Average air consumption of an engine.

[Belatti]

Particularly that bit about a good operator... 120% VE for a NA street engine at 7kRPM? You're beyond a bit optimistic - I'd revise your understanding of VE.

How do you know its a street engine?

Come one men! the guy asked how to estimate it:

Average air consumption...
Any info, URL, Calculators, formulae or comments are very welcome!

I gave him an answer wich he doesn´t even have to get dirty hands, the guy is not a F1 engineer.

So:
1. model the engine
2. estimate VE
3. calculate air cons (estimated)

No need to install or buy a sensor

[SZ]

Particularly that bit about a good operator... 120% VE for a NA street engine at 7kRPM? You're beyond a bit optimistic - I'd revise your understanding of VE.

How do you know its a street engine?

Come one men! the guy asked how to estimate it:

Average air consumption...
Any info, URL, Calculators, formulae or comments are very welcome!

I gave him an answer wich he doesn´t even have to get dirty hands, the guy is not a F1 engineer.

So:
1. model the engine
2. estimate VE
3. calculate air cons (estimated)

No need to install or buy a sensor

Calculators are great if you know how to use them.

120% VE is wildly optimistic for anything but a hard-tuned race engine.

No point in exacting numbers if they're not accurate.

[Belatti]

Calculators are great if you know how to use them.

120% VE is wildly optimistic for anything but a hard-tuned race engine.

No point in exacting numbers if they're not accurate.

Well, I showed this model as an example, it is a hard-tuned race engine, a V10 to be precise =D>



Numbers have been accurate in this case

[SZ]

And this proves what?

Showing it CAN be done via 1D gas analysis is a totally different thing to showing HOW it's done.

Your example citing 120% VE is fraught with assumptions you seem not to understand. This makes it very pretty but ultimately useless unless your assumptions fluke every critical aspect of the practical engine it simulates. 120% VE? for a street engine your numbers are up to 40% off. For a race engine they're still optimistic. There's a good chance your V10 example wouldn't hit that much.

I'm not saying your approach is fundamentally flawed; 1D gas analysis software is used in industry every day to simulate various aspects of engine performance for research and development purposes. I've used 1D gas analysis myself (LES, WAVE, GT-power) and they're all great if you know how to setup a simulation properly - but that's far from simple in a race engine and very complicated in a street engine.

And even then they're just that - a simulator, not the calculator the thread originally asked for.

You might find some well-thought-out equations suitable for use in a calculator in Heywood (Internal Combustion Engine Fundamentals), though it's been a while since I've looked.

And PS:

What are you trying to show by the top graph? Sure this isn't more useless info?

The second one is easy to understand but the first is a little nondescript and possibly irrelevant - you're showing pressure variations within <15% of atmospheric in what? An intake runner, and airbox, what? Aside from the initial bit being totally useless information (flat line, atmospheric, nothing is happening) - your pressure variations under power in any duct of significance are way more than that, which is where you're really likely to test how accurate your simulation is.

[Conceptual]

SZ,

I appreciate your driving this way beyond anything that is actually necessary. I am perfectly happy with the estimate of 400CFM that has been shown not only here, but cross refereced in other forums.

I appreciate everyones contributions, especially Belatti's since it was the simplest and most direct way to figure out what I was after.

SZ, If I ever have a need to use a car engine as a life support air pump, and need to know how many RPM I need to maintain to allow my family to breathe, then I will be sure to contract you, because I'm sure that you would take it right to the absolute.

But this was simply to see what the round about number in CFM is used under normal aspiration, and to extrapolate turbo boost usage estimates.

Thank you all, I have my answer, and you can stop driving it to the quantum level.

Thanks!!

Chris

[SZ]

Chris;

The point is to have reasoned, educated estimates.

Air demand/consumption for IC engines is a question that gets thrown around a lot in engineering classes... as there's no easy answer to work it out, and a good answer requires good thought - and in that, an appreciation of what's at play.

I'd certainly not pass a student on an answer given on basis of it appearing on various internet forums. Nor one providing an answer that, while exacting, has a significant degree of error associated that's poorly understood.

No "take it right to the absolute" bullshit - just apply basic theory that you understand. Belatti does put forward a very well known and very correct approach - ideal air pump times volumetric efficiency - but shows a poor understanding of just what kind of VE numbers you could expect to apply and why... rendering the method useless (which isn't to take away that his theory wasn't solid up to that point).

"to extrapolate turbo boost usage estimates" - would be interested to see how you calculated this much - there's a few approaches I'd use. A crude calculation that'd be quite reliable to scale the difference to an NA engine has already been given here. If you want to make a decent calculator, you can go a little deeper.

[Conceptual]

Chris;

The point is to have reasoned, educated estimates.

Air demand/consumption for IC engines is a question that gets thrown around a lot in engineering classes... as there's no easy answer to work it out, and a good answer requires good thought - and in that, an appreciation of what's at play.

