Honda Power Unit Hardware & Software

[Thunder18]

The last three pages of posts can be deleted with nothing lost. I know it's getting close to testing, so maybe off to the testing or team thread? I have to scour through pages of posts to find one or two meaningful ones.

I think we all have to say thanks to Wazari for posting what seems like credible and honest information over the past few months.

A few tidbits he posted over the last few weeks:

1. ICE gain around 3 to 6%
2. MGUH, turbine/compressor changes
3. Improved MGUH output
4. Engineering seems to be pleased with improvements

Now we wait for testing.

+1 for Wazari input, the only reason I keep coming back, (hoping he's legit)

On the speculation side of things, I imagine if McHonda are tight-lipped about potential progress, the media start planting seeds in order to generate a reaction from McHonda, therefore enabling them to gather more info for more speculation.

E.g. The 223bhp gain, McHonda reacted to that story, and if they bit once, they might bite again

[Wazari]

The items I have posted are what I understand to be IMO accurate and from reliable sources within Honda. Also unless it's coming from Honda or McLaren personnel, all the other stuff I would consider to be pure speculation which media is infamous for doing. The items I have posted I really wouldn't consider "earth shattering" news either.

The 223 HP figure was very clever. IMO it's just another way of saying Honda found 60 HP increase from their ICE which is plausible but again IMO maybe a little optimistic.

Really testing will be the initial tell all. Everyone at McLaren-Honda is working at a frenzied pace. I understand that initial dyno results have met or exceeded expectations. Few minor issues as expected, from what I understand mostly sensor related issues.

I am excited to see the results of the first test days and see what the gap is between Mercedes and the others.

[hemichromis]

The items I have posted are what I understand to be IMO accurate and from reliable sources within Honda. Also unless it's coming from Honda or McLaren personnel, all the other stuff I would consider to be pure speculation which media is infamous for doing. The items I have posted I really wouldn't consider "earth shattering" news either.

The 223 HP figure was very clever. IMO it's just another way of saying Honda found 60 HP increase from their ICE which is plausible but again IMO maybe a little optimistic.

Really testing will be the initial tell all. Everyone at McLaren-Honda is working at a frenzied pace. I understand that initial dyno results have met or exceeded expectations. Few minor issues as expected, from what I understand mostly sensor related issues.

I am excited to see the results of the first test days and see what the gap is between Mercedes and the others.

To be honest i find it easier to believe you - keeping the above disclaimer in mind - then all these reports we hear from dubious sources reporting news from dubious sources.

[ollandos]

my view is ...honda make the hardware and desing changes and now is on stage to max. the software controls and find the tunes ....journalist they dont now for what they are speaking ..........i think honda can see very soon mistakes and dont loose time to push wrong ideas.... ..for sure is back from mer-fer...but they can find more and more faster and faster....if they dont blow up and make the right job i wait them to others level after symmer break maybe.....

[Andres125sx]

http://www.marca.com/motor/formula1/201 ... b45a5.html
power normal, reliability bad

Just in case someone is interested but can´t read spanish, the article basically says what Lucky sumarised, but a bit more extended.

They claim power is good, but some turbos have failed on dyno tests. Also claim overheateing is still an inssue, and turbine speed has been decreased from past year´s 120k to 100k rpm, but overall they´re happy with ERS perfomance wich has improved significantly

Not sure how accurate this info is, but when I read details about specific parts, I tend to believe the article a bit more. I know they´re only journalists, but they´re paid to get information before anyone else

[Per]

You're the one that keeps saying you can't think of these engines as just turbocharged engines, you are not even considering why the turbo needs to be sized in this specific way. I'll let you explain why moving to larger more efficient turbos would help heat recovery then. I promise I won't say another word on the subject, just please answer me this, why does a larger more efficient turbo aid heat recovery?

Sorry to dig up this rather old post but I didn't have the chance to reply sooner. You really need to treat the compressor and turbine as two separate components when you discuss their sizing. The only thing they have in common is their RPM.

