Boozehound

I think he's talkin' about a ' Jet Assisted Take - Off' rocket. (See Darwin Awards)

Boozehound

I agree that losses do to drag along the surface of the pipe are going to be negligible due to the cross sectional area versus the surface area of pipe for MOST vehicular applications - this probably isn't the case if the pipe gets extended very far. There's more fluid dynamics in an exhaust stream than just assuming fully developed flow with negligible losses at the wall. Think 8 mile water hose vs. the 8 ft. variety. Still, this isn't a simple area and volume calc, there's entry length, reynolds numbers, etc - that need to be accounted for.

At any rate, the cat and the muffler are going to be significant in lowering the velocity of exiting gasses, and building pressure in the system. I'm not up to speed on what kind of volumetric flow rate you can get from most 'modern' cats, but forcing a gas through a porus ceramic substrate isn't likely to be neglible (see aftermarket high-flow cat replacement industry).

As for the muffler, I don't propose strapping on the "cookie jar" muffler (I think they're ugly and sound awful, but that's just my opinion ), but there is something to adding a larger diameter exhaust tip. It acts as a diffuser. Sure, it lowers the velocity of the exiting exhaust, but the pressure drop at is what we're after, not velocity. If it were velocity, it'd make sense to strap a nozzle on there and blast it out of there like the exhaust coming out of an afterburner or turbofan (if you like the subsonic analogy). Its the whole "try blowing through a funnel in both directions" effect.

Another thing to consider would be headers. I'm not sure about Mazda's current design, but I would guess it's a cast piece that's meant to retain heat and be cheap to manufacture while still flowing a decent amount. I dont know if their design takes into account any 'scouring' effects, but you can time the pulses in the exhaust from one header to slightly lead the other pulses in the other tube coming from the other port, and effectively make use of entrainment and get the first pulse to "suck" the second one along. This is something that's mainly seen in X-pipes and on some header systems, but I have no clue how the RX-8 will flow.

I'm not sure why heat loss in the exhaust after the cats is a bad thing, and if you dont care about emissions, then it shouldn't matter what temp the exhaust falls to upon exit. I wouldn't think that compressability due to temp change is a real big problem.

As for the weight, you're right, I think that poorly done exhaust systems can in effect be detrimental to gains, or at best they take away possible improvements, but if you're replacing factory parts, then you probably have the advantage in the fact that you're not paying for mass production, cost, heat shields and the like. The whole weight thing is a system by system determination.

Long story short, I think that even though the Renesis is optimized for certain operations, it is still a mass produced engine/exhaust/powertrain that can probably be "improved" in regards to whatever your goal may be, whether it's performance, sound, emissions, whatever.

Technical discussions are great aren't they? It makes me really miss school.... No wait, there was all that homework and projects and test and... ah screw it, I'm gonna go watch TV.

wakeech

the length of the exhaust pipe in the RX-8 isn't more than 10 ft anyways

although the cat isn't as unrestrictive at very high rates of flow as a straight-pipe, it's really not nearly as bad as they were even as little as 5 years ago. the muffler (given its level of muffle, as reported by people who've seen the car) is probably the biggest culprit, and definitely a place to knock 10lbs off the back of the car.

the reason that hitting an optimal velocity is important is because the faster a gas moves in a closed volume, the lower its static pressure (either static or dynamic, i can't friggin remember) drops propotionately (according to some relation i can't remember): it's the Brenoulli Effect, and that's how AFMs work, isn't it??

anyhoo, the analogy between the large tip and diffuser... i can see where your thinking is going (i used to think the same thing). if we compare a diffuser on the back of a race car in a wind tunnel (instead of on a track, because we're blowing air over/through it, not moving the tip through the exhaust), it reduces the pressure of the air along the surface of the curving surface (which expands the volume up, away from the ground) it is accelerating air ONLY in an upward direction, which means the air being pulled up speeds up, to lower its pressure and exert an equal and opposite force on the diffuser downward, creating downforce. this action of changing the velocity of the air (from straight underneath the car to upward) does in fact slow the velocity (which in the windtunnel is how you get your drag metric) of the air passing through it.
the oversize tip, on the other hand, only expands the volume the gas is to fill in a closed volume. basically, this expansive force (in a round tip) is equal in all directions, and doesn't help the exhaust system create any suck out the back (ambient pressure outside the exhaust pipe stays the same), just a slight amount of suck inside the tip against itself... if the tip is abnormally shaped, then you'll end up with abnormally distributed forces around the tip, but still won't help generate a force pulling exhaust straight out of the tube, just pulling it apart to the sides of the tip (still inside the pipe). blah, i'm long winded

