Honda Power Unit Hardware & Software

[Thunder]

Alright Guys. This Thread gets a 24hr Timeout for cleaning up (again) and for everybody to think about what and what not to post here.

To quote myself from 5 Days ago:

Ooooook we all knew it was coming. There has been way to much OT chatter on this Subforum here lately (and especially in this Thread).

So from now on the PU Threads will be more closely moderated and i started with a quick and dirty clean up of this one.

Remember this Thread should focus on the mechanical Side of the PU (including speculation with a solid technical Background of Future Developments of course), NOTwhen Spec 4 will be released, who has the most dnf's, if RB helps TR with Integration, what Renault does or what PU TR will run in the Post Season Test....

There are Team Threads for such posts:
McLaren Honda Team Thread: https://www.f1technical.net/forum/viewt ... start=6705
Scuderia Toro Rosso Team Therad: https://www.f1technical.net/forum/viewt ... &start=240

Speed Trap figures (as a main Subject of the Post) also don't belong in here. There are however slight exceptions. If for instance someone was able to point out that Vandoorne and Alonso had different types of PU, but otherwise same set up and aerodynamic profiles, than that would be acceptable as we can scrap all significant variables in that case. Just don't Post "Verstappen 322, Vandoorne 310 "

And 2 Days ago:

MGU-H problems are related to operating temp issues.

Not sure if I am allowed to mention anything about the "Spec 4" PU, but the complete Spec 4 PU will not be run in any races this season. Some components may be installed in the current "Spec 3" platform but not the complete "Spec 4". Yes, there are Spec 4 PU's built and tested.

Of course you can, as long as there are some technical bits in your post about said Spec that make it worth discussing.

It's just the "hey, rumours are spec 4 will get introduced" Posts followed by 2 pages of Fun and Games about Honda that will get deleted.

To add to that from what i've seen on the last page. Points target for McLaren, Spec 5 jokes and what McLaren focuses on for the remainder of the Season are NOT to be discussed here.

Stay Technical

[turbof1]

Reopened. Please stay on topic. That means no gratuitous negativity, no discussion about PU deployment strategies and finally speculation is only allowed if staved by a solid technical background.

[FW17]

Does the MGU H need bearings?

[godlameroso]

Does the MGU H need bearings?

Internally yes however since the mgu-H is unlimited, they can use any material or technology so sky's the limit as far as bearing type.

I would wager that they use silver windings, and other neat tricks to boost efficiency. I would love to take an MGU-H apart and see what makes it tick. The types of magnets they use, etc. A motor so light and compact and capable of 100+ kW has to have some fancy internals water cooled or not.

[Tommy Cookers]

the fancy externals seem to a step ahead of anything else

varying voltage and frequency to keep at the load angle from instantaneous machine position at this stonking current and these 20000 - 120000 rpm
just another constituent functionality of the PU

[roon]

the fancy externals seem to a step ahead of anything else

varying voltage and frequency to keep at the load angle from instantaneous machine position at this stonking current and these 20000 - 120000 rpm
just another constituent functionality of the PU

Was there any comparable applications featuring such parameters prior to this?

[Tommy Cookers]

some time ago I checked and posted somewhere that it seemed clear .......
on offer even from specialist providers seems to be this rpm capability at lower power or this power capability at lower rpm
it's a power electronics issue - the ability of the transistors to switch well this current at this frequency

this seems to be the most useful of the sources
http://www.highspeedgenerator.com/
tbf they seem to have done one rather design that is rather outstanding at 100 kW/60000 rpm

another manufacturer highlights the use of magnetic bearings - presumably this is usual

[amho]

[...]
MGU-H bearing failure is a result of unforeseen multi-directional stress on the shaft and entire casing that houses the bearings.
[...]

Just a reminder of a previous post abut the MGU-H, to add to the latest.

Well spotted, I missed that!

The shaft ca be loaded in torsion, radially (bending) and axially (tension). Torsion does not affect bearings however both axial and radial loads are reacted by angular contact bearings. I suspect this is what Wazari means by multi directional stresses.

The axial loads are likely to be produced by the pressure acting on the compressor and turbine wheels, while radial loads are the result of unbalance, EMF and misalignment.

Another factor could be the mounting of the turbo and MGUH to the block. It is very bad practice to bridge the LH and RH banks in high performance Aluminium engines due to the high thermal expansion and Honda might have done just that.
In this case the housing will experience the loads reacted by the bearings as well as the thermal loads due to the block expansion.

you mentioned about mounting turbo & mgu-h in cylinder head block, it leads me to this imagination:
temperature gradient from turbine to compressor is very large so the mounting of bearing in turbine side has larger expansion than the mounting that houses compressor bearing which causes a bending load on shaft and extra radial load on bearings, in normal turbo configuration(non split turbo) there is only one mounting ( turbine flange) so another side is free to expand without extra load on bearing, but here in Honda engine there are two mounting with different expansion-strain that put extra stress on shaft and bearing.
I guess that another problem is that when u want to have turbo with long and narrow shaft u should watch for rigidity of shaft to avoid bending and vibration of shaft in critical shaft speed, to reach good rigidity u pick a bearing with tight internal clearance but the problem is that bearings with tight internal clearance are not suitable for large temperature change and when temperature goes high put excessive load on bearing.