I'd certainly not pass a student on an answer given on basis of it appearing on various internet forums. Nor one providing an answer that, while exacting, has a significant degree of error associated that's poorly understood.

No "take it right to the absolute" bullshit - just apply basic theory that you understand. Belatti does put forward a very well known and very correct approach - ideal air pump times volumetric efficiency - but shows a poor understanding of just what kind of VE numbers you could expect to apply and why... rendering the method useless (which isn't to take away that his theory wasn't solid up to that point).

"to extrapolate turbo boost usage estimates" - would be interested to see how you calculated this much - there's a few approaches I'd use. A crude calculation that'd be quite reliable to scale the difference to an NA engine has already been given here. If you want to make a decent calculator, you can go a little deeper.

Please don't tell me what the reason for my post was. I would call you presumptious for such a thing.

And you are correct, you can make the perfect calculator, and heck, with all of the argument you have presented on this board, you could have made a spreadsheet calculator and posted it to prove your point.

I was looking at how much air (ballpark) an engine used, and a simple way to calculate that estimate. I am happy with the results I got, and changed my design direction accordingly. I dont have the capacity to directly feed the engine with my device, but I can control 10%, and to me, that is really all I need to know.

Thanks for your time, but this thread, and its arguments are unnecessary.

I have my answer, and can care less about the method to compute the exact figures.

Thanks!

Chris

[Ciro Pabón]

Well, Chris, in the same spirit, you can care about what you wish, we can also do the same, don't you think?

I also think that my point is not to make an educated guess. If you have the understanding SZ asks from his students, you should be able to build a model.

The model can be simple, as the volumetric one I started with.

It can be a little more complicated, like the stoichiometric one that finds the amount of gas used and deduce the oxygen needed for reaction.

It can be even more complicated, like the one SZ proposed, which deduces the gasoline used from the power of the engine.

Anyway, all of them are fundamentally wrong. We did not take in account the change in pressure and density of air when it is inside the cylinders.

I also do not recommend to take the model by Lotus and apply it verbatim: it's much more fun to understand it than to apply it, except when you're in a hurry. If you're really in a hurry, you could measure instead of modeling.

BTW, I have a page on engine models here (it includes the Lotus one):
http://www.ciropabon.co.nr/Simuladores_eng.htm

Ciro,

By the time I got to the end of your answer I had forgotten what the question was!

Welcome, John. You're right: it happened to me while I was writing, why it shouldn't happen to you while you were reading?

[Belatti]

Belatti does put forward a very well known and very correct approach - ideal air pump times volumetric efficiency - but shows a poor understanding of just what kind of VE numbers you could expect to apply and why... rendering the method useless (which isn't to take away that his theory wasn't solid up to that point).

Belatti understands perfectly what kind of VE numbers you could expect to apply and why... and WHEN.
Belatti showed graphics (AS AN EXAMPLE, ONLY AS AN EXAMPLE OF HOW EXTREME FIGURES F1 CAN GET) of a modelled, tested and built F1 World Champion Engine from 2005.

SZ called the work of Mr. Virr, Mr Lever and Mr Early from Renault F1 a fraught.

Your example citing 120% VE is fraught with assumptions you seem not to understand. This makes it very pretty but ultimately useless unless your assumptions fluke every critical aspect of the practical engine it simulates. 120% VE? for a street engine your numbers are up to 40% off. For a race engine they're still optimistic. There's a good chance your V10 example wouldn't hit that much.

SZ: don´t get me wrong again, please. I was giving an example of how pros work and not saying 120% VE is achievable in any engine. I think any clever person with a basic knowledge of how and engine works, and using the assumptions the program makes as default, can easily modelate a +/- 5% example to get an approach of air cons.

And PS:

What are you trying to show by the top graph? Sure this isn't more useless info?

I was trying to show how accurate a predicted 1D model can be, no need to complicate things to 3D in the level Conceptual wants.

The second one is easy to understand but the first is a little nondescript and possibly irrelevant - you're showing pressure variations within <15% of atmospheric in what? An intake runner, and airbox, what? Aside from the initial bit being totally useless information (flat line, atmospheric, nothing is happening) - your pressure variations under power in any duct of significance are way more than that, which is where you're really likely to test how accurate your simulation is.

Check the time scale: Thats the pressure wave that occurs when exhaust valve opens, the model is from a 5-in-1 juntcion pipe (one of the two "branches" of the V10)

For more details, contact Renault F1

[SZ]

Belatti, that's what I call selective reading.

Re read what was posted in detail, and have a good, objective look at the first plot. Possibly forget where it's said to come from for a second; and re-read what's inferred about understanding what kind of VE numbers to expect as opposed to believing what numbers a simulation spits out.

[Belatti]

Re read what was posted in detail, and have a good, objective look at the first plot. Possibly forget where it's said to come from for a second;

From a model I have in Lotus Software and use to play with.
I know from the people who made that Model, that with that graphics, its perfectly possible to calculate air consumption as I said in a previous post.

