Honda Power Unit Hardware & Software

[tom101]

Eric Boullier:

"So I think it's going to be complicated for Honda to understand what's wrong, it could maybe be coming from some batch problems. I don't believe it is a car build problem but we can't find what's wrong... We can do two laps or we can do two days [worth of running].

It is a little bit less hot today, by the way. Maybe these five or six degrees are good enough, but I don't know. It's a bit strange."

http://www.espn.co.uk/f1/story/_/id/192 ... ry-bahrain

[dren]

https://twitter.com/andrewbensonf1/stat ... 3068909568

Plus the time they are losing to the drivability issues with the upshifts. It sounds like some of this was sorted during the last test.

[godlameroso]

Ben'sonso' doesn't know what he's talking about. That would mean the Sauber is also down 120hp, which means the Mercedes engine has a fair bit over 1,050hp.

[Vortex37]

OK everyone I've been reading this forum for a long time but the fantasy in this thread made me finally register and post.

Turbos typically use 2 forms of bearings - Journal bearings or Ceramic Ball Bearings.
Any turbo worth its weight these days uses Ceramic bearings as they offer lower friction and thus less rotational drag on the turbo.
If you think any of the F1 turbos use anything but Ceramic bearings then you need to step away from the pipe.
A "Magnetic Bearing" would not be able to handle the axial thrust loads that a compressor wheel generates. If you can't visualize is a compressor wheel is like a screw trying to pull its way through the air - thus the "Thrust Bearing" that is needed on a turbo.

I'd also be willing to bet that the MGU-H problems Honda suffered the weekend were a seal failure - probably from a machining defect / tolerance problem. EB hinted at this with his remarks yesterday about it maybe being a "batch problem"

I'd also like to add that while it is frustrating to see where Honda are at I think most of you seriously underestimate the skill needed to even do as badly as they are right now. They will get there. They just need time.
And remember - just because you have a theory about how or why they are doing something doesn't mean you have a clue about any of it.

Modern turbos use ceramic hybrid bearings.

[Vortex37]

The more I think about this the more I think J.A.W. may have hit the nail on the head!
An EM bearing would explain why Honda was so confident of a significant step change in performance for the new PU, and it would also explain why (while the technology is in its infancy) they are having terrible trouble on the track that was hard to detect on the test-bed.

Is it really feasible that a marginal decrease in friction of the MGUH shaft due to the type of bearing causes a step change in whole PU performance. It doesn't seem realistic to me.

Research shows that ball bearings give a faster response time for spin up.

[shady]

OK everyone I've been reading this forum for a long time but the fantasy in this thread made me finally register and post.

Turbos typically use 2 forms of bearings - Journal bearings or Ceramic Ball Bearings.
Any turbo worth its weight these days uses Ceramic bearings as they offer lower friction and thus less rotational drag on the turbo.
If you think any of the F1 turbos use anything but Ceramic bearings then you need to step away from the pipe.
A "Magnetic Bearing" would not be able to handle the axial thrust loads that a compressor wheel generates. If you can't visualize is a compressor wheel is like a screw trying to pull its way through the air - thus the "Thrust Bearing" that is needed on a turbo.

I'd also be willing to bet that the MGU-H problems Honda suffered the weekend were a seal failure - probably from a machining defect / tolerance problem. EB hinted at this with his remarks yesterday about it maybe being a "batch problem"

I'd also like to add that while it is frustrating to see where Honda are at I think most of you seriously underestimate the skill needed to even do as badly as they are right now. They will get there. They just need time.
And remember - just because you have a theory about how or why they are doing something doesn't mean you have a clue about any of it.

If you watch the Siemens video on the EM bearing they specifically speak about axial thrust loads, and how its not affected the same as other EM bearings. I dont think this is what Honda is doing, but it would seem as though this particular weakness had been explicitly addressed.

Also I would have to assume they would have to work in concert (EMB/MGUh), as an MGUh is already EM it would seem natural for them to want to lower the coefficient of friction. The interference between the two doesnt seem compelling as a constraint as it would more than likely be a single integrated unit, plus you get damping from the EM bearing opposed to a hybrid ceramic.

Like i said i dont think they are doing this, its merely an exercise, though some of the shortcomings mentioned dont really lean towards it being infeasible.

[MrPotatoHead]

If you watch the Siemens video on the EM bearing they specifically speak about axial thrust loads, and how its not affected the same as other EM bearings. I dont think this is what Honda is doing, but it would seem as though this particular weakness had been explicitly addressed.

I'll watch it later when I get a chance - busy programming a 5-axis cnc machine right now

[MrPotatoHead]

Modern turbos use ceramic hybrid bearings.

Exactly. It's an angular contact ball bearing assembly.

[PlatinumZealot]

@ EM bearings topic..

