Late model wide case exhausts

[machten]

I've been doing some more reading on this subject.

" The Inertia of that gas front moving down the pipe pulls a depression in the pipe behind it, that removes the residual gas, and reduces the pressure in the cylinder. "

____ Clearly I interpret this to mean that once the expelled bullet of ex.gas is being shot-out through the ex.pipe, it's momentum must leave-behind a relative-vacuum (to do pretty-much as stated).


Yes if you want to call it that way.

This is referred to as the Kadenacy effect. There are some differences of opinion (or interpretation) of the relative contribution of the various effects between the end of the inlet trumpet and the end of an exhaust. A seminal work on this was by Professor Gordon Blair " Design and Simulation of Four Stroke Engines". If you want to invest a few hours of your life, this user manual for a four stroke engine simulator program has an excellent description by a colleague of Professor Gordon Blair's (who did not always agree with him) of the gas wave and particle flows in it starting at page 234. Be warned, it's a big document if you download it - 30MB. Interesting though, and lots of good diagrams and graphs explaining things.

Capture.JPG

Capture2.JPG

http://www.motionsoftware.com/downloads/Dynomation-UsersManual.pdf

Dunno if it is right, but it sure is interesting to me! I haven't finished it yet.

Kev

[DewCatTea-Bob]

[quote= Harvey ...
" The usual term used for this action is " gas front" or "gas plug", "

____ Okay then, thanks for explaining that.
It -("plug") didn't seem like a tech.term to me, so I just assumed it was your-own word (having been chosen only due-to a lack of a more appropriate term coming to mind at the time).



" If we used a pipe that was long enough for the exhaust gas "plug" to develop a depression at low rpm, it would be too long for the sound wave to work at the desired rpm. So unfortunately it is one or the other. "

____ I suppose that could be so, with a mere ordinary straight-pipe,, but certainly a long enough megaphone could still allow both effect advantages, or-else rather with a long extended straight-pipe which includes a 'flute' pressure-relief appropriately located midway (for creating the desired neg.pressure sound-wave).



" The positive sound wave starts at the same time. It has to travel to the open end, expand to change signs to a negative wave, and travel back up to the cylinder arriving at over-lap "

____ Okay, I can certainly accept that those two wave-fronts exist & do as expected,, however I've been left at a loss for full explanation as to how both waves combined can possibly be properly timed (for the v.overlap-period), to make the round-trip if they both travel at the full increased speed of sound.



" to induce the inlet flow.
( as both valves are open at over-lap, and the piston is at TDC, it continues to travel up the inlet tract to start the inlet wave action.) "

____ I suppose anywhere within the v.overlap-period, that must be true,, however in my opinion, it would be most perfect if that neg.wave-front just-happened to be timed to reach the intake-valve just as it has begun to open-up.



" the speed of sound depends on temperature of the gas it is moving through.
as the temp changes the speed does too.
in the average road exhaust about 1500ft-sec,
at race conditions, about 1700ft/sec. "

____ I don't find it too easy to accept that ex.gas-temp could vary so greatly enough to yield such a wide tolerance-range in sound-speed,, but if accepted, then the tuned-length of your 41.5" ex.pipe expansion-point could then possibly vary up to 13%, thus varying the actual tuned-length within a range of 36.5 to 47 inches.



" We work the length out with the formula: 180 X 1600 / 6500 = pipe length in ". "

____ That incredibly overly-simple formula certainly must be a fully simplified magic/short-cut formula, as it obviously cuts-out & bypasses all the more detailed interdependent factor-steps related to reaching the bottom-line result.



" The 180 is the number of engine degrees for the sound wave to do the return run down and back to the cylinder, the 1600 is for the speed of sound in that gas temp. The 6500 is the engine rpm that we want the maximum torque to occur. "

____ That way-simple/abbreviated-formula sure seems pretty-clever though, as it apparently even includes the factor for the valve-timing duration between the ex.valve opening and when the in.valve opens !
How incredibly fantastic is that (if that abbreviated formula really always works) !



