they have to get it right, first time and every time. That, for me, will be progress.

[gruntguru]

Agree. "Incredibly rich" is difficult to ignite - particularly without spark. Best implementation of TJI would have slightly richer than stoich' in the pre-chamber - probably 0.9 - 0.95

Exactly. In my mind the only way to get it to work with any kind of precision is for the injector to be inside of (or to be) the pre-chamber. The idea that you could somehow convince the fuel from the main combustion chamber to go hide inside a pre-chamber with exactly the right ratio... is far fetched at best.

When you start to think about the time window available for these events to take place it's mind boggling. At 10500 rpm 1 stroke takes only 2.7ms. At 12000 rpm it's only 2.5ms.

Not sure I agree. In fact injection precision/resolution is a reason not to have the injector in the pre-chamber. Remember the pre-chamber only gets about 2% of the total fuel load. I refer back to the idea I posted earlier.

Direct injection is quite capable of stratification and producing a relatively consistent mixture strength in a selected region of the combustion chamber. What end up in the pre-chamber depends on mixture strength in the adjacent zone - during the compression stroke (because the pre-chamber is "filled" during the compression stroke). So I am becoming increasingly convinced that TJI can be implemented/controlled by:
1. Creating a homogeneous, lean mixture in the main chamber by injecting during the intake stroke.
2. Creating a rich kernel in the region of the pre-chamber during the compression stroke - resulting in a richer mix in the pre-chamber.

[PlatinumZealot]

With those fine nozzle holes in the pre-chamber.. Why would the fuel mist move inside the pre-chamber easier than the main chamber which has no such restrictions to moovement? Worse it is driving against a dead end and possible high concentration gradient. Not sure getting inside the pre-chamber richer would happen without alot of momentum in that direction..

[Sasha]

Injector is always in pre-chamber under the rules of only one per cylinder.

[mwillems]

Injector is always in pre-chamber under the rules of only one per cylinder.

Probably this a is a stupid question, but is it possible that the one injector could span two chambers, with two exits? What exactly constitutes a second injector?