designing muffler for formula cars......

[Nishant_iitd09]

hi frnds,
we are trying to design our own muffler....but still we haven't got any proper theory and concepts , how to design mufflers??
if there is anyone who can help us in this regard....please. help us

we its really hard to design mufflers, but we are fully determined to prepare this year.


Nishant Jaiswal
Team Axlr8r.. IIT delhi, INDIA

[Chris Allbee]

Start by researching acoustic attenuators (or filters). Get familiar with the 3 main types band-pass, high-pass, and low-pass. And just because I know someone will bitch about it: Helmholtz.

Once the math makes sense to you then you should be able to figure out the rest. Fair warning, the math is complex (and sometimes imaginary http://fsae.com/groupee_common/emoticons/icon_smile.gif ) and may take some serious study if you haven't progressed that far in your schoolwork.

[Jimmy01]

I've looked at the theory a few times and realised that I didn't care enough about acoustics to waste my life learning how to get realistic numbers for a system which you probably have not modelled correctly anyway.

My muffler design technique usually involves pulling something out of my a**, last year didn't quite work so we welded an extension on at comp and passed noise. I've found the engine noise (as apposed to exhaust noise) is quite a large contributor to sound level for fsae cars. Muffler design won't help with this, so it's a good idea to have your exhaust outlet as far away from the engine (given packaging and exhaust length).

Get yourself (or borrow) a reasonable sound level meter, make a simple muffler which you think might work, test it. If it fails you could just buy one (from a wrecker if money is an issue) or make one with more restriction. More restriction generally equalls quieter, but you will maybe lose power. But I'd rather have a week of my life to make the car better/ more reliable than wasting it designing an ideal muffler.

[BWill_Inc]

stock mufflers are really lightweight and it might be hard getting use from doing a custom muffler

[Bazanaius]

If you're on a budget, or worrying about noise, generally the stock silencer from the 'full fat' version of your engine (so R6, rather than fazer) is a good bet. A bit heavy for sure, but the restriction really isn't a problem given that you've only got 20mm in front of everything.

[RenM]

most stock mufflers have a very high backpressure and thus reduce your engine performance significantly

[Chris Allbee]

Originally posted by RenM:
most stock mufflers have a very high backpressure and thus reduce your engine performance significantly

"Significantly"? More than a 20mm restrictor? Considering the stock muffler from a modern sportbike is good enough for close to 100hp, I fail to see the logic of your statement.

[RenM]

Mufflers using reflection have a high back pressure because they divert the gas flow.

The air u suck through the restrictor is cold, the one you blow out of your exhaust is hot, thus having a much higher volume. And having 2 flow restrictions does not improve your performance.

[Chris Allbee]

Still calling it a lie. Sorry, but the stock muffler can flow enough air at high RPMs to produce 100hp will be able to flow enough air not to be the main restriction in a FSAE engine. In fact...that's why I think they put the restrictor on the intake side...

Now to put a little more of some mud in the air. Your hot exhaust also has a larger orifice to exit from as in greater than 20mm even on a super-duper restrictive stock muffler (that makes 100hp).

RenM, I'm sorry, but if you are having a hard time making reasonable power exhaust with a 20mm intake restrictor, I can guarantee the problem isn't the muffler.

[RenM]

The thing is, that just because u have a restriction in your intake that is more influential does NOT mean that any other restrictions dont matter at all. It is not as influential as the air restrictor but it does have an effect:

A Muffler that increases your back pressure WILL result in a lower power output for 2 reasons: First of all with a high back pressure level you will find it more difficult to fill your cylinder with fresh gas and secondly your charge-cycle work will increase resulting in higher pumping losses.
That is a matter of fact and has nothing to do with my ability to build mufflers. Actually its the reason why i built my muffler myself with very good results.

It is the same for bends in your manifold or to low pipe diameters. They will increase your backpressure with the same results.

[Chris Allbee]

Ren, you're beginning to get the idea. My beef with your original statement was the unqualified use of "significantly", because I believe it places undue importance on the muffler when it came to making engine power.

Beyond that, you've come close to revealing a few truths about the system, I don't know if you are intentionally hiding them or you just don't realize how close you are. It all has to do with the concept of back-pressure and pumping losses, though its not nearly as cut and dry as you make it seem. I maintain my position that properly designed and tuned elsewhere, one can build a competitive engine with a stock muffler.

[repeatoffender]

Originally posted by Chris Allbee:
Ren, you're beginning to get the idea. My beef with your original statement was the unqualified use of "significantly", because I believe it places undue importance on the muffler when it came to making engine power.

Beyond that, you've come close to revealing a few truths about the system, I don't know if you are intentionally hiding them or you just don't realize how close you are. It all has to do with the concept of back-pressure and pumping losses, though its not nearly as cut and dry as you make it seem. I maintain my position that properly designed and tuned elsewhere, one can build a competitive engine with a stock muffler.

For all this talk Chris you have yet to reveal the truths about the system. I dont know if your intentionally hiding them with some hope of taking the intellectual high ground.

Care to share with the rest of us?

[Chris Allbee]

I don't need any high ground, thank you much. And I'm more than happy to answer any intelligent questions on the subject. But the original poster has not asked any and I only took exception to RenM's blanket statement that stock mufflers have "very high backpressure and thus reduce your engine performance significantly".

