Renault Power Unit Hardware & Software

[noname]

Rotational energy dissipated in the impact is 0.5 x inertia x shaft speed ^ 2. Where is the tip speed in that ?

Knowing tip speed can help you assess the risk.

2 turbos, 50 and 100mm wheels, 100.000 RPM. How much margin (to burst) do you have with each of them?

[63l8qrrfy6]

Rotational energy dissipated in the impact is 0.5 x inertia x shaft speed ^ 2. Where is the tip speed in that ?

Knowing tip speed can help you assess the risk.

2 turbos, 50 and 100mm wheels, 100.000 RPM. How much margin (to burst) do you have with each of them?

Both hoop and radial stresses are proportional to the product of the squares of angular velocity and disc radius. The maximum stress is 4 times larger in the 100mm wheel. Of course, mathematically you can substitute in tip speed but my point is that you can describe any mechanical phenomenon without ever having to invoke this derived quantity.

As a counter example, if you have 2 overhanging rotors of equal diameters, one of which has 2 times higher unbalance than the other, tip speed is equal in both cases yet shaft and bearing loads are clearly higher for the rotor with the larger unbalance.
Similarly, 2 wheels of equal diameters and different trim will have different natural frequencies. Would you plot the Campbell diagram with respect to shaft speed or tip speed ?

The maximum speed of modern turbochargers is governed by bearing durability in the vast majority of cases and F1 is no different as shown by the number of bearing failures in the current engine era. Other than aerodynamic considerations, there is no reason to bring up tip speed.

As a side point, there is a very good explanation why centrifugal loads are easily dealt with in radial compressor and turbine wheels. In a rotating thin disk the maximum hoop and radial stresses are equal and occur at the centre of the disk. Fortunately the hubs in radial compressor and turbine rotors have large blend radii to the back disks which mean that the blade roots tend to be far from the high stress area. In addition, the static pressure on the blade is higher towards the tip which mean that the root stress due to pressure is much lower towards the centre.
Of course, wheels with high backsweep angles are slightly more problematic but they still benefit from the same favorable conditions.
Finally, the centrifugal stresses have relatively higher amplitude but much lower frequency compared to dynamic stresses, therefore they contribute less fatigue damage over a typical duty cycle.

[MrPotatoHead]

Rotational energy dissipated in the impact is 0.5 x inertia x shaft speed ^ 2. Where is the tip speed in that ?

Knowing tip speed can help you assess the risk.

2 turbos, 50 and 100mm wheels, 100.000 RPM. How much margin (to burst) do you have with each of them?

The material of the turbine wheel could be very critical here.
The Titanium Aluminide turbine wheel on the Borg Warner EFR turbos have some limitations when it comes to rotational speed. They reduced the size of the turbine wheel on the 9180 turbo to create the 9174 turbo because of exploding turbine wheels.

[noname]

(...)

All you said is absolutely right. Maybe except of what I would call underestimating aerodynamics; aero-induced instabilities could have nasty effects, but could be I am just too picky. That being said you would have hard time finding product engineer working on automotive turbochargers being as meticulous as you. Or, maybe, I was not lucky enough to came across too many of them. Their development process is far simpler and can be summarized as “it was always done like this”. Turbochargers those days became almost a commodity, and so I understand why people are not too creative. Price is the king.

Wheels sizing, or to be more precise compressor wheel as it is being selected first, is limited by max tip speed. There are no written rules defining it strictly, but people are hard-wired to certain number and can freak out if you will tell them you can safely go above.

There are good and legit reasons why the limit is set where it is, but this knowledge is not common.

To make the long story short, process of developing automotive turbochargers can be quickly described as follow: select compressor wheel (bearing in mind max tip speed can not be higher then …), select turbine wheel (its size is around certain % of compressor wheel diameter), select shaft diameter (and thus bearing size). LCF and HCF are usually covered by checking if blades’ 1st natural frequency is higher than x*speed. It is good first approximation, however it should be followed by further evaluation as fatigue is more complicated than this. Quite often they’re finding this when hardware testing starts.

[noname]

The material of the turbine wheel could be very critical here.
The Titanium Aluminide turbine wheel on the Borg Warner EFR turbos have some limitations when it comes to rotational speed. They reduced the size of the turbine wheel on the 9180 turbo to create the 9174 turbo because of exploding turbine wheels.