yes, definitely. the current spec header is a twin skinned (very good) log style (very very bad) header, which was manufactured with simplicity and some of the architectural constraints of the engine bay in mind.
an equal-length header (with those runners tuned in diameter and length to flow optimally a given volume at a given rpm... no more difficult to figure than the math i did above) would definitely show the most gain, short of opening the engine. ... ummmm... there are some kind of advantages to log-style headers, but i can't remember what they are.... anyways, real headers are best.

oh, and what you were saying before about having no tail-pipe after the header is absolutely true and right: that's what they do in F1, but having the length so short is what they need as they run 18-19k on their engines

what i've read is that by insulating the exhaust gasses until they've exited the system, you increase the enthalpy (...or is it entropy?? it's been so long, i've confused them: one's "heat", the other "randomness" *shrug* whatever, you know what i mean) difference, which by keeping the exhaust hotter is almost like making the ambient air cooler, hence improving the thermal efficiency of the engine (however marginally: i don't really know how important that is/could be), which is unarguably a good thing.

yeah, but what i mean is the same thing with performance wheels: if you take a specific design (ie: canister design, bends in the tube, length, etc) and compare two different diameters, obviously the smaller one will be a tiny bit lighter.

no doubt there can be some comprimizes made, but again, with a header that is setup for a slight gain in torque at 7500rpm, which could equate into 10hp or somehting like that, you may sacrifice fuel consumption rate (somethign i'm sure you're not concerned with), emissions (again, not a concern), and (get ready for it) bottom end torque *duunnn dunnn duuuunnnnnn* :p... and i dont' know how you feel about that... of course you could (in theory) go the other way, with a longer header, smaller diameter exhaust system throughout, and get really killer off-the-line acceleration (only to choke it up top)

Boozehound

There's a whole lot more to these flow equations, than just plugging in a volume of escaping gas and and the area of the pipe. You have to worry about things like the reynolds number, entry effects, blah blah blah (Do a search for the Navier-Stokes equation online, and that equation will accurately predict the nature of any flow - solve it, and you'll be my hero). Things like mandrel bending and joints in pipes affect the velocity profile for any fluid flow through a pipe, and also mean a lot in a situation like this.

As for the diffuser, quit thinking of planar surfaces. the nozzle/diffuser is a very common effect that is used everywhere. If you don't believe that adding a cone to the end of hose will accelerate your flow by way of a pressure drop, go with the experiment - get a small funnel, blow from the large end to the small (hard isn't it - and the flow out the other side is strong, yes?). Now turn it around and blow through the small end - doesn't feel like you can keep up right? I'd bust out with my fluids mechanics book & equations, but I dont have them with me - maybe in a week or two, I'll go find them.

As for thermal efficiency, I would think that your control volume (boundary around the system) does not include the exhaust system. I dont think that when you calculate thermal efficiencies of the Otto cycle that you need to worry about when that heat is lost after it's exhausted from the system. I can't think of any useful work that wasted thermal energy from the combustion process is going to complete. And as such, if you're using your temp/pressure at the end of the exhaust system (and I think it's wrong), you're just adjusting your efficiencies synthetically - it doesn't actually change how efficient the engine is at converting chemical energy into kinetic and thermal energy.

And I think the word you were looking for is enthalpy. Entropy is the 'randomness' of things. http://www.mchawking.com/multimedia...._function=mp3z
Let MC Hawking 'splain it (d'load Entropy.MP3)....

You seem to dig the engineering thing, you should drop the economics and come join the dark side. Or is it dork side? Nah, I think we're nerds not dorks....

wakeech

that's true, but assuming that all is as optimized as you can get it between two comparable systems (the welds are nice, and the bends are done with the mandrel thingy, not one of those gawdawful hydrolic presses, the interior surface of the tubing is very similar) then the largest determinant impacting the performance of your engine would then just be sizing. if you're comparing two systems which have a huge disparity between the quality, then of course this won't work. my origional point was that over-sizing your exhaust is certainly not of universal benefit, and doens't magically make horsepower: i think i've illustrated that.

sounds good, fill me in.

hey, i dunno... just what i've read. i'll look more closely into it.

good, i was right then

a lot of people have suggested it, and i thought about it coming into school too, but i don't think that my triste of posts on a forum will really please the Admissions Officer enough to ignore my entry grades :p (that jerk)
...math is the only problem. but whatev, money is just as cool, but there doesnt' seem to be any kind of recreational study of economics around, and thus no forums kinda sucky.