Compressor sizing is driven by what the ICE needs. It depends mainly on fuel flow, air-fuel ratio and desired in-cylinder pressure. All of this defines the required compressor work.

The turbine is different. Its massflow also comes from the ICE, but whereas in a conventional turbocharged engine the turbine work only needs to equal compressor work once the desired turbo RPM is reached, in the current F1 engines the turbine work simply needs to be as large as possible because any excess work can be harvested by the MGU-H.

Therefore the link between turbine size and compressor size in F1 is not as strong as in other turbocharger applications. When discussing Honda's redesign of the turbo to increase MGU-H harvesting, the compressor isn't even necessarily part of the story. If you get the compressor right once [to provide the necessary air conditions to the ICE], you can change the turbine capacity without changing the compressor [as long as you don't change the turbo RPM].

This is also why Mark Hughes is right when he says that the desired turbine capacity is related to the efficiency of energy transfer on the path MGU-H - controller - MGU-K. Taking more work out of the turbine inevitably means higher back pressure and lower ICE output. It's only worth it if this lower ICE output is compensated by higher MGU-K output - it is always a trade-off.

So like Mark Hughes says, if you can get the power electronics to work more efficiently, you will also want to increase turbine work. Whether or not this means actually having to make the turbine larger or smaller has already been commented on by other people in this topic who are more knowledgeable than I.

[dren]

Yes, when looking at individual components' needs. But the compressor's needs may change when you start increasing the loading the MGUH puts on the turbine due to higher back pressures and delta Ps. It's one great big intertwined efficiency optimization puzzle. Change something upstream, it affects something downstream. Change something downstream, you'll have to make changes upstream.

[godlameroso]

It's all linked together, you have to design components to maximize efficiency. You can get 120kW from the mgu-k period, no more but also less. The MGU-H can harvest whatever it wants, you want to be able to harvest as much heat as possible, I think the limit is probably around 170kW that can be harvested from the turbine. In order to reach that you have to use a bigger turbo in general, not just because more air will allow you to burn fuel more efficiently and generate more heat which in turn generates electricity, but also because a larger turbine will have less back pressure which mitigates pumping losses aiding the cycle. Since the ES takes care of any lag it doesn't matter if you have a huge laggy turbine, it just has to produce enough electricity when the engine is at full throttle on the straights(or whenever it can harvest) to be worth it. Another benefit from a large turbo is that it will deliver it's intended boost at a lower than the max allowed rpm.

Why would you want this? Simply put, you could give your MGU-H an overdrive ratio, which would also create gear reduction in motor mode, this would improve compressor output which is unlimited, further mitigating turbo lag creating more exhaust pressure at lower ICE rpm and more available energy to harvest over a broader rpm range.

It all ties in together, you need to create more cylinder pressure to create more electricity to give you more power.

If you have 480kW from the ICE and 120kW from the K for 45 seconds are you better off than someone with 500kW and 120kW from the K for 38 seconds?

[gruntguru]

You really need to treat the compressor and turbine as two separate components when you discuss their sizing. The only thing they have in common is their RPM.

. . . . and the massflow. Turbine massflow is always equal to compressor massflow plus fuel flow (about 4 or 5% extra)
. . . . and the pressure ratio. Turbine PR cannot be a lot different to compressor PR.

Exhaust flow, pressure ratio and temperature dictate the turbine size.

[godlameroso]

It's almost as if the new engines are treated like turbine engines with the ICE serving as the cold and hot stages of a conventional axial compressor turbine.

Something like this but with an engine in the middle with an energy recovery system
https://m.youtube.com/watch?v=-1cN5di4osE

The engines are very clever, orchestrating all the interrelated parts takes time. It's a shame that the engineering that goes into making these power plants is so secretive.

[pgfpro]

You really need to treat the compressor and turbine as two separate components when you discuss their sizing. The only thing they have in common is their RPM.

. . . . and the massflow. Turbine massflow is always equal to compressor massflow plus fuel flow (about 4 or 5% extra)
. . . . and the pressure ratio. Turbine PR cannot be a lot different to compressor PR.