So? What´s the problem? Can´t I have a paper from a F1 team? Can´t I make a model of a V10 using info from that paper and other I estimate thanks to this forum? Can´t I compare my plots with the plots in that paper? Can´t I correct assumptions, like heat exchanging modelization, for instance, to approach numbers and GUESS how they (Renault) work?

See, SZ: you are in a 1st world country uni, with a proper lab, resources, plenty of books, etc.

I have nothing, except internet. The only Uni subject where I studied Engines we have to share time with turbines and pumps, we spent 3/4 year studying water turbines because that was what the teacher did for a living. There is no lab, nor dyno, nor nothing in public unis arround here. Just a classroom, underpaid well predisposed people and some chalk.

If I want to understand how does a 3 liter NA engine to generate 1000HP I have to investigate. A LOT. In my humble models I made a lot of assumptions, sometimes wrong, and corrected them. But 120% VE is not a weird thing in F1 world. I have sources like this:

http://forums.autosport.com/showthread. ... id=2697327

OK, they maybe wrong, but thats all I have got. Gathering 1000s of tiny clues I keep on improving my models and designs (THAT OF COURSE I NEVER BUILD, because with what money???)

Also have some OLD OLD sources, like this:



A NASA paper transduced from an Italian study made in... 1927!!!
Should I trust? I dunno... but there it says 140% VE can be achieved in weird conditions in a 4 cyl engine.

SZ, I appreciate your posts cause you take the most out of me for posting here. I think you know a lot about engines, but I´m not sure about F1 ENGINES. You keep on discussing some numbers and also, as my mother tongue is not English, I may missunderstand some of your comments and it may happens viceversa.

So, its late and I want to get more "pragmatic" if you let me:

Volumetric Efficiency = (Actual CFM / Theoretical CFM) * 100

If you know HP at a given rpm you can approximate your VE (sea level):

VE = (HP * 792001.6) / (AP * CR * CID * RPM)

AP = atmospheric pressure [psi]
CR = compression ratio
VE = volumetric efficiency
CID = cubic inch displacement
RPM = revolutions per minute
(US units to make Chris life easier )

[Belatti]

Chris, take a look at this:
http://www.rpmoutlet.com/formula.htm

[Carlos]

Hope this isn't OT but Chris is one Willey Coyote. Reading his threads from the last few weeks is starting to make sense. It's like that scene in Apocalypse Now where Captain Willard says to Colonel Kurtz " Method? I don't see any method." Except in reverse. I do see a method ... if I'm wrong stick a newspaper between my cheeks and light it on fire! A vortex tube would lower the temperature of intake air for a denser mixture and a Tesla Turbine would pack it in the motor like a Turbo. That's his interest in the average air consumption of an engine. He's got to do his calculations. Did I get it right Chris?

[SZ]

Belatti;

You'd be surprised how cheap and inexpensive my resources are. Don't assume. And you'd be totally wrong to assume I have no working familiarity with racing engines and other extreme engine designs. Just like I'd not assume everything you read on the internet is equally applicable everywhere.

The most important lesson ever passed onto me by people that know more about engines than you or I ever will is that objectivity costs you nothing. You can read whatever you want but what information you pick up from it isn't worth squat if you can't critique it. (Have a good look at that first plot.)

A race engine is a lot easier to simulate than a street engine. Without more detailed information about what affects VE, a simple model you can build in LES isn't going to be as representative as it is for a race engine. Chris' initial specs didn't hint at F1, did they?

A more likely VE is in the 80-90% range, and turbocharging can change this substantially.

Props for researching as you do though... very few students anywhere have that kind of initiative - without sounding condescending (e.g. very sincerely) you'll go far with that kind of initiative.

Do you have access to a copy of Heywood? You'd very likely enjoy it.

[Belatti]

Hey man, sorry if I assumed wrong things
The fact that the only Uni here with decent material charges you 600USD a month and is far away from home, sometimes make me wish public Unis where better. Still, I prefer public Uni teachers

About the first plot: I have missed telling you (and the other forum members) that that graphic is not from a F1 V10 (wich would reach max VE at 18000-20000 rpm and not 7000-8000 rpm). I am fully aware that in street engines more likely VE is in the 80-90% range as you say. The case that these Renault F1 engineers presented in the paper was a very fine tuned one. Maybe that paper confuses a little bit cause there are several cases, examples and plots about different engines all mixed up and you have to read it very carefully.

Still, I have read in lot of places that F1 engines reach 120%VE and that also Honda Vtec engines can achieve similar figures at 8000rpm.

I share and welcome your lesson that "objectivity costs you nothing", thanks.
Also, I know that I can´t assume everything I read on the internet is equally applicable everywhere, all I can do is try to make the pieces fit and sort out the gaps.

@Carlos
This conversation has been very fruitful to me, maybe more than to Chris.