" There are a few variables here, if we accept the speed of sound in the pipe to be 1600ft/sec, we then have to settle on how many degrees for the trip.
We say 180* as that is approximately the number of degrees from the exhaust valve opening, to over-lap. So if the effective exhaust timing is open at 70* BBDC and the inlet opens at 30* BTDC, it is about 180*. "

____ Once again, it seems that you & I don't practice the very-same methods for figuring bottom-line result-figures,, as for this particular example you've given, I-myself get '220' degrees, (not "180"),, and that's quite a significant difference !
So I'm wondering if the '180' is actually rather the pre-factored/magic factor-number (that's a reduced end-result which allows your presented formula to be such an abbreviated-formula) ?
____ (Another more common example of an abbreviated-formula, is the one for figuring the capacity of a cylinder...
__ Instead of dividing the bore-diameter by 2 & squaring the radius & multiplying it by 'pi' & multiply that result by the stroke-length,, ya can rather simply just use the associated 'abbreviated-formula' with it's reduced-factor/magic-number of '.785', so as to rather calculate cylinder-volume by more-simply just squaring the bore-diameter & multiply by the stroke-length, and-then merely multiply by just the '.785' 'reduced-factor'.)



" In the example of 41.5" for 6500rpm being the length to the middle of the cone. The Conti is about 22" long, so there is 11" each side of the middle. The longest end would be 52" to resonate at 5540 rpm, and the shortest end of 29.5 would resonate at 9600rpm. "

____ According to this figuring, the point at which the ex.header-diameter becomes increased to the diameter of the cone-opening,, that length of ex.pipe to the expansion-point, is too-short for a 450 to take full advantage of the Conti's entire cone-length, (as it would be considerably better to have the cone's shortest-end resonate nearer to just 8500-RPM).

My interest in reading has been in the physics of the rarefraction and scavenging and reverse scavenging aspects, pipe lengths etc. I'm coming from some way behind you on understanding and I needed to understand the pressure wave motion (in both directions) a little better. One of my degrees is in applied mathematics, I just needed to get a base line into the theory. I now feel I understand the theoretical physics of the effect of the straight pipe and the megaphone,

" One of my degrees is in applied mathematics "

____ Great, as I could use the likes of yourself to confirm that my-own math-figuring is without any faults !
____ Concerning the presented '6500RPM' example, I submit the following set of math-reasonings...
__ At that eng.speed, the ex.valve opens 3250 times per minute, (which is 54.167 times per second), and-so one Otto-cycle takes just .01846-sec, and during that very-short time, the accepted* speed-of-sound -(* 19200in/sec) travels near 354.5-inches !
Now please correct me if I'm wrong,, but in order for the SOUND/pressure-waves to travel the round-trip (from the ex.valve to expansion-point & back, [during 6500RPM]), they'd each then take a route that's 177+ inches long. _ And certainly that length sure seems a good-bit excessively longer of a trip than just the "41.5" inches, isn't that correct ?
__ Anyhow, that's what the extent of my thinking was before,, before Harvey's last post merely-clued me that he wasn't always really in regards to just the standard 'overlap' term (which refers-to the overlap-time when both valves are open near TDC) every time, but rather actually often referring-to the space between the opening of the ex.valve and the opening of the in.valve (which I had been failing to keep in-mind). _ This previously-overlooked realization enlightened me to realize that my math-figuring had been (stupidly!) failing to include the additional step of further reducing a near complete Otto-cycle down-to merely-just the duration between both valve openings (rather instead-of nearly the entire time it takes between consecutive exhaust-cycles). - (A pretty-stupid metal-block, actually ! _ [I'll have to get into the habit of turning-off the TV whenever I'm trying to work with such tech.figuring that I'm not real familiar with !] )
So-thus that correctly-altered factor relatively greatly reduces the round-trip cycle-length (from near-around 720~660, down-to near the 180-degrees), thusly further cutting-down the time/length by around 74%.
So with that correction,, the unrealistic full 177-inch length then becomes reduced down-to a much-more acceptable 46-inches, which thus brings both our presented result-figures together within the same expected ballpark figuring-range !
__ If Kev doesn't post such first,, I intend*to try doing the long/drawn-out math-figuring to pin-down the exact result-figure, and then compare it to the result-figure resulting from Harvey's presented abbreviated-formula.
(* Later,, after I've gotten sufficient sleep, and then don't have anything on TV worthy of distracting me.)


Enlightened-Cheers,
DCT-Bob

[Harvey]

I've been doing some more reading on this subject.