Statements like those are misleading and perpetuate "bolt-on" engineering with aftermarket bits and pieces without understanding why they work (or not as the case may be).

And frankly speaking, to say that a stock bike muffler has too much back pressure is a lot like saying its better to tune for high RPM horsepower instead of low RPM torque.

If someone has a specific question about silencer theory or pressure drop and tuning effect then we can delve into specifics, but until then we can get back to the pointless pissing match.

[VFR750R]

Calm down boys. The best muffler is a turbine. Does work while it muffles http://fsae.com/groupee_common/emoticons/icon_smile.gif

That said, if there is a 110 db limit, there is no point in making the car quieter then 110db.

Stock mufflers will have less then 110db and thus have more then required back pressure. Will it be signficant, depends on your definition; will it be measurable, yes. And there are weight and size disadvantages over an optimized design.

Stock muffler can get you by in a pinch, and you could use a muffler for a larger bike to decrease the back pressure. Or modify the stock system like removing some of the chambers.

And exhaust gas due to its temperature (750K) and the fact that the mass flow stays constant means the muffler has to flow ~2.5 times the volume of the restrictor. So for the same speed, ~31.6mm(1.25") exhaust outlet. Don't take that to mean I recommend a 1.25" outlet. I would size this much bigger as long as sound levels can be met.

I recommend to anyone building a muffler to use bernoulli's equation to estimate backpressure using est. mass airflow from restrictor calcs, and estimated or measured exhaust gas temp. You can then calculate pumping loss, by applying that pressure to the pistons working on the exhaust stroke. This should shine some light on what kind of power numbers you're dealing with and how important it is to design a fresh system concerning yourself with backpressure as a primary design consideration. At an intermediate step in the calculations you should be able to tell the outlet velocity of the exhaust flow. I would think about what speed you would like and why.

And finally, why does burns stainless sell megaphones for racing exhaust systems.

[RenM]

Originally posted by Chris Allbee:
I only took exception to RenM's blanket statement that stock mufflers have "very high backpressure and thus reduce your engine performance significantly".


I specified it to mufflers that are using reflection chambers. And there is no blanket statement about that, it is simply a fact.

Significantly is more then 2 NMs of Torque for me.

I also never said that u can not build a good engine with a stock muffler, but if you have the capability to build a good engine you´ll probably have the ability to build a muffler aswell and make the engine even better. Especially when considering that actually manufacturing a muffler is fairly easy.

[Chris Allbee]

Good input VFR. That is a good way to start looking at a muffler design. And if after looking at those rudimentary numbers a team decides to use an aftermarket pipe, then thats just dandy and good for them.

I seriously doubt that most teams know when their muffler becomes a restriction to begin with. Everyone gets so hung up on full-throttle torque numbers that they never stop to think that you hardly get a chance to reach full throttle on FSAE track. Looking at some old data from a one of the Cali competitions I think for a every 60 seconds on track we had less than 3 at full throttle. Thats less than 5%. Not an insignificant amount to be sure, but how often do people lament about chasing the last 5% of something and how much time it wastes etc?

If you're not at full throttle then your upstream pressures will be a little less, meaning your muffler won't be restricting power as much or at all depending on your set-up. In the case of 90% of your track time being spent at less than 80% throttle, does it make sense to spend the time, money, and manpower to create your own muffler? Thats for each team to decide on their own, but other than the weight issue I would argue that a well set-up engine with a stock muffler will not see very tangible performance loss in a real life on-track situation when compared to the custom muffler. If your making decent torque then you both will have to lift on the throttle anyways, so what does it matter?

I understand you will always have "number engineers" who try to eek every last bit out of a system under a certain controlled lab scenario, but fail to ultimately provide real benefit because the need to perform within a complex system wasn't taken into account.

Ren, you attached a number to your "significantly", thats good, but WHERE does it matter? and when? under what condition? The muffler, engine system, or anything on a car for that matter, is not a point or a one dimensional line. If you consider the complexities of the situation and still arrive at the same conclusion with all the data and options having been taken into account, then have confidence in your decision and run with it and ignore what a random jack-ass like myself has to say.

[Mikey Antonakakis]

This is going to be a pretty open-ended, vague question, but you guys that seem to know what you're talking about, what do you think of using software to design a muffler? For instance, Ricardo WAVE? Or not even necessarily designing the muffler, but analyzing its effect on the system. Any insights on accuracy of its impact on the torque curve? Sound output? Anything? How accurate does the model need to be? How fine of a mesh? Long, rambling answers are encouraged.

[RenM]

Originally posted by Chris Allbee:
Looking at some old data from a one of the Cali competitions I think for a every 60 seconds on track we had less than 3 at full throttle. Thats less than 5%.

We have never been to the Californian Competition but in Michigan, the European and the Australian Competition we had full throttle from 25 to 35% of the time. We are running a 4 cylinder engine and are definitely not power limited (acceleration times under 4 seconds).

When i take your reasoning, the air restrictor isn't a part to worry about as well, because you are anyways running on partial load most of the time, not hitting the maximum airflow....

If i consider the complexity of the situation it still gets down to one thing:
F=m*a
The more Power i get the faster i will be on the track (of course only if you can get the power down on the street).