I guess they just changed material without touching the design. You can achieve the same speed with Titanium Aluminide turbine wheel, but you have to modify design slightly.

[PlatinumZealot]

(...) the point of the conversation was the effect of turbo speed on reliability. In that respect tip speed is almost meaningless (...)

"Almost" makes a difference
https://www.youtube.com/watch?v=7Y01Ed4Sg3U

Rotational energy dissipated in the impact is 0.5 x inertia x shaft speed ^ 2. Where is the tip speed in that ?

There are compression effects at the tip of the compressor wheel. You know, the speed of sound and all that.

[63l8qrrfy6]

(...)

All you said is absolutely right. Maybe except of what I would call underestimating aerodynamics; aero-induced instabilities could have nasty effects, but could be I am just too picky. That being said you would have hard time finding product engineer working on automotive turbochargers being as meticulous as you. Or, maybe, I was not lucky enough to came across too many of them. Their development process is far simpler and can be summarized as “it was always done like this”. Turbochargers those days became almost a commodity, and so I understand why people are not too creative. Price is the king.

Wheels sizing, or to be more precise compressor wheel as it is being selected first, is limited by max tip speed. There are no written rules defining it strictly, but people are hard-wired to certain number and can freak out if you will tell them you can safely go above.

There are good and legit reasons why the limit is set where it is, but this knowledge is not common.

To make the long story short, process of developing automotive turbochargers can be quickly described as follow: select compressor wheel (bearing in mind max tip speed can not be higher then …), select turbine wheel (its size is around certain % of compressor wheel diameter), select shaft diameter (and thus bearing size). LCF and HCF are usually covered by checking if blades’ 1st natural frequency is higher than x*speed. It is good first approximation, however it should be followed by further evaluation as fatigue is more complicated than this. Quite often they’re finding this when hardware testing starts.

Well I admit that I lack a good understanding of aerodynamics - my background is limited to mechanical design and numerical/computational methods. However I know that the only aerodynamic instabilities that are accounted for in turbocharger rotor-bearing system design are the cross-coupling effects due to variations in wheel - housing clearance. Neither of the two most popular formulations - the Alford and the Wachel formulae - explicitly relate the cross-coupling stiffness to the wheel tip velocity. Even more bizarre, the Alford formula does not include a velocity term at all.
The logarithmic decrement and stability threshold are again routinely plotted against shaft speed rather than tip speed.

Unlike HCF, LCF has nothing to do with the natural frequency. Typically in a turbocharger context the 'low cycles' refer to thermal cycles but can be expanded to include high loads that occur rarely such as violent kerb strikes.
It is no longer a sufficient condition for the 1st natural frequency to be higher than the synchronous speed as twin entry turbines for example produce strong 2nd order excitations. When a motor is coupled to the turbo, there will be an additional super-synchronous excitation based on the number of poles.

And now a few lines on one of my biggest bugbears:
I have noticed that the competence of engineers has been steadily declining and I am convinced that the fault lies with large companies that insist on tightly controlling every single process with rigid standards, practices, etc. The engineer becomes very proficient at navigating through mountains of documents that dictate in the greatest detail the size, aspect and characteristic of each component and in doing so loses his/her capability of deciding when and how to apply engineering principles.
Nowadays companies such as Ford, Renault, JLR, etc have become incapable of designing a complete engine in-house,yet they each employ hundreds if not thousands of so-called powertrain engineers. And to think that half a century ago a handful of Cosworth engineers designed an engine that would dominate F1 for over a decade using nothing more than good old drawing boards..

[MrPotatoHead]

The material of the turbine wheel could be very critical here.
The Titanium Aluminide turbine wheel on the Borg Warner EFR turbos have some limitations when it comes to rotational speed. They reduced the size of the turbine wheel on the 9180 turbo to create the 9174 turbo because of exploding turbine wheels.

I guess they just changed material without touching the design. You can achieve the same speed with Titanium Aluminide turbine wheel, but you have to modify design slightly.

I don't think you understand what I was saying. Borg Warner physically changed the size of the turbine wheel making it smaller to lower the tip speed of the turbine for the same rpm at the compressor side.
This way they could achieve the rpm needed on the compressor side without the turbine exploding.
This is why they released the 9174. 74mm Exducer on the turbine vs the 80mm on the 9180.