Boozehound

WHAT!?! No accounting forums?! I won't hear of it!

zoom44

jato car

wakeech

jesus, that looks like a good waste of an hour or three

cool link

[edit] half-done, and that's seriously the funniest thing i've read in a long time.

wakeech

:p!! ECONOMICS forums... i've never come across one :p

Boozehound

My bad - but on the plus side, accts and economists both get to talk about money all day long.

wakeech

no biggie: but the difference between the stuff accountants and economists talk about can be qualified as "pretty boring" to "positively interesting" respectively

babylou

How about spinning a turbo?

We shouldn't be too hard on Wakeech. He will probably be the guy that gives the world a commercial cold fusion unit since that bastard in Cali that is running his A/C off of one won't let anyone see it.

Oh, and don't kid yourself, engineers are nerds, dorks and geeks all rolled into one pocket protected, short sleeved dress shirt, unmatched belt and shoes package. I am the exception. I am a GQing cool cat, baby. I imagine that maybe, just maybe a combined Economengineer would be the king of the aformentioned group.:D :D

Boozehound

What? No tie with the short sleeved shirt? Or is that just for the GQ engineers?:D

babylou

Yah, I forgot about the tie, a narrow solid color one at that. What about the balck plastic framed birth control glasses.? Images of NASA's mission control during the moonshot are dancing in my head.

Boozehound

Just so long as we're not using slide rules anymore... that's a quantum leap in style - unless you have a holster for your calculator of course...


Babylou-

I wasn't trying to bust Wakeech's chops on all of this stuff, my point is that most of this stuff is much more detailed and involved than it appears at first glance. And if you dont want to do all of the math and legwork, then the only way to estimate performance from alot of these addons is on a dyno. I also think that it's outstanding anyone whos not trying to make a living out of engineering would get involved and actually look into the why of things, like cars. I'm all about more enthusiats who dont want to pay someone to think for them ("uh, just make it go faster - here's my amex.")

DTECH-RX

and now back on topic....... :p

wakeech

for sure. i'm not trying to pretend i know a lot more than i do, but i do understand a few things... but yeah, you guys give me way too much credit anyways...

but yes, back on topic. as i admittedly know nothing close to everything, i'm really really looking forward to this whole diffusion effect thing...

babylou

Yes, the ol'e calculator in the holster look. That is as GQ as one can get.

I know you are not cracking on Wakeech just letting him know that the devil is in the details.

Yes, back on the topic of N/A bolt on power adders. This basically means intake or exhaust.

1. Just by looking at pics it seems the intake system is highly developed with what seems to be factory CAI for the primary snorkel. A freer flowing cotton or foam filter may be worth 3 hp. I say "may" because Mazda had a goal of 260 hp and surely worked their tail off on the intake side since there are no emmission compromises required. Is the secondary air snorkel also CAI? If not then that could be worth 2 hp.

2. The header definetly has some performance compromise to meet emmissions and cost constraints. It appears to be a dual walled. This does not affect power, improves emmissions but is heavy. The header also forces the exhaust to make an abrupt turn. I believe a good mandrel bent header will make worthwhile power increases, maybe 15 hp. Of course 90% of the improvement will be at 7,000 rpm and up.

3. The exhaust seems to be quite simple looking in pics but the inlet into the muffler may be restrictive. Maybe 5 hp here?

4. Someone earlier mentioned weight reduction is as good as power increase. Weight reduction is better. Weight reduction adds power throughout the rev range, like a displacement increase. As a bonus, one gets better fuel efficiency, lower emmissions, better handling, better brake fade resistance and improved crash safety (non-structural weight reduction only). There is a thread on weight reduction floating around. I think once this car comes out and we dig into the bowels some more there will be relatively easy methods to reduce mass by 100 lbs. It won't be cheap but it will be doable. The cool thing is sometimes one can get weight reductions for little cost if the change also improves other areas. For example, a fabricated header will usually be lighter. Or when buying wheels for vanity also look to get some lightweight ones at the same time.