Building your own muffler isn't a hard task at all. From your statements it seems like you need a lot of manpower to build your own muffler but that is not true at all if you do it clever. And Manufacturing is fairly easy as well

[RenM]

In GT Power we modeled our absorption muffler as Helmholtz Resonator with good results on the torque curve. You can not draw any information about the sound output though.

[Professor Gas Can]

"I seriously doubt that most teams know when their muffler becomes a restriction to begin with. Everyone gets so hung up on full-throttle torque numbers that they never stop to think that you hardly get a chance to reach full throttle on FSAE track."

Couple comments:
-To clarify: Backpressure of any source/magnitude has a negative effect on the engine; and it doesn't suddenly become a significant restriction if you happen to have a well flowing intake manifold.
If you're not at WOT, then it shouldn't be a concern for you anyway because if you want more torque, then you push the "skinny pedal" more, regardless of restrictions. So therefore, when considering your manifold and exhaust design, you should be designing at WOT (where they will be significant). To bracket "significant", here is a rule of thumb I've used: 1% peak power for every 1kPa induction pressure loss, and 1% peak power for every 6.8 kPa of backpressure. A good race exhaust (basically straight pipes like the pro's use) is <6-7 kPa. With a muffler, you're probably around the 15-16 kPa, depending on your muffler.
Also, you gotta be careful...last I remember, the competition was still running an acceleration competition, where you are operating at high RPM WOT.
People are hung up on full-throttle torque curves because that's what you design for when you are developing an engine. You want to make the most amount of HP possible with trading off the least amount of torque. So for acceleration you want the most high RPM HP possible, because that's where you're operating.


To quote you a second time: "And frankly speaking, to say that a stock bike muffler has too much back pressure is a lot like saying its better to tune for high RPM horsepower instead of low RPM torque."

In FSAE's case it might not be such a bad idea to tune for some high RPM HP. You need it for acceleration. And like you've said, on track you are at WOT maybe 5% of the time? Why would you tune for all that torque if you're throttling the engine 90% of the time?
Again, on a professional racing level, they don't just tune for low rpm torque, they're looking for the best trade off of HP and torque at an RPM range where they will be the most. It also highly depends on their base engine. There will probably be different strategies for a Corvette C6R than a Ferrari F430.

Mikey: to answer your question: Sure you can use analysis to quantify the pressure drop of your muffler. Probably the best way to do it is use CFD with boundary conditions (pressure, temp, density, mfr etc) from your 1D model. There are ways you can model mufflers in 1D to be pseudo representative (there are auto-meshing tools that cut up the muffler into many flowsplits - maybe half inch cubes) but it doesn't really make any difference to your power and torque predictions than if you had just used a back pressure valve/orifice (maybe a little bit of tuning differences due to valve being a reflective surface, but should be minor). Basically all a highly discretized muffler would do is slow down your 1D simulation.
Of course backpressure's impact on the HP/TQ curve gets increased with the higher you go in RPM (because there is more mass flow rate to create restriction). Basically all backpressure does is tilt the HP/TQ curve up or down at high RPM (large effect at high RPM, low effects at low RPM). See rule of thumb values mentioned above for impacts on performance. Hopefully that was long and rambling enough.

Ian

[VFR750R]

An option not listed yet, is to flowbench a variety of mufflers (stock included) and measure pressure drop at car estimated flow rates. Flow benches aren't free, but you can make your own rudimentary flow bench with a fan and a couple of manometers (which can be made for free for basically nothing). Or use a dyno and your manometer to get real world numbers (be careful to understand where you're measuring pressure and to account for it in your results).

[Mikey Antonakakis]

1% peak power for every 6.8 kPa of backpressure. A good race exhaust (basically straight pipes like the pro's use) is <6-7 kPa. With a muffler, you're probably around the 15-16 kPa, depending on your muffler.
That's about the percent power loss I was imagining with vs without a muffler, and runs in WAVE seem to generally agree with that. My opinion is that, while yes, you can get more power with a custom muffler over a stock muffler, a good deal of time is probably spent somewhere else. For us, if it will meet noise the first time and have less than say 5% impact on power, it's probably good enough. Our team, at least, needs to be spending time on things other than getting every last bit of performance out of the car. And even so, a 2% increase in power with a good muffler will have much less of an impact on the performance of the car on the track than the improvements we can make elsewhere, mainly in the "duh" category, such as "did we tune the suspension," "can our drivers control the car without sliding forward a foot in the cockpit," "does our shifter not break," etc.

[Professor Gas Can]

Yea, I'm by no means an advocate for a custom muffler. Just buy a light weight one that packages well and is cheap and meets the sound requirements. When considering the potential gains you can get from an intake system, exhaust headers and a cam, a muffler is peanuts.
Still, it's nice to know what you leave on the table though, especially when you're asked by the design judges.

[VFR750R]

Two thoughts. Engine guys always have weight hanging over their heads when designing car systems. Which is what drives things like inconel, and although it was hinted at before, weight is a factor. And a suspension guy ought to say, "hey, if you can make that muffler lighter, that's a lot easier to do then make the uprights, or frame lighter, and the muffler isn't structural!" And a muffler that is properly designed can package at nearly ground level, whereas a 'street' muffler will be much harder to package. So as a team, you'll also have to decide what the potential weight savings are worth.

