Turbocharger for a little F1 car

[lightsun]

I'm a student in the last year of automotive engineering, and this year I'm going to build a little F1 car with a group of 15 other students. It's an open-wheel car, with an engine of 120 HP, 300 Kg weight including driver to have a little idea of it. And what I'm most interested is in aerodynamics, but the car doesn't go enough fast to be a very rellevant thing, so I was thinking about developing a turbocharger, but I don't know with which programs can I develop this component, if it could be too much difficult for a student, and how many HP can I obtain with that turbocharger (or other advantatges). I remember that many years ago, when we had turbocharged engines in F1 the engine's HP were amazing, so maybe it could be a key component to beat other cars. Can anyone help me???Thanks!!!

[flynfrog]

why not build a FSAE car then you have a set of rules to build to

your question is way to openened to answer.

If you are asking if you could build your own turbo charger im going to say no It would take years just to develop the blade profiles let alone the manufacturing process. But you can buy one from a junk yard for the cost of a nice lunch $5USD

[lightsun]

In fact, we are going to take part in Formula Student 2009 competition .
A member of the theam which is going to take part in FStudent this year told me that the car can have a turbocharger, but only if a student develops it. So, my idea is to develop a basic turbocharger, very simple but effective. I could work on it full time during the next 10 months more or less, only designing it, and manufacturing it could be done for one of our collaborating companies. What do you think, is the project feasable?

My second question is, which programs may I use?? Design with Catia and simulation with ANSYS? or can I make the simulation with any module of Catia? Or I would have to use Fluent?

Finally, anyone has a little idea of how many HP could I obtain with this turbocharger? (The car has more or less 120 HP, and weights about 300Kg with the driver included).

[bazanaius]

If you could get 120hp out of an FSAE car you'd be laughing - to whack is about 102 at the moment.

Before you start thinking about FS, read the rules - http://students.sae.org/competitions/fo ... ies/rules/

Engines are restricted and there are specific rules on where the turbo can go. This leads to other complications that will affect your design. I'm pretty sure you dont have to design your own turbo, just the implementation of it.

B

[SZ]

It's an open-wheel car, with an engine of 120 HP, 300 Kg weight including driver to have a little idea of it.

I'd suggest you do some calculations involving fully choked flow through an ideal 20mm restrictor using best and worst case ambient conditions. If your maths and theory's done correctly you'll work out that anyone claiming 100+hp with an FSAE racer is:

- not running a standard iso-octane based fuel
- using smaller horses than normal in their power calculations
- (more likely) talking sh*t

Would also suggest peak power is not what the competition is about.

If you wanted to find out more about what kind of engine you're actually after (more to the point what kind of static/transient power/torque numbers you're chasing), I'd suggest a few people in your team write a vehicle simulator to start playing with as much.

Sounds hard, but easy to do in MATLAB, and it's been done many, many times before by many students before you. There's a wealth of papers, books, complete theses, even free codes out there to do as much.

You'll find it's not all about peak power or torque. Quite far removed from it, actually.

300kg with driver is an interesting idea if your lightest driver weighs 100kg+. With Newton's second law being fairly reliable and many of the cars you'll compete against being considerably lighter than what you're proposing, why aim to be second best? (If you're a first year team) some might pass that off as an excuse, but that's poor - many other teams in the long/considerable history of the competition have delivered well-executed, successful sub-200kg in a variety of materials from steel to exotica, complete with full-house 600cc engines. Study specifications and photos - if you want to be competitive from the word go, there's no reason you can't - reinventing the wheel won't help you though. Learn from history.

And what I'm most interested is in aerodynamics, but the car doesn't go enough fast to be a very rellevant thing,

Total crap. I'd fail any of my aero students for making such a remark - prove it so.

FSAE cars travel through a fluid medium - I'm yet to see one run in a vacuum - so fluid mechanics are important. Sure, there's better return to be had in other areas, but there's significant advantage to be had in aero.

I would argue in FSAE particularly there's some good advantages to be had in design, not because there's significant forces to be explored but because on the whole aerodynamics aspects are so poorly executed in the competition. Aero is more than downforce (though there's ways of getting that out) - many other aspects often get overlooked (for instance cooling, judging by the number of overheated entrants over the years).

Though it helps you don't need a wind tunnel or mega CFD either. Just good thinking and good, ingenious use of whatever resources you have at hand.

so I was thinking about developing a turbocharger, but I don't know with which programs can I develop this component, if it could be too much difficult for a student, and how many HP can I obtain with that turbocharger (or other advantatges). I remember that many years ago, when we had turbocharged engines in F1 the engine's HP were amazing, so maybe it could be a key component to beat other cars.

Don't know how you'd think developing your own turbocharger is going to be more rewarding for your efforts vs aerodynamics! Turbocharging isn't simple in practice (as in just matching a reciprocating and turbine system to give acceptable performance is a hard study in itself). The components are difficult to design, difficult to manufacture, are made to exacting tolerances, they undergo severe thermal loads... I can go on and on. You'd need an expert level of proficiency - and understanding - in design, manufacturing, fluid mechanics and thermodynamics to design one, let alone expertise in packages in each, let alone a very solid idea of what kind of engine it's being mated to and how you could best exploit it's performance with a turbocharger. This is why turbochargers aren't designed by single person enterprises, they're designed by teams of professionals.

You'd find it easier designing your own engine and using an off-the shelf turbocharger in such a project than in trying to build a turbo to suit an existing engine. In fact, the only relevant project in the field did just that.

Garrett/Honeywell and others make (and if they still do, it's been a while since I was a student) and may sponsor you with a turbo (or, as mentioned, a wrecker - they're auto-spec units) - if you need it.

But do you?

I remember that many years ago, when we had turbocharged engines in F1 the engine's HP were amazing, so maybe it could be a key component to beat other cars.

You don't beat other FSAE racers by having a turbo or going quicker. It's not a "best turbocharger" or "fastest car" competition. It's a competition wherein teams of students compete to best deliver a large scale project. (Say it over and over to yourself before it sinks in).

The most common mistake I see student teams make (and I've seen a few) is that the project is started with fixed ideals, fixed design solutions - "we must have a turbo!" "aero is ---!" etc.

If you're going to revolutionize anything, be one of the few teams that sets about doing the design process properly. Start with concepts. Ask yourselves questions. Don't stop until you have technical answers. Develop a specification for your car in terms of what it must do - not in terms of how you're going to do it - if you don't know where you're going, how are you ever going to get there? How much power and torque do you need (let alone how much can you have?) What other aspects of engine performance are important? Do you know? Sounds like no - get yourselves some concrete ideas before moving forward! You'll thank in the design review when you have to explain your choices, and throughout your year as your better defined ideas of "how" and "why" let you make real engineering gains a lot quicker. Many of your competitors (most FSAE teams) won't have this edge.

The competitive edge in any large-scale project isn't technical, it's in good management and what technical solutions such a process allow you.

[Conceptual]

To be honest, developing a turbocharger isnt really that hard, if you get away from the centrifugal compressor blades.

If you use 2 Tesla turbines on the same shaft, you can effectively tap the exhaust pipe, and feed it into the first half of the Tesla charger, and then the other half will spin, and suck in more air, and pressurize it. At 4" round, and spinning at 30K RPM, you can produce 30psi at the outlet.

The cool part of this is that the used exhaust is actually recirculated into the incoming aircharge, so you don't need an exhaust return. PM me if you are interested, since I have already sourced all of the materials needed to build this. The motor discs are made from teflon, and the compression discs are laminated mylar. I think the entire thing sourced at about $60USD.

The bad part is that it is my concept, and though I have done many hours of research, I haven't built one yet, so I cannot say that it would work perfectly. Unfortunately with a Tesla Turbine, there are several thousand combinations that you can explore, and I expect that what I have is pretty close, but may need further refinement after some testing.

If you are interested in my research, and want some drawings, PM me. It wouldn't cost you anything (except sourcing the parts), and I would just like to see how well it works.

Chris

[SZ]

To be honest, developing a turbocharger isnt really that hard, if you get away from the centrifugal compressor blades.

If you use 2 Tesla turbines on the same shaft, you can effectively tap the exhaust pipe, and feed it into the first half of the Tesla charger, and then the other half will spin, and suck in more air, and pressurize it. At 4" round, and spinning at 30K RPM, you can produce 30psi at the outlet.

The cool part of this is that the used exhaust is actually recirculated into the incoming aircharge, so you don't need an exhaust return. PM me if you are interested, since I have already sourced all of the materials needed to build this. The motor discs are made from teflon, and the compression discs are laminated mylar. I think the entire thing sourced at about $60USD.

The bad part is that it is my concept, and though I have done many hours of research, I haven't built one yet, so I cannot say that it would work perfectly. Unfortunately with a Tesla Turbine, there are several thousand combinations that you can explore, and I expect that what I have is pretty close, but may need further refinement after some testing.

If you are interested in my research, and want some drawings, PM me. It wouldn't cost you anything (except sourcing the parts), and I would just like to see how well it works.

Chris

Tolerances in a Tesla setup are critical. This and actual manufacture are a significant hurdle for a university workshop, let alone a university student.

The bigger challenge in a turbo setup don't concern the turbocharger, it concerns matching the turbine to the reciprocating system. It's a huge challenge - more than enough for a few university students to achieve on a concept as simple as an FSAE racer. I'd suggest the lack of data in this area would render it a unsuitable for an FSAE project - there's more than enough to do in calibrating an NA engine, let alone a turbo, let alone a turbo that uses a turbo whose performance characteristics are unknown and unique from common literature.

Doesn't answer the bigger question though - do you need it to be competitive? Qualitative evidence (successes in the competition) would indicate otherwise; plenty of quantitative sources too. Best thing at this stage would be to start simulating requirements developing an engine spec and then seeing what kind of powertrains exist that will meet those requiremnts - turbo or otherwise.

Your choice of materials is interesting... how structurally sound's the mylar at 30k? The metallurgy for an automotive Tesla on the turbine side wouldn't be something to sneeze at. Maybe a tentative approach would be to keep a conventional compressor? I remember a few years ago someone was keen to try something similar, I don't have the link handy or any idea if it went ahead.

But for a student project you'd need to be able to justify why you're using a Tesla setup over a conventional centrifugal setup, which involves knowing a good deal about both to start with and justify the work, etc etc...

Don't get me wrong Chris, Tesla was a genius (a special one at that) and the pump is technically very sweet.

I'd be keen to see a double pump setup for turbocharging use - would imagine such a design could be made to be scaled relatively easily for various mass flow rates...

[Conceptual]

To be honest, developing a turbocharger isnt really that hard, if you get away from the centrifugal compressor blades.

If you use 2 Tesla turbines on the same shaft, you can effectively tap the exhaust pipe, and feed it into the first half of the Tesla charger, and then the other half will spin, and suck in more air, and pressurize it. At 4" round, and spinning at 30K RPM, you can produce 30psi at the outlet.

The cool part of this is that the used exhaust is actually recirculated into the incoming aircharge, so you don't need an exhaust return. PM me if you are interested, since I have already sourced all of the materials needed to build this. The motor discs are made from teflon, and the compression discs are laminated mylar. I think the entire thing sourced at about $60USD.

The bad part is that it is my concept, and though I have done many hours of research, I haven't built one yet, so I cannot say that it would work perfectly. Unfortunately with a Tesla Turbine, there are several thousand combinations that you can explore, and I expect that what I have is pretty close, but may need further refinement after some testing.

If you are interested in my research, and want some drawings, PM me. It wouldn't cost you anything (except sourcing the parts), and I would just like to see how well it works.

Chris

Tolerances in a Tesla setup are critical. This and actual manufacture are a significant hurdle for a university workshop, let alone a university student.

The bigger challenge in a turbo setup don't concern the turbocharger, it concerns matching the turbine to the reciprocating system. It's a huge challenge - more than enough for a few university students to achieve on a concept as simple as an FSAE racer. I'd suggest the lack of data in this area would render it a unsuitable for an FSAE project - there's more than enough to do in calibrating an NA engine, let alone a turbo, let alone a turbo that uses a turbo whose performance characteristics are unknown and unique from common literature.

Doesn't answer the bigger question though - do you need it to be competitive? Qualitative evidence (successes in the competition) would indicate otherwise; plenty of quantitative sources too. Best thing at this stage would be to start simulating requirements developing an engine spec and then seeing what kind of powertrains exist that will meet those requiremnts - turbo or otherwise.

Your choice of materials is interesting... how structurally sound's the mylar at 30k? The metallurgy for an automotive Tesla on the turbine side wouldn't be something to sneeze at. Maybe a tentative approach would be to keep a conventional compressor? I remember a few years ago someone was keen to try something similar, I don't have the link handy or any idea if it went ahead.

But for a student project you'd need to be able to justify why you're using a Tesla setup over a conventional centrifugal setup, which involves knowing a good deal about both to start with and justify the work, etc etc...

Don't get me wrong Chris, Tesla was a genius (a special one at that) and the pump is technically very sweet.

I'd be keen to see a double pump setup for turbocharging use - would imagine such a design could be made to be scaled relatively easily for various mass flow rates...

I have about 130 hours invested in the Tesla Turbine setup, and about 60 emails from sourcing companies that claim that the laminated Mylar discs (actually they are 4" OD washers with a 2" bore. I have been told by the manufacturer of the Mylar and Teflon "washers" that they will withstand 150K+ RPM without coming apart. And as long as you drill the mounting holes through the entire stack at the same time, and have the mylar/teflon small washers to go between the discs, you will maintain disc spacing tolerances.

I am working on something similar that I hope to release as a new product line when my acquisition of a certain aftermarket performance manufacturer is completed.

So, I will be sure to video document the progress, and we will see what happens.

And, back to the topic: Why would you have to explain ANYTHING as to your choice of turbo setup? If simplicity is not a good enough answer, then I think that I wouldn't even bother.

Chris

[SZ]

I have about 130 hours invested in the Tesla Turbine setup, and about 60 emails from sourcing companies that claim that the laminated Mylar discs (actually they are 4" OD washers with a 2" bore. I have been told by the manufacturer of the Mylar and Teflon "washers" that they will withstand 150K+ RPM without coming apart. And as long as you drill the mounting holes through the entire stack at the same time, and have the mylar/teflon small washers to go between the discs, you will maintain disc spacing tolerances.

I am working on something similar that I hope to release as a new product line when my acquisition of a certain aftermarket performance manufacturer is completed.

So, I will be sure to video document the progress, and we will see what happens.

How do balance and rigidity stack up at 150kRPM after you've drilled holes through these things?

Still think it's a good idea worth exploring though. I understand Tesla himself failed to popularize it though shortcomings in contemporary material processes.

And, back to the topic: Why would you have to explain ANYTHING as to your choice of turbo setup? If simplicity is not a good enough answer, then I think that I wouldn't even bother.

Simple answers to complicated problems - if shortsighted - don't win the students any prizes.

Having a turbocharger that's based around simple concepts is one thing - having the processes to develop, build, test and design it is another, as is the ability to do all that and then justify it in a more holistic context: Did it cost less? Does it weigh less? Is there a tangible performance benefit? Did you find building and designing such a device simpler than purchasing an existing unit off the shelf? What was your development plan for the unit - what where the variables you explored, the test processes, the parameters important to you, etc? And on and on. What about the implications in the wider context of the powertrain: Does is offer better thermofluid/matching characteristics with the engine? In the wider context of the project: how do you justify devoting resources to (and accepting significant risk in) delivering this aspect of your vehicle as an original solution to a problem peripheral to the main thrust of the project?

It's not a "best turbocharger concept" competition. I don't doubt the concept wouldn't be a contender in one, to be fair. I mean no sarcasm in that - quite sincerely, it's a good very idea, and if this student likes turbocharging more than he likes FSAE cars, it'd be a great project worth investigating further that could very easily bring together some solid fundamental knowledge and lead to either further study, a highly employable graduate or both.

But it's barely a competition where turbocharging is required. To be crude under average conditions it's extremely difficult to get enough air through the restrictor to generate 90hp (if that, realistically); there are many NA powerplants allowable that will deliver as much without a turbocharger. Aspects giving rise to "driveability" are more important in such an engine anyway; whilst there are certainly advantages in turbocharging, developing a system and calibrating it such that a turbocharged engine is "racecar driveable" is a very serious task. I've seen very good students spend a year on solid projects in developing better calibrations for a single cylinder NA engine for the same competition and getting serious results out of as much - multicylinder, turbocharging, Tesla turbos they're all good ideas with their own potential advantages, but they add significant complexity, cost, other resources and risk (and that's just the powertrain, let alone the whole car!)

While fear of complexity certainly isn't a reason alone to say "don't do it", in any real-world engineering enterprise when time and money are at stake there's a reasoned pause wherein your team looks at things as says "should we assume this task - do we really need it to get where we want to go" and completes a solid period of structured research to effect as much.

Far from simple though easy to follow if you understand the rationale.

[Conceptual]

When I source these items, I may just order 50 sets of parts at the same time, so that way I have extra parts to play around with the original prototype, and if it delivers the desired results, I will donate one to the team of your choice for testing...

Thanks!

Chris

[SZ]

When I source these items, I may just order 50 sets of parts at the same time, so that way I have extra parts to play around with the original prototype, and if it delivers the desired results, I will donate one to the team of your choice for testing...

Thanks!

Chris

A shame I'm leaving University to head to industry, or I'd take you up on that and gladly look into finding a few good students, an engine and a dyno to make use of it.

[lightsun]

Maybe the idea of developing a turbocharger is not the best... To be honest, I really like aerodynamics, but I'm a student of Mechanical engineering, not aeronautical, so I have studied aerodynamics theorically, but I haven't studied any programs of CFD. I have studied Catia and Ansys, but I don't know their possibilities in CFD, I don't know if I could use them to study the aerodynamics or if I will have to learn Fluent or something like this. Any suggestions? Thank you for your answers!

[Conceptual]

When I source these items, I may just order 50 sets of parts at the same time, so that way I have extra parts to play around with the original prototype, and if it delivers the desired results, I will donate one to the team of your choice for testing...

Thanks!

Chris

A shame I'm leaving University to head to industry, or I'd take you up on that and gladly look into finding a few good students, an engine and a dyno to make use of it.

I spoke to the manufacturer again today, and he once again assured me that his products meet my needs.

This will take shape by the end of summer, so maybe you can set up some students with it to play with after you are gone?

Chris

[SZ]

Maybe the idea of developing a turbocharger is not the best... To be honest, I really like aerodynamics, but I'm a student of Mechanical engineering, not aeronautical, so I have studied aerodynamics theorically, but I haven't studied any programs of CFD. I have studied Catia and Ansys, but I don't know their possibilities in CFD, I don't know if I could use them to study the aerodynamics or if I will have to learn Fluent or something like this. Any suggestions? Thank you for your answers!

Read the posts carefully. You need specialists in fluid mechanics, materials, design and manufacturing to develop a turbo. Catia and Ansys is the tip of the iceberg. You'll need proficiency in CFD and FEA/FEM at a level most in industry won't have until they're using it all day, every day, for three years at the least with proper industrial training (not what you do at uni as a ugrad). You'll need experience in metallurgy and relevant manufacturing processes. You'll need access to those manufacturing processes. You'll need to have access to reasonably accurate test benches to map and further develop the turbocharger, which means a top understanding in fluids, $$$ in test equipment, more time and $$$ in developing the thing (you won't be building it once), a spares budget and a program to build various units, life estimation calcs, etc etc

Let me put it in perspective for you:

The only student that ever built a turbocharged FSAE engine from scratch... didn't build the turbocharger - too complicated, no rewards.

Or a bigger perspective:

Of all the teams to use turbocharging in F1, not one ever built the turbo themselves. Too much of a specialist area, there are suppliers that'll do that better than you ever could.

What you want to be asking yourself now is:

"What kind of engine do we need for this project"

What you don't want to be telling yourself now is:

"We need to build a turbocharger for this project"

As you seem to have no idea why. Which is fine! Much of the challenge in FSAE is finding out why you'll make the design choices you'll choose. So go research. Read my earlier posts in detail - don't skip the initial work just to get to something you think (now) is more exciting - there's nothing more exciting than turning up to comp with a car who's performance you understand and are confident in. There's nothing more demeaning that doing so with something that doesn't work, is unreliable and that you have little understanding of. Trust me, I've seen both.

If you ever get to the stage where you think you'll need a turbocharger, I suggest you read "Turbocharging the Internal Combustion Engine by N. Watson".

[SZ]

Maybe the idea of developing a turbocharger is not the best... To be honest, I really like aerodynamics, but I'm a student of Mechanical engineering, not aeronautical, so I have studied aerodynamics theorically, but I haven't studied any programs of CFD.

And FWIW;

Plenty of mechanical engineering graduates go on to successful careers in aerospace (even motorports aerodynamics).

CFD is not a requirement - that you can drive a CFD package means that, well, you can drive a CFD package. And nothing more. Possibly less actually - I'm yet to meet an undergraduate student that could use CFD to a level good enough to model turbocharger flows with any relevant accuracy. That's not because undergrads are stupid - more because the problem is very, very complicated, and the computational resources you'd need are extreme.

Skill in aero is fundamentally in understanding flow mechanics - knowing the theory. There's no smoking gun - what tools you use to deconstruct this (experimental, computational, whatever) don't help or hinder you nearly as much as your understanding of the work you're trying to complete and what goals you're trying to achieve therein.

Forget learning packages to give you answers. It's not what they're there for anyway.

Get out a notebook and pencil and start developing your understanding - start with questions. Most of a competitive FSAE racer really can be designed "on the back of an envelope" - many cars at comp would be better if they were!

If you're keen to learn any package seriously, start with those that automate mundane tasks - get a handle on MATLAB or Octave.