Exhaust flow, pressure ratio and temperature dictate the turbine size.

From my own experience the compressor PR and turbine PR don't always run close PR numbers. Depending on rpm, boost there can be as much as a 30% difference.

My DD turbo Talon, compressor PR verse turbine PR is
30% PR difference @ 4500 rpm@25 lbs.
12% PR difference @ 5500rpm@28lbs.
9% PR difference @ 6500rpm@28lbs.
0% PR difference @ 7500rpm@28lbs.
9% PR difference @ 8500rpm@28lbs.
Now on this setup its kinda a hybrid turbo, but even when I match the compressor size more to the turbine size the numbers on my turbo simulator show a higher PR % difference.

I have a lot more examples from customer cars with data logs showing more of the same.

[gruntguru]

You really need to treat the compressor and turbine as two separate components when you discuss their sizing. The only thing they have in common is their RPM.

. . . . and the massflow. Turbine massflow is always equal to compressor massflow plus fuel flow (about 4 or 5% extra)
. . . . and the pressure ratio. Turbine PR cannot be a lot different to compressor PR.

Exhaust flow, pressure ratio and temperature dictate the turbine size.

From my own experience the compressor PR and turbine PR don't always run close PR numbers. Depending on rpm, boost there can be sometimes as much as a 30% difference.

My DD turbo Talon, compressor PR verse turbine PR is
30% PR difference @ 4500 rpm@25 lbs.
12% PR difference @ 5500rpm@28lbs.
9% PR difference @ 6500rpm@28lbs.
0% PR difference @ 7500rpm@28lbs.
9% PR difference @ 8500rpm@28lbs.
Now on this setup its kinda a hybrid turbo, but even when I match the compressor size more to the turbine size the numbers on my turbo simulator show a higher PR % difference.

I have a lot more examples from customer cars with data logs showing more of the same.

30% is not a lot in the context of my post.

[Abarth]

[...]From my own experience the compressor PR and turbine PR don't always run close PR numbers. Depending on rpm, boost there can be as much as a 30% difference.

My DD turbo Talon, compressor PR verse turbine PR is
30% PR difference @ 4500 rpm@25 lbs.
12% PR difference @ 5500rpm@28lbs.
9% PR difference @ 6500rpm@28lbs.
0% PR difference @ 7500rpm@28lbs.
9% PR difference @ 8500rpm@28lbs.
Now on this setup its kinda a hybrid turbo, but even when I match the compressor size more to the turbine size the numbers on my turbo simulator show a higher PR % difference.

I have a lot more examples from customer cars with data logs showing more of the same.

To be fair, your example shows an engine with a very large RPM range, compared to how the actual F1 engines are used.
But 30%, PR of 3.5 to 2.5, I'd say that's pretty high in any context.

[GoranF1]

@Wazari,opinion?

http://thejudge13.com/2016/02/17/honda- ... -continue/

[Postmoe]

We keep on talking about the gross output of these PU's , but taking into account the complexity of the tuning they entrail, I bet the killer punch is within the software.

The more predictive and automatic, the better. I do not know how much telemetry and mapping are regulared but these engines ask for constant managing, it's in their essence. When managing how much do you harvest at any point within the track, without a powerful software, you'll end spending a lot of workforce, budget, time, in the same task your competitor spends a mouse click.

I mean, there's a lot to tweak: fuel, fuel flow, harvesting in it's various forms... How much grip, how cool is the air... And then more strategic layers, what do we want, how do we defend against X. All comes in the end on how much and how easily you can translate that into fine track-specific tuning.

As I see it, those factors are well known individually, but difficult to manage as an orchestra without a tool.

And this software must be fed with good data from experienced engineers. This new formula is certainly a challenge, no wonder Honda prefers keeping everything in house.

This is why I think Honda struggled so much in the beginning. It's not their prowess as engineers, it's perhaps they lacked a team dealing with the organizational engineering part.