Second, if you get 1hp from the muffler, and 1hp from the air filter, and 1hp from the perfect radius on the runners, and 1hp from the oil, and one from .....point is, little gains add up, if you're willing to work on them all.

[exFSAE]

Originally posted by VFR750R:
Two thoughts. Engine guys always have weight hanging over their heads when designing car systems. Which is what drives things like inconel, and although it was hinted at before, weight is a factor. And a suspension guy ought to say, "hey, if you can make that muffler lighter, that's a lot easier to do then make the uprights, or frame lighter, and the muffler isn't structural!" And a muffler that is properly designed can package at nearly ground level, whereas a 'street' muffler will be much harder to package. So as a team, you'll also have to decide what the potential weight savings are worth.

Second, if you get 1hp from the muffler, and 1hp from the air filter, and 1hp from the perfect radius on the runners, and 1hp from the oil, and one from .....point is, little gains add up, if you're willing to work on them all.

Just imagine the potential gains to be had if they went from carbs to EFI!

...oh wait, thinkin' bout wrong series.

[VFR750R]

Originally posted by exFSAE:


Just imagine the potential gains to be had if they went from carbs to EFI!

...oh wait, thinkin' bout wrong series.

Be nice, we might be going FI in 2011, but it won't be lighter!

[Mikey Antonakakis]

Originally posted by VFR750R:
Two thoughts. Engine guys always have weight hanging over their heads when designing car systems. Which is what drives things like inconel, and although it was hinted at before, weight is a factor. And a suspension guy ought to say, "hey, if you can make that muffler lighter, that's a lot easier to do then make the uprights, or frame lighter, and the muffler isn't structural!" And a muffler that is properly designed can package at nearly ground level, whereas a 'street' muffler will be much harder to package. So as a team, you'll also have to decide what the potential weight savings are worth.
Up to this point we still don't have a dyno http://fsae.com/groupee_common/emoticons/icon_frown.gif but we have a very dedicated, organized person making sure the dyno will be running in the near future.

Also, another way to lose some weight is to go from, say 0.065" tubing to 0.035" or 0.049" (the strategy so far this year).

Oh, I also doubt that we'll be using a stock muffler, and having a running dyno is going to be great because then I can experiment http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[VFR750R]

Originally posted by Mikey Antonakakis:

Up to this point we still don't have a dyno http://fsae.com/groupee_common/emoticons/icon_frown.gif but we have a very dedicated, organized person making sure the dyno will be running in the near future.

Also, another way to lose some weight is to go from, say 0.065" tubing to 0.035" or 0.049" (the strategy so far this year).

Oh, I also doubt that we'll be using a stock muffler, and having a running dyno is going to be great because then I can experiment http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

I don't think anyone in Nascar runs this thin of tubing but some sportycar series cars i know of run .028 on secondaries, and as thin as .019 on tailpipes. To the point of having to weld in a ribbon section of heavier gauge to welds mounts and O2 bungs to.

[Mikey Antonakakis]

Where do you get that tubing???

[VFR750R]

Not positive, but I'd try Pro-Fab in NC.

profabrication_dot_com

They don't list much on their website, but I know they have a lot of stuff, and they build exhaust systems for all kinds of racecars.

For straight sections of tailpipe you could always roll sheet metal and seam weld it. We have used this technique quite a bit for prototype tapered secondary sections.

[Nishant_iitd09]

hi genius...
Is it so that longer mufflers are better, is there any constrain on the length and radius of muffler (other than space contraints)..

[Nishant_iitd09]

hi friends....
we are trying our best to design our own muffler this year. can u help us out in some respect. we are lacking in how to decide the density, radius of perforations in absorption muffler. i got report on this but, it is for 6 cylinder engine. i need it for 4 cylinder.

does the attenuation(transmission loss) depends on thickness of absorbing material, how we can decide it (parameters to decide).

length of muffler is an important task............we hav not got any proper information for this.

thanking in aticipation.

[Nishant_iitd09]

hi friends....
we are trying our best to design our own muffler this year. can u help us out in some respect. we are lacking in how to decide the density, radius of perforations in absorption muffler. i got report on this but, it is for 6 cylinder engine. i need it for 4 cylinder.

does the attenuation(transmission loss) depends on thickness of absorbing material, how we can decide it (parameters to decide).

length of muffler is an important task............we hav not got any proper information for this.

thanking in aticipation.
nishant
AXLR8R,
IIT Delhi.
nishant09iitd@gmail.com

[The_Man]

OK thats good... follow up on this thread.. not take off you multiple posts and avoid starting your posts with: hi genius http://fsae.com/groupee_common/emoticons/icon_smile.gif

About the length yes you can have any length and any radius you like but go through the rules as there is one specifying how much your muffler can hang behind your car which has been changed this year so make sure its the new rules you are referring to. Of course, bigger your muffler more is the weight, that can't be good.

[Nishant_iitd09]

hi friends,
could u help me out,...
how we can decide about the radius of perforation and distance between the perforations for designing the absorptive muffler.


thanking in aticipation.
nishant
AXLR8R,
IIT Delhi.
nishant09iitd@gmail.com

[Nishant_iitd09]

what is the normal range of noise delivered by our formula cars, and whats the range of transmission loss that a muffler can give.

[andrewd]

Originally posted by Nishant_iitd09:
what is the normal range of noise delivered by our formula cars, and whats the range of transmission loss that a muffler can give.