New Rheometers in the food industries use Electromagnetric bearings for extremely low friction so that even the tiniest vsicosities can be measured. Their predecessors were air bearings. These EM bearings work great especially in the axial direction.. but there is a catch... The air bearing still has to be used for lateral stiffness. The position of the shaft is sensed in both the horizontal and axial position and EM controllers alter the current supplied to the actuators (basically electromagnets) to position the shaft axially and laterally.. It is all very complex.. and HEAVY for the forces applied.

[gruntguru]

https://www.youtube.com/watch?v=n5MMGZSi60Q

So conventional bearings have speed exclusion windows due to resonance, that have to be driven through quickly. Hmm, sounds like what we have already with the current PU setup. If you want to hold on to that idea of the em bearing, it could have been that Honda went conventional in the hope of integrating something like this later on, and ran into untuned harmonics.. I dont know. If power is lost, even for a moment that the MGUH would be the first thing to fail, but they would see that in the data.

Wazari did say they tried to stuff as many gidgets into the PU up the last minute, though this seems far fetched to me.

Interesting, in the video it says that there are speed exclusion windows around rotor resonances that must be driven through quickly or risk damaging. So perhaps they started with conventional bearings, and they've been working on MGU-H bearing design ever since? Since they consume electricity wouldn't it be counter productive to use EM bearings though?

That would be an interesting scenario given that compressor and turbine P.R. are both almost exclusively a function of rotor speed. ie restricting speed to a band above a resonance would also force PR to a band above particular values for T and C. At least on the compressor side, the manifold pressure can be regulated by throttling whereas the exhaust back pressure could only be reduced by opening the wastegate which is undesirable in most circumstances. (Although throttling is also undesirable for the same reason - reduction in overall efficiency)

[gruntguru]

OK everyone I've been reading this forum for a long time but the fantasy in this thread made me finally register and post.

Turbos typically use 2 forms of bearings - Journal bearings or Ceramic Ball Bearings.
Any turbo worth its weight these days uses Ceramic bearings as they offer lower friction and thus less rotational drag on the turbo.
If you think any of the F1 turbos use anything but Ceramic bearings then you need to step away from the pipe.
A "Magnetic Bearing" would not be able to handle the axial thrust loads that a compressor wheel generates. If you can't visualize is a compressor wheel is like a screw trying to pull its way through the air - thus the "Thrust Bearing" that is needed on a turbo.

Agree except to say the axial load on a centrifugal compressor is applied towards the turbo - likewise the turbine so these two thrusts oppose and cancel to some extent. This is due to the change in momentum of the gas - from axial to radial at the compressor and vice versa for the turbine.

Compressor thrust will reverse if the compressor is throttled upstream.

[MrPotatoHead]

Agree except to say the axial load on a centrifugal compressor is applied towards the turbo - likewise the turbine so these two thrusts oppose and cancel to some extent. This is due to the change in momentum of the gas - from axial to radial at the compressor and vice versa for the turbine.

Compressor thrust will reverse if the compressor is throttled upstream.

When the thrust bearing wears out in a journal bearing turbo it's usually always on the inside because of the larger thrust forces axially towards the compressor wheel.

[MrPotatoHead]

Agree except to say the axial load on a centrifugal compressor is applied towards the turbo - likewise the turbine so these two thrusts oppose and cancel to some extent. This is due to the change in momentum of the gas - from axial to radial at the compressor and vice versa for the turbine.

Compressor thrust will reverse if the compressor is throttled upstream.

Thrust forces visualized:

[godlameroso]

OK everyone I've been reading this forum for a long time but the fantasy in this thread made me finally register and post.

Turbos typically use 2 forms of bearings - Journal bearings or Ceramic Ball Bearings.
Any turbo worth its weight these days uses Ceramic bearings as they offer lower friction and thus less rotational drag on the turbo.
If you think any of the F1 turbos use anything but Ceramic bearings then you need to step away from the pipe.
A "Magnetic Bearing" would not be able to handle the axial thrust loads that a compressor wheel generates. If you can't visualize is a compressor wheel is like a screw trying to pull its way through the air - thus the "Thrust Bearing" that is needed on a turbo.

Agree except to say the axial load on a centrifugal compressor is applied towards the turbo - likewise the turbine so these two thrusts oppose and cancel to some extent. This is due to the change in momentum of the gas - from axial to radial at the compressor and vice versa for the turbine.

Compressor thrust will reverse if the compressor is throttled upstream.

Would throttling with a blow-off valve/bypass-valve be worth it?

[gruntguru]

Depends on what you are trying to achieve. If you want to reduce pressure and airflow from the compressor the most energy-efficient way is to reduce the shaft speed. The next best would be variable guide vanes. Next is inlet throttling. Last of all is blow-off valves - the compressor does the same work and much of that energy is dissipated through the blow-off valve.