DTECH-RX

Excellent response that makes perfect sense babylou! It's nice to know that there are a few more "possible" somewhat easily attainable HP and torque available as part of a customizing process a lot of us will be doing.

My question about the intake is that if it is true that forced induction is being developed for a Mazdaspeed RX-8 as mentioned by Rotary News, what is the most logical way they are probably going about this? I know some people have mentioned a turbocharger being difficult, if not "impossible" given the location of an engine mount or other structure, so how would they go about supercharging the RENESIS? I've been staring at the RENESIS for a while now, and I'm thinking that it would have to be a centrifugal supercharger of some sort because there is no way to replace the intake manifold with a positive displacement SC for obvious reasons. The intake piping and actual mounting still looks like a very interesting puzzle although I have a couple of possibilities visualized.

This brings up another interesting alternative.....good 'ol NAWS. :D It would seem that NOS would be the easiest and possibly even the "friendliest" power adder the RENESIS could have. hmmmmm......

Boozehound

So long as we're talking about FI, how about a centrifugal supercharger that's engaged with an electromagnetic clutch (a la C32 AMG and old school blown MR2s (~circa 88))? HP and torque on demand, with increased reliability and fuel economy. And so long as we're going down the exotic solutions route, how about Ford's new supercooler on the Lightening concept, to act as an intercooler?

DTECH-RX

Very kewl alternatives Boozehound.....I'll take the electromagnetic supercharger one step further. Have a switch with OFF / AUTO / ON on it to allow for it being off, computer controlled, or always on... :D The supercooler sounds neat too, but let's not forget that we don't want to spoil the RX-8's weight distribution by adding too much weight up front. Another reason to keep the compressor and related piping compact and light. Weight is yet another argument for a couple of strategically placed and wired "Sneaky Pete" NOS cylinders.... :D

wakeech

*ahem* seeing as this thread is about NOT going the FI route, lets continue with nitros... i dont' know nearly anything about it (just can't remember: too boring), sooooo, Lou??

Sputnik

Okay, what is this "supercooler" thing? Is it an A/C backed air/water IC like the "Intracooler"? Is it a mere air/water IC?

Remember, the MR2 MKI SC had a roots SC, not that it makes much of a difference.

Most roots-type blowers are available with a bypass valve now, which will open at cruising/idle, very similar to bypass valves on turbo setups. By the time you add all of the rotational weight of such a clutch system, and the complexity (and therefore potential expensive problems), you gain very little from not having to turn an unloaded SC. Remember, if the SC is unloaded, then it's not sapping as much power as when it's up to speed, pumping air.

Don't forget, the ports, intake system, exhaust system, etc. are going to have to either be optimized for full rpm WOT power, optimized for best economy, or a compromise between power and economy. Regardless, you will have to open up the intake and exhaust systems, so that you don't "choke" the SC when it's going full song. This means that you will not have the optimum setup for fuel economy at cruise/idle, so even if you are successful in bypassing the SC restrictions at cruise/idle, you still won't have the same fuel economy as an NA car.

---jps

DTECH-RX

Well, the nice thing about nitrous is that it introduces more oxygen as well as cools the intake charge, both at the same time. This would sound like an ideal situation for a rotary since as long as you get the fuel mixture right, you can eliminate the possibility of detonation and damaging the apex seals, at least in theory. It also lets you cruise around with a minimum of added weight which can be mounted to maintain a 50/50 weight distribution as well as keeping fuel economy and everything else stock while not actively "spraying".

Sine the RX-8 is already set up to dominate the twisties. A little boost of "juice" would help it squirt down the straights just a little faster making for some really impressive lap times! You could even make the activation of nitrous throttle dependant so after you "arm" the nitrous switch it will automatically spray once the throttle reaches a certain position and even meter the amount going in based on throttle position. Definitely some interesting possibilities.

:D

Sputnik

If you get the fuel mixture right. How're you going to go about doing that?

Quote:

...You could even make the activation of nitrous throttle dependant so after you "arm" the nitrous switch it will automatically spray once the throttle reaches a certain position and even meter the amount going in based on throttle position...

Given you could control the flow enough so that it didn't come on/go off abruptly, how do you make it last long enough for more than a couple of laps?

---jps