Your mum has the best transmission loss per length ive seen in a long time. Not to mention radius!

[t21jj]

Step one take crappy stock 6.5lb Harley muffler that you have laying around the shop.
http://www.sae.stuorg.iastate.edu/zenphoto/cache/formula/engine/around-the-shop/P1160010.JPG_595.jpg
Cut it up to remove the stock crappy baffles and keep outer perforated tube and packing. Grind off stock mounting bungs and shorten by 4.5in.
http://www.sae.stuorg.iastate.edu/zenphoto/cache/formula/engine/around-the-shop/P1160016.JPG_595.jpg
http://www.sae.stuorg.iastate.edu/zenphoto/cache/formula/engine/around-the-shop/P1160015.JPG_595.jpg
Weld it back up add mounting tab and paint over areas that had the chrome sanded off and mount to car. Final weight 4lb 1oz, the titanium Hindle muffler it replaced was 3lb 4oz.
http://www.sae.stuorg.iastate.edu/zenphoto/cache/formula/engine/around-the-shop/P1160014.JPG_595.jpg
http://www.sae.stuorg.iastate.edu/zenphoto/cache/formula/engine/around-the-shop/P1160017.JPG_595.jpg

*disclaimer do not attempt on a car that still needs to pass tech with out testing.

I was back in Ames visiting and several of us needed to maintain the illusion of our 07 car still having a muffler and do so using free materials that we had in the shop. We did not actually start the car so we don't know how it will sound or effect the car yet (had to use the original muffler for a different project).

[t21jj]

Placeholder for post with photos pending approval.

[Adambomb]

Placeholder for video of gutted Harley muffler in action...or at least as much "action" as you can get with a formula car in Iowa in January... http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Adambomb]

Here's the offending muffler on our '07 car. Sounds surprisingly mellow. My "calibrated ear" guesses it comes in somewhere around 115 dB. Can't wait to autocross it now!

ISU 2007 car with bastardized Harley muffler (http://www.youtube.com/watch?v=j0mMzAT0j9k)

[Gaanja]

@nishanth

there are many e-books that you can get off gigapedia to help you design the muffler. There is a book by a prof from IISc that explains the acoustics of a muffler. i forgot the exact name but i think its listed on google books too. try and get it. it will help you out a lot. We too got a muffler done last time that gave 108 dB. Understand the working of the various frequency filters and how to realise them.

[Nishant_iitd09]

thanx sriganesh...

may i know ur mail-id, i m nishant from IIT Delhi.
hope this year we will also design our own muffler.

[Adambomb]

In a tad bit more seriousness...

The muffler can be a pretty simple component to design. I've had a pretty good amount of fun making various ones over the years, and never really put more than a half hour's worth of thought into any of them. Some of them sounded gorgeous, others awful.

From an analytical standpoint, I think chambered mufflers (similar to Flowmasters) are pretty cool. You could put together a simple Helmholtz model with some assumptions and guesses at parameters, weld some junk together, and if you have some sort of running engine, a tape recorder, and Matlab, it's not hard to do frequency response analysis on it. I made a set for my '88 Ninja 600R that made it sound like a Ferrari.

If you just want to make one that "works," the above pictures of the Harley abomination actually shows most of what goes into a performance motorcycle muffler. If that one was longer and had a larger OD to accommodate more packing, it probably would have met FSAE noise requirements. To get an idea of the hole radius and spacing, you can see a stub of the original outer perforated tube on the left end of the baffle in the second pic above. For scale, that tube is approx. 50 mm in diameter. In my non-professional opinion, in terms of sound transmission, the critical factor with that is not hole radius, but the percentage of open-area in the perforated tube. Hole radius should be kept small to keep the packing from blowing out. For example, a cheap "Cherry Bomb" muffler has only a few large-radius holes. It is loud, and all the packing blows out after about a week of driving.

Making one isn't terribly hard either. For packing, look for people that sell motocross parts (motocross guys tend to re-pack their mufflers often). I've also heard of steel wool working; never tried it myself. For the perforated tube, either find some laying around or you can buy perforated sheet from McMaster, roll it into a tube, and either weld or make some sort of plate to butt the ends together. Trying to drill an effective number of small holes in a piece of tubing would probably drive anyone insane. For the outer case thin wall Al tubing is good, then the end caps can be made of steel and pop riveted in place.

[RenM]

Originally posted by Adambomb:
Trying to drill an effective number of small holes in a piece of tubing would probably drive anyone insane.

I did that for the Titanium Pipe in our Muffler.
Were about 2000 holes but "only" took me 6-8 hours.
But i was probably insane before so i cant really tell if it had any other effect than a lighter muffler http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Adambomb]

Originally posted by RenM:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Adambomb:
Trying to drill an effective number of small holes in a piece of tubing would probably drive anyone insane.

I did that for the Titanium Pipe in our Muffler.
Were about 2000 holes but "only" took me 6-8 hours.
But i was probably insane before so i cant really tell if it had any other effect than a lighter muffler http://fsae.com/groupee_common/emoticons/icon_wink.gif </div></BLOCKQUOTE>

Got to thinking...perhaps laser cut it? If nothing else a CNC with an indexing head. Hmm...