[63l8qrrfy6]

"Almost" makes a difference
https://www.youtube.com/watch?v=7Y01Ed4Sg3U

Rotational energy dissipated in the impact is 0.5 x inertia x shaft speed ^ 2. Where is the tip speed in that ?

There are compression effects at the tip of the compressor wheel. You know, the speed of sound and all that.

The static pressure towards the blade tip is a tad over half the pressure at the diffuser outlet. In terms of the stress it produces, that is not very much at all, especially given that the centre of pressure is well away from the tip.

[noname]

(...) And now a few lines on one of my biggest bugbears: (...)

I wish this problem would be limited to automotive.

The strength of people like the ones you've mentioned was coming from really good understanding of what they are doing. Common answer you can get nowadays is modern machines are far more complex and sophisticated, and thus much more efforts and people are required. That to some extend is right, although I see the biggest difference in a fact todays' engineers are given cookbook and are being discouraged from having doubts and asking too many questions. Accountants are happy as you can use less skilled (read: cheaper) workforce, however neither development cost nor time went down (even with the advent of digital tools). Instead we're seeing refinements and growing defragmentation, our ability to innovate was greatly diminished and we're flirting dangerously close with critical complexity.

[noname]

I don't think you understand what I was saying. (...)

Maybe I was not clear enough.

Assuming base turbocharger was well matched this change could negatively impact turbine performance. It would be avoided, or change would be greatly reduced, if different material properties of Titanium Aluminide were accounted for during development process. Changes would be subtle and efforts almost meaningless in the grand scheme of things. I suppose process could be faster and cheaper if common sende would be applied before failures happened.

I guess they opted to change material spec on existing drawings instead.

[PlatinumZealot]

Horner said Renault is 25bhp to 30bhp down in engine power. Didn't say if it was race or Qually mode though.

[dren]

And now a few lines on one of my biggest bugbears:
I have noticed that the competence of engineers has been steadily declining and I am convinced that the fault lies with large companies that insist on tightly controlling every single process with rigid standards, practices, etc. The engineer becomes very proficient at navigating through mountains of documents that dictate in the greatest detail the size, aspect and characteristic of each component and in doing so loses his/her capability of deciding when and how to apply engineering principles.
Nowadays companies such as Ford, Renault, JLR, etc have become incapable of designing a complete engine in-house,yet they each employ hundreds if not thousands of so-called powertrain engineers. And to think that half a century ago a handful of Cosworth engineers designed an engine that would dominate F1 for over a decade using nothing more than good old drawing boards..

I am, well was, a mechanical engineer in the power generation industry; in management now. I see the exact same thing. It's not just in engineering, either. I see equipment operators with zero knowledge or understanding of the equipment they are operating; they are just following a procedure. We've dumbed things down to a point that nobody has to think. If somebody wrecks a piece of equipment it is never their fault, it's because we don't have a procedure. Some people think cleaning 550v motors, that are in service, with a water hose is a good idea.

Back to the PU...I'd expect the rotational speed, and especially tip speed, to fall into the mechanical design somehow. Well, I'd guess it'd be more inertia related than tip speed related. Surprising you only worry about shaft speed.

[Sieper]

but then again, plonk an engine "that dominated for a decade" in a current car (for arguments sake let's disregard it wouldn't meet current regulations) and see the car struggle. The current level of integrations of different specialties is just on such a different level. I do agree with your point that the creative power of a small team is lost, but how can you still meet the requirements of a designing a current gen engine with a small team, it is just undoable. If you could find a way, have such a great team then yes, you could still gain a great advantage by harnassing their skills.

In my own field of expertise (ERP Software) we are also way past this point, you can only know/understand a small fraction of the workings.

[NSBiker]

In a recent interview, Christian Horner mentioned that the Renault engine was effectively a diesel.
The implication would be that the compression ratio is in the range of a compression ignition engine.
Questions ....
Are they running the turbo system hard enough to be able to get the compression ratio that high.?
I would expect it is capable of compression induced ignition, but a spark system would be needed for timing control.... or would it.?
Could they get the clearance volume low enough to make this work without the turbo.? I doubt it.
If they are relying on the ultra high CR (for a spark ignition engine) is this part of he posturing that is going on relating to the elimination of the MGUH for the new engine formula.?
With the supplementary electric drive for the turbo, there are engine mapping and control options that would not otherwise be available.