[godlameroso]

I wonder if the mgu-H takes advantage of the max rotating speed allowed by the regulations, or if it's quite a bit lower? The compression ratio and max voltage allowed are unrealistic numbers as it is. So I wonder if actual mgu-H speed is lower as well.

[roon]

Another factor could be the mounting of the turbo and MGUH to the block. It is very bad practice to bridge the LH and RH banks in high performance Aluminium engines due to the high thermal expansion and Honda might have done just that.

Isn't this de facto the case for everyone? Each bank is rigidly attached to the chassis & gearbox via the heads and crankcase. Re: the MGUH installation, would there be anything against using compliant bushings or mounting methods to suspend it within the vee?

...failure is a result of unforeseen multi-directional stress...

Story of my life.

[HPD]

Hasegawa: "Coming to Mexico we expected to have limited power due to the high altitude, but I’m pleased to say the effect is much less than anticipated. I appreciate the efforts made by our colleagues back in the factory who have simulated various methods and found an effective way to minimise the power loss in the tough conditions.”

[PlatinumZealot]

Well spotted, I missed that!

The shaft ca be loaded in torsion, radially (bending) and axially (tension). Torsion does not affect bearings however both axial and radial loads are reacted by angular contact bearings. I suspect this is what Wazari means by multi directional stresses.

The axial loads are likely to be produced by the pressure acting on the compressor and turbine wheels, while radial loads are the result of unbalance, EMF and misalignment.

Another factor could be the mounting of the turbo and MGUH to the block. It is very bad practice to bridge the LH and RH banks in high performance Aluminium engines due to the high thermal expansion and Honda might have done just that.
In this case the housing will experience the loads reacted by the bearings as well as the thermal loads due to the block expansion.

Remember the chassis also bridges the lh and rh cylinder banks. They want that to be as stiff as possible in certain directions i supposed. Now if what you are saying is indeed a big factor... To play devil's advocate.. Suppose Mlaren Made the top engine mounting of the chassis too stiff in the transvers direction? That wounld be a McLaren caused problem now.

[PlatinumZealot]

Another factor could be the mounting of the turbo and MGUH to the block. It is very bad practice to bridge the LH and RH banks in high performance Aluminium engines due to the high thermal expansion and Honda might have done just that.

Isn't this de facto the case for everyone? Each bank is rigidly attached to the chassis & gearbox via the heads and crankcase. Re: the MGUH installation, would there be anything against using compliant bushings or mounting methods to suspend it within the vee?

...failure is a result of unforeseen multi-directional stress...

Story of my life.

Haha. Good one!

But marriage problems aside...

Andy Cowell notably said that the distribution of stresses through the engine block is the biggest factor on designing the split turbine housing. Did he say it took Mercedes maybe two years to solve or something?

[fellowhoodlums]

Did we ever find out what the 0.1 update gave us? New cylinder head was mentioned yet nothing official.

[63l8qrrfy6]

Their compressor can easily deliver in excess of the boost pressures necessary. In order to maximize the potential of the ERS it has to be this way.

I am sure it can at normal altitude, not too sure about 2000m though.

the fancy externals seem to a step ahead of anything else

varying voltage and frequency to keep at the load angle from instantaneous machine position at this stonking current and these 20000 - 120000 rpm
just another constituent functionality of the PU

Was there any comparable applications featuring such parameters prior to this?

Sure are:
https://integralp.com/integral-e-drive/

There are quite a few other examples, I though the IP products were a bit special as the company was founded by Keith Duckworth's son together with other ex Cosworth people. Unsurprisingly, a few of their engineers moved to Mercedes a long while ago.

Another factor could be the mounting of the turbo and MGUH to the block. It is very bad practice to bridge the LH and RH banks in high performance Aluminium engines due to the high thermal expansion and Honda might have done just that.

Isn't this de facto the case for everyone? Each bank is rigidly attached to the chassis & gearbox via the heads and crankcase. Re: the MGUH installation, would there be anything against using compliant bushings or mounting methods to suspend it within the vee?

The cam covers and blocks have big buff ribs to take whatever load goes through the mounting points. Similarly, I think that the chassis side mounting can be easily designed to accommodate the thermal loads.

The obvious problem with a very compliant turbo/MGUH mounting is low natural frequency (vibration) and turbo twist under boost. The reason I brought up mounting is because unlike Merc, where the turbo sits very low, the Honda turbo sits quite high up and would need big old girders to mount to the block (bottom of the vee). That's why I though it's possible they might have mounted it to the heads or somewhere up the banks where relative thermal deformation is high (the LH and RH banks tend to come apart at temperature).

Someone brought up the bumpy nature of the circuit - I don't think this was a factor.
The shocks produced by hitting kerbs are significantly higher than those produced by track protuberances. I don't know how harsh kerbs are on COTA, but I can't recall drivers complaining about them.