[RenM]

I tried laser cutting on a machine with rotating axis but the guy who owns the machine was not able to get it working.
I dont think its worth the cnc time, if you can do it manually. You´ll probably spend a lot of time setting the machine up and in the end you might have been faster manually anyways and i guess any team has more manpower then cnc time.

You only have to scribe the tube good enough, which is fairly easy with a "height scribe" (havent found a good translation for this tool http://fsae.com/groupee_common/emoticons/icon_wink.gif) and the rest you need is some time....

[Mikey Antonakakis]

http://en.wikibooks.org/wiki/E...n_and_Implementation (http://en.wikibooks.org/wiki/Engineering_Acoustics/Filter_Design_and_Implementation)

[Mikey Antonakakis]

Placeholder for a link. Or you can just google "Engineering Acoustics/Filter Design and Implementation" and click the first link.

[kapps]

Something like that is a prime candidate for laser cutting with a 4th axis rotary. The programmer should be able to set up a table of point locations and translate it. My buddy runs the laser cutter at our company and has even made some 4th axis programs that extend/retract the quill because the part isn't a perfect cylinder. This is on a 30 year old laser and programming g code mostly by hand. That said, it's still going to be quite a run time. Depending on the size of the holes you are looking for and the power of the laser, it might be possible to fire on the run and punch the hole in one shot without stopping any of the axis. If you could do this, it cycle time would be much faster.

[Nishant_iitd09]

hi friends...

what is the normal frequency range and noise produced by an unmuffled formula car.


nishant jaiswal
IIT Delhi.

[Mikey Antonakakis]

There are lots of frequencies... Here's a place to start though:
First of all, I'm assuming you're using a 4-stroke (as per rules). This means each cylinder fires once every two crankshaft revolutions. If you have an even fire 4-cylinder (most sport bike engines), there are two combustion events every crankshaft revolution. To find your RPM in Hz, divide rpm/60. Now again, for every revolution there are 2 combustions, so multiply 2*rpm/60, or rpm/30 Hz. This is the frequency of exhaust events, which is a significant contribution to the noise produced by your engine (intake and mechanical noise are also important, but I'm assuming we're only focusing on exhaust). Logically, the more fuel and air in your mixture, the more torque you make. That torque comes from an increase in pressure in the combustion chamber, and I don't think I'd be incorrect in saying that this pressure is in on the order of 10^3 psi (anyone feel free to correct me here). This pressure is relieved when the exhaust valve opens, so the more pressure, the stronger the pulse coming out of your exhaust ports. The stronger the pulse, the louder, generally. So at least disregarding any exhaust system at all, your exhaust noise will roughly follow your torque curve. Placing any amount of tubing (including your ports) changes the characteristics, however. A longer tube will resonate at a lower frequency, and these resonance effects make an impact on the frequencies produced by your engine.

All that said, what SOUNDS like the frequency range of your engine? When you rev your engine up, if you know anything about music, you can get a general idea of what SOUNDS the loudest. Also, you can get a microphone, and a computer, and download any one of a number of free programs online that will give you a spectrum of sound. Some softwares will allow you to apply weighting curves. Then you run your engine without a muffler, record, and analyze the data. You'll find that there are a whole lot of frequencies produced by your engine... so then just pick the ones that you think are most important, and try to attenuate (or not) them.

Also, I think you'll be seeing around 130dB from maybe a meter or so away from the exhaust outlet.

Finally, I hope you realized how easy any of the math I did in this post was. Generally, if you show that you took some initiative and tried to figure it out first by yourself, you will get a very good response on this board. For instance, saying something like "I figured out that the frequencies of exhaust pulses coming from my engine varies from 50Hz to 500Hz, and I know there are resonances at higher frequencies, but how do I start determining their importance?" will get you a much better response than "What are the frequencies and sound levels produced by an unmuffled formula car," especially when you don't specify the type of engine used.

[Tim.Wright]

Originally posted by Mikey Antonakakis:
There are lots of frequencies... Here's a place to start though:
First of all, I'm assuming you're using a 4-stroke (as per rules). This means each cylinder fires once every two crankshaft revolutions. If you have an even fire 4-cylinder (most sport bike engines), there are two combustion events every crankshaft revolution. To find your RPM in Hz, divide rpm/60. Now again, for every revolution there are 2 combustions, so multiply 2*rpm/60, or rpm/30 Hz. This is the frequency of exhaust events, which is a significant contribution to the noise produced by your engine (intake and mechanical noise are also important...

....Also, you can get a microphone, and a computer, and download any one of a number of free programs online that will give you a spectrum of sound. Some softwares will allow you to apply weighting curves. Then you run your engine without a muffler, record, and analyze the data. You'll find that there are a whole lot of frequencies produced by your engine... so then just pick the ones that you think are most important, and try to attenuate (or not) them...

pfft... you could do that.

Or you could just ask for the answer on the FSAE forums http://fsae.com/groupee_common/emoticons/icon_wink.gif

Tim

[samphlett]

Amusing thread.

Ask you faculty to buy "Acoustics of ducts and mufflers", M.L.Munjal

Or run simulations.

[Nishant_iitd09]

what is the use of WASH PLUG IN PACKING OF A MUFFLER.

[Nishant_iitd09]

COULD U SUGGEST SOME BEST POSSIBLE PACKING MATERIAL FOR MUFFLER....