" The Inertia of that gas front moving down the pipe pulls a depression in the pipe behind it, that removes the residual gas, and reduces the pressure in the cylinder. "

____ Clearly I interpret this to mean that once the expelled bullet of ex.gas is being shot-out through the ex.pipe, it's momentum must leave-behind a relative-vacuum (to do pretty-much as stated).


Yes if you want to call it that way.

This is referred to as the Kadenacy effect. There are some differences of opinion (or interpretation) of the relative contribution of the various effects between the end of the inlet trumpet and the end of an exhaust. A seminal work on this was by Professor Gordon Blair " Design and Simulation of Four Stroke Engines". If you want to invest a few hours of your life, this user manual for a four stroke engine simulator program has an excellent description by a colleague of Professor Gordon Blair's (who did not always agree with him) of the gas wave and particle flows in it starting at page 234. Be warned, it's a big document if you download it - 30MB. Interesting though, and lots of good diagrams and graphs explaining things.

Capture.JPG

Capture2.JPG

http://www.motionsoftware.com/downloads/Dynomation-UsersManual.pdf

Dunno if it is right, but it sure is interesting to me! I haven't finished it yet.

Kev

There is a lot written on the subject now days, but a lot of it is not too clear, of which action their are talking about. They seem to mix the two different actions in together.
The diagram that you posted, is of the Inertia action, but is referred to as the Kaadency effect, which is the action of the sound waves.
A lot of the work that you read in Forums is from US engines that develop there torque at a low rpm, 4500/5000, which is where the low speed Inertia action is most useful, but on a Ducati single that will be developing torque around 6500 7500, it is not the action that will produce the best torque, as the sound wave action is better suited to that rpm range.

That's why I suggest you read the "Bible" to get the factually basics of the process of the two actions, so you can decipher the facts from the crap.

[Sam]

Here is a link to a wee chart that will help with your calculations (or a comaprison to your calcs) this has been taken from Graham Bells book (which I used many moons ago when I raced sidecar)

http://www.mez.co.uk/mezporting/exhaust_length.html

[machten]

Harvey,

I do see a lot of conflicting information. The difficulty I have with thinking that "sound waves" have any significant impact (as opposed to pressure waves moving at the localised speed of sound - I suspect this is the source of some confusion) is that it appears to me that sound waves amplitudes are so small as to be irrelevant.

Am I reading that right?

Kev

[DewCatTea-Bob]

[quote= machten ...
" Harvey, "

____ I've now given Harvey sufficient time to reply before bedtime on that side of the globe, (and I've been meaning to submit a post since the same hour that Sam submitted his).
So I don't wish to delay any further, as I had been already waiting to see what Kev would have to post next.



" The difficulty I have with thinking that "sound waves" have any significant impact
is that it appears to me that sound waves amplitudes are so small as to be irrelevant. "

____ I've already brought-up the concern of differentiating between actual sound-waves & pressure-waves previously, and Harvey apparently wasn't inspired to offer any insight on this relevant issue.
__ Actual sound-waves themselves, couldn't be what they are without 'vibrating' in both positive and negative orientation, and-so naturally tend to self-cancel (thus leaving no residual effect).
Furthermore, we can hear the report of the exhaust-note and it certainly is-not consisting of just a single-frequency note-tune, but rather a fairly large hodge-podge group of various different frequencies combined and all mixed-together !
So if actual sound-waves themselves possibly had any significant collective effect, it certainly would be a vast collection of minor effects (that would be a severe nightmare to figure their bottom-line result) !
__ So I'd certainly agree Kev, that you've raised a very relevant concern and-also reached a fairly valid conclusion.



____ I was sorta hoping that Kev's next post would include some math-figuring of his-own that would either help support or otherwise relate to Harvey's presented formula for determining a tuned ex.pipe-length for a given eng.RPM.
Last-night, I was finally going to get-around to posting my-own (non-formula) math-figuring worked-out for determining the tuned pipe-length at the established '6500-RPM',, however I then found that Sam had just posted a link to an active workout-sheet that apparently is supposed to solve such problems (I assume with it's-own built-in formula-workings).
So I then decided to hang-back and leave it in the limelight for a fair while, to see what it might inspire to bring-up next (that might make my-own intended post more relevant [as it would've been, if placed before Sam's post]).
__ I'll now next try to list all the involved math-figuring for working-out the bottom-line result, as had been intended.

__ If Kev doesn't post such first,, I intend to try doing the long/drawn-out math-figuring to pin-down the exact result-figure, and then compare it to the result-figure resulting from Harvey's presented abbreviated-formula.

____ First-off, to carry-out Harvey's own presented (supposed)- formula (of '180x1600/6500'), the result comes-out to 44.3 (supposed inches), which curiously is near 3 inches longer than Harvey's originally stated "41.5", and (coincidentally?) within the expected ballpark tolerance-range.
__ Now I next present the actual math.figuring-steps normally needed to reach the ACTUAL bottom-line inch-length figure !...
_ First, I divide the 6500-RPM established figure by 2, to get 3250 ex.valve openings per minute...
Step-2 _ Then divide by 60 to get 54.167 v.openings per-second, (which converts to 1 complete 720-degree Otto-cycle, that lasts .01846-sec).
Step-3 _ Then-next determine the number of degrees between the exhaust & intake valve opening-times, (which after choosing the std.DESMO-cam valve-timing), thus providing length-duration of 80+110 (with respect to BDC), plus half of the 'overlap' period, gives a total-result of 257-degrees ave.duration.
Step-4 _ Then-next, that result in degrees has to be divided by the 720-degrees (of the single Otto-cycle) to get the resulting fraction-factor of '.35694'.
Step-5 _ Then-next that resulted-factor must be applied against the .01846-sec (time of the 1 Otto-cycle at 6500RPM), in order to get the actual time that the sound-speed wave-fronts have to make the round-trip,, which is: .0065897-sec.
Step-6 _ Then-next, that fraction of time is actually how long the accepted speed of sound has (to make the round-trip through the ex.pipe-length),, which if "1500ft/sec"-(18000in/sec), it then travels 118.6-inches, or if rather "1700ft/sec"-(20400in/sec), then it travels 134.4-inches (for the round-trip).
Step-7 _ Lastly,, the actual tuned-pipe/exhaust-length obviously rather has to be half that 'round-trip' distance, which thus means that the exhaust-pathway needs to have a tuned-length of 59 to 67 inches long (depending on whichever the speed of sound actually is, through the ex.pipe full of hot ex.gas) !
__ So if we logically*split-up the Conti's 22-inch length in-half (* as was also done with the overlap-duration),, then the ex.header-length itself can be shorter by 11-inches, (and-so the ex.header-length should be about 52-inches +/- 4"),, thus-then allowing the full-cone of the Conti to come-into-play centered over the center of the overlap-period of the DESMO-cam !
__ This unexpected out-of-the-ballpark result -(referring-to the rather extensive 59 ~ 67 inch 'tuned-length'), leads me to suspect that there must be another factor that slows-down the speed of sound,, and so I'm now suspecting that the speed of the ex.gas-flow may significantly work against the sound-speed.
____ Hopefully now that this math-figuring includes particular valve-timing such as that involved with Kev's G&W.camshaft that's now employed in his 450, he will finally soon become sufficiently inspired to bother himself with double-checking the actual validity of all my math-figuring ! - (As we-all certainly ought-not let his math.degree go untaken advantage of !)

____ I've now seen that I've fallen-behind in this thread and have another new post by Harvey to respond to. _ So I'll now quite this post at this point and move-on, (and perhaps finish-up my current/leftover thoughts within another post later).


Enlightening-Cheers,
-Bob

[Harvey]

Harvey,

I do see a lot of conflicting information. The difficulty I have with thinking that "sound waves" have any significant impact (as opposed to pressure waves moving at the localised speed of sound - I suspect this is the source of some confusion) is that it appears to me that sound waves amplitudes are so small as to be irrelevant.

Am I reading that right?

Kev

What you are saying is conflicting, " "sound waves" have any significant impact (as opposed to pressure waves moving at the localised speed of sound

I have disabilities that make it slow to write, so it would take a long time to answer all your questions as I would like, so it would be a slow process.
That's why I suggest you read the book, you know you want to
Here you go $42 of the best read.
http://www.bentleypublishers.com/automotive-reference/engineering-and-motorsports/scientific-design-exhaust-and-intake-systems.html

[Harvey]

[quote= machten ...
" Harvey, "

____ I've now given Harvey sufficient time to reply before bedtime on that side of the globe,

Didn't realise I was working to a dead line here.

" The difficulty I have with thinking that "sound waves" have any significant impact
is that it appears to me that sound waves amplitudes are so small as to be irrelevant. "

The amplitude of the wave depends on the speed of the engine, the gas pressure at valve opening, and the rate of the exhaust valve opening, at our speed it would be about negative 4 to 5 psi, at higher speeds and with higher, faster lift, it would be higher.

____ I've already brought-up the concern of differentiating between actual sound-waves & pressure-waves previously, and Harvey apparently wasn't inspired to offer any insight on this relevant issue.

Yes it probably will be an ongoing problem. They are both referring to pressure, just produced by a different means. The Inertia gas pressure, and the sound wave pressure both combine to produce the negative pressure that is used at over-lap. That is the way I referee to them, Exhaust Gas pressure, Sound wave pressure.

__ Actual sound-waves themselves, couldn't be what they are without 'vibrating' in both positive and negative orientation, and-so naturally tend to self-cancel (thus leaving no residual effect).
Furthermore, we can hear the report of the exhaust-note and it certainly is-not consisting of just a single-frequency note-tune, but rather a fairly large hodge-podge group of various different frequencies combined and all mixed-together !
So if actual sound-waves themselves possibly had any significant collective effect, it certainly would be a vast collection of minor effects (that would be a severe nightmare to figure their bottom-line result) !
__ So I'd certainly agree Kev, that you've raised a very relevant concern and-also reached a fairly valid conclusion.

To vibrate there has to be more that one cycle, we only have one sound wave cycle working in the pipe per cycle. So it can't cancel itself out.
I see Bobs added more, later.

[DewCatTea-Bob]

[quote= Harvey ...
" Didn't realise I was working to a dead line here. "

____ Of-course not Harvey !
It seems you've mis-taken the intended jest, as it was rather in regards concerning the turning of the globe, (which of-course even you-yourself are-not expected to be in control of, [I assumed] ).
I was actually merely meaning that I held-off responding-to Kev's post (which was addressed at you) so that my-own response-post would-not unnecessarily interrupt the chain-link you may've chanced to complete before bed-time.

" The difficulty I have with thinking that "sound waves" have any significant impact
is that it appears to me that sound waves amplitudes are so small as to be irrelevant. "

The amplitude of the wave depends on the speed of the engine, the gas pressure at valve opening, and the rate of the exhaust valve opening, at our speed it would be about negative 4 to 5 psi, at higher speeds and with higher, faster lift, it would be higher.

____ It seems fairly clear to me, that while Harvey always remains in reference to THE 'sound-wave', Kev otherwisely seems to be in reference rather more-so to mere noise-waves. _ At-least that's how I-myself might distinguish the difference.
Is that not at-all fair to say ?



" They are both referring to pressure, just produced by a different means. The Inertia gas pressure, and the sound wave pressure both combine to produce the negative pressure that is used at over-lap. That is the way I referee to them, Exhaust Gas pressure, Sound wave pressure. "

____ I've been thinking that the ex.gas bullet/plug pressure-wave has fallen out of relevant contention for what's become the main-topic of this offshoot discussion. _ So that leaves separating mere noise-waves from the actual workhorse pressure-wave that travels at the speed-of-sound.
I'm not sure if it's really fair to assume that any wave-front that happens to travel at the speed of sound should automatically be assumed classified exactly as an actual/true bonafide-legit SOUND-wave. _ However the way I recall it being put (back when I first read on this specific subject), the desired neg.pressure report-front travels from the exhaust-pathway expansion-point back up in-through the ex.port 'at the speed of sound'. _ So I suppose it may possibly be fair to call that wave-front a 'sound-wave',, but I'd really think not, since a real sound-wave is much same as a 'double-edged sword', in that along with it's neg.side, must follow a pos.side, (or-else it can't really be considered as a true sound-wave !). _ And what of that-pos.side, and it's effect on exhaust-flow ??

__ Actual sound-waves themselves, couldn't be what they are without 'vibrating' in both positive and negative orientation, and-so naturally tend to self-cancel

" To vibrate there has to be more that one cycle, we only have one sound wave cycle working in the pipe per cycle. So it can't cancel itself out. "

____ Well then that pretty-much admits that the desired neg.pressure wave is really not a TRUE single 'SOUND-wave cycle