I HAVE HEARD OF "SPEC 30".... HOW IS IT??

KINDLY REPOSND SOON.

[Drew Price]

COULD U PERFORM N EXPERIMENTATION?

RESPOND ON THIS TOPIC KINDLY.

Please read Mikey's response above about what constitutes a good way to frame your question.

Or use his earlier suggestion and try some steel wool - and see all the pretty sparks!

Best,
Drew

[Mikey Antonakakis]

I followed my advice, and it sounds pretty good! And the muffler is only tacked together so far.

[Adambomb]

I MAKE IT A POINT NOT TO RESPOND TO DEMANDS POSTED IN ALL CAPS. EVEN LESS SO WHEN THEY END IN "RESPOND SOON."

[Nishant_iitd09]

hi friends..

can any1 help me with d data that.... what is the approximate range of frequency of sound(i repeat frequency of sound, not the frequency of engine) that is emitted without a muffler.


Nishant Jaiswal
nishant09iitd@gmail.com

[Nishant_iitd09]

what is this quantity.... "rayl" and "rayl/m"...

Nishant jaiswal
nishant09iitd@gmail.com

[Drew Price]

I will help you out, because this thread has had me thinking:

Recall that the rules limit the amplitude of sound.

You need to (at minimum) find the frequencies of your specific engine/exhaust combination which exceed the 110dBa limit at the test RPM and try to minimize those.

Those frequencies will differ (some, at least) between every single car present, based on exhaust length, shape, engine configuration and calibration, etc.

The easiest way to find what you need to minimize is probably to follow Mikey's (excellent) advice above, and get some free frequency analysis programs and measure your engine. You don't even need an expensive microphone, since the first order frequency ranges you need to attenuate should be very clear, even with a webcam mic and free software.

Best,
Drew

[Mehul Botadra]

Originally posted by Drew Price:
I will help you out, because this thread has had me thinking:

Recall that the rules limit the amplitude of sound.

You need to (at minimum) find the frequencies of your specific engine/exhaust combination which exceed the 110dBa limit at the test RPM and try to minimize those.

Those frequencies will differ (some, at least) between every single car present, based on exhaust length, shape, engine configuration and calibration, etc.

The easiest way to find what you need to minimize is probably to follow Mikey's (excellent) advice above, and get some free frequency analysis programs and measure your engine. You don't even need an expensive microphone, since the first order frequency ranges you need to attenuate should be very clear, even with a webcam mic and free software.

Best,
Drew


Good suggestion! You might be just stuck up if your values range around 110 db! Ours was around 108db so you dont wanna take a risk!

[Mikey Antonakakis]

I'm not sure if anyone heard our car at competition... besides the low-rpm rattle (broken tack weld in the muffler) what did you guys think? I was trying to produce a specific sound from the muffler, I think I was fairly successful. It was too loud, but it was hardly packed. I made sure I left room to be able to quickly add a second muffler if needed, which despite adding a few pounds worked like a charm.

[AxelRipper]

We tried designing our own muffler this year, and actually had it built and working (seemed to be quiet enough, but didnt have a good area or conditions to test it). However, the carbon fiber case cracked (almost like UofM dearborn's during endurance) during testing before competition. With a suzuki muffler on our 450x we came in at 105db, which was louder than the muffler that we used at cali last year by 2 (?) db.

[coastertrav]

Flowmaster was one of our sponsors and provided us with a prototype muffler, which we believe had too much backpressure elevating our egt's too much leading to this...


http://sphotos.ak.fbcdn.net/hphotos-ak-ash1/hs553.ash1/32239_426452724202_691079202_5563847_2780335_n.jpg

Could be something silly we did, and we are not laying blame on Flowmaster just yet, but we switched to an oem Suzuki muffler and didn't have any problems (we also swapped motors due to burnt valves so that could have something to do with it as well).

[VFR750R]

I was amazed by the difference in sound level from the various cars in endurance going by. Several cars (I remember Lafayette) were verrrrry loud, and I was surprised they passed. The singles are loud too, but obviously at a different frequency. The turbo cars are ridiculously quite in comparison (except UTA's 250 turbo), and I wonder sometimes how they need mufflers at all.

[Nishant_iitd09]

Hi friends,

How to decide the perforation radius of the muffler. Somewhere i found that it is not the size of pores which matters, its just that we should keep it small to save packing material from direct burnout. But, still there would be some dimension which will be preferable for muffler perforation.

what consideration needs to be take to decide the pores?

What would be better option for fabricating these holes. EDM or normal drilling.

[Mbirt]

The two parameters you're concerned with are porosity and transparency. Porosity is the ratio of open area to closed area and is not affected by perforation hole radius. Transparency, at a given porosity level, is greater for a perforation pattern with smaller holes. If I remember correctly, high-frequency attenuation increases with greater transparency.

And, of course, smaller perforations will hold your absorptive material in better.

Use whatever method you like to create the holes, but take care to make sure the inside of the tube is burr-free. Burrs create flow noise.

[Mbirt]

I found an awesome pdf explaining the concept of transparency:

w w w . iperf . org / IPRF_ACUSES . pdf

You could skip staright to section "III. The "Transparency" Approach", but I found the whole document to be golden.

I'll tell you that 4mm is the auto industry standard "small" size. Choose a reasonable hole spacing, as it's a shame to see cars DNF endurance with muffler failures.

[Nishant_iitd09]

Thanks a lot Mbrit.
That PDF was really awesome. It explained very clearly how to decide various parameters.

But still, it is not clear that whether we have to go for higher TI or lower TI. Since TI increases with smaller attenuation of sound and effective muffler is which has higher attenuation. this clears that we should go for smaller TI, and in the literature in PDF it is mentioned, many times, to go with higher TI.

what do you say in this case?

Do you have any idea, generally, in what range of frequency, sound is produced in the cars we fabricate.

Nishant Jaiswal
nishant09iitd@gmail.com

[Mbirt]

Mikey Antonakakis's long post on page 3 of this thread is excellent advice. The sound created by your primary firing frequency at the noise test rpm for your engine's stroke is the place to start.

Let's say you have a CBR600F4i. Its noise test speed is 11,000 rpm. (11,000 rpm/60 s)(2 pulses/rev)=366.67 Hz. This is referred to in the auto industry as the "second order". Due to exhaust cam timing, multiples of this frequency will also likely be prominent in your exhaust's acoustic signature. I'm not a 4-cylinder guy, but I would focus on the 2-8 orders to find my frequency range to attenuate. You'll find that the 8th order frequency is not high enough for the transparency index play much significance. There is much more to be gained from adding volume within the muffler shell.

A single's primary firing frequency is the half order and, at noise test speed, is severely squashed by the curve of the A-weighted dB scale. I think we'll be seeing more creative muffler designs in the future for singles. I've got one on the dyno right now.

[Wesley]

High-frequency noise from turbulence is, from most of the data I've seen, often a major reason FSAE cars fail noise tests versus second order exhaust notes - poorly fabricated headers or mismatched diameters, steps in the exhaust system, any small protrusions that cause turbulence will significantly increase the high frequency noise that you generate.

It's really easy to tune out the 11k RPM drone with a properly designed band-stop filter. Chamber mufflers are excellent at this - but most FSAE teams don't use chambered mufflers because packed mufflers attenuate a broader range of high frequencies (and restrict less, to boot!) which are usually the problem, especially considering the A-weighting scale in which high frequencies have considerably higher gain (~ -10dB gain for 300Hz sounds compared to ~2dB gain for 3k Hz sounds, relating to the equal loudness Fletcher-Munson curves of the human ear)

If you're going so far as to actually tune your exhaust note aesthetically(which in itself isn't a bad study, since most all manufacturers spend countless hours subjectively designing the "tone" of their exhaust systems) then you have even more study ahead, but from a simple "pass noise tech" standpoint, high frequencies generated by turbulence are killer, followed by the second order frequency.

[Nishant_iitd09]

Sir,
We are running short of time for designing our muffler. We are falling in deciding the perforation parameters for Muffler Tailpipe. The graph we have got is for 10KHz frequency but our requirement is below 1000Hz. We are not getting the graph in our range. TI (transparency index) of pipe can be calculated according to various iteration of perforation diameter and bar size. We are working with material SS 321 pipe of thickness 1.5mm.
Sir, please help us in this regard.


Thanking in anticipation.

Yours truly,
Nishant Jaiswal
nishant09iitd@gmail.com

[Adambomb]

Not to sound like too much of a jerk, but if you have spent 17 months on muffler design and you are roadblocked by not having a graph (in a pdf someone else provided for you, no less) in the proper frequency range, then I just have to say, I don't think the real problem has anything to do with perforation parameters.

Have you built something, anything yet?

[Mbirt]

1. Use a slightly smaller diameter perf tube than you think you need.
2. Use a smaller-diameter "tailpipe" than that.
3. Use a slightly higher-quality fiberglass than you think you need.
4. Use a larger-diameter shell than you think you need.
5. Make the muffler and call it a day.

^^Says the guy who wasted all day at work playing with mufflers in GT-Power...

[Adambomb]

+1, I'm still proud (in a redneck sort of way) of the modified Harley muffler on page 2 of this post. It is too loud, primarily because it goes against all of principles 1-4 that Mbirt just mentioned, but it was free, and if it wasn't for having to grind off those stupid humongous mounting bungs we probably could have finished the job in an hour, which was a smashing success at #5. Plus it's chrome. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

If you want this problem to just "go away," do exactly like Mbirt says, and as long as it's sturdy and relatively well-sealed you should have something that works well, reliably, with no calcs involved.

[manifold]

hey.....
regarding exhaust tuning its not a perfect art....partially dependent on assumptions you take...there are parameters to determine, but there are no hard and fast rules to follow.
first of all if you are designing a exhaust system i.e complete unit, then please tell me what approach you took for designing the manifold?. regarding muffler its all about wave dynamics...
every thing is dependent on what power band you want your max power ....that will determine your dimensions for muffler ..in the same wat the convergent cone and the divergent cone has its own theory....and have specific angle acc to what you want from the engine...!! it cant be described in a para..

[dafxsua]

We are designing our own muffler this year, the building of it has been a challenge not doubt. However in terms of theory, I would look at Transmission Loss (TL) and Insertion Loss (IL). The muffler you design will have a TL for any given frequency. I suggest looking
into exactly what frequencies you want to attenuate for your engine and design your muffler around that frequency range so that you have adequate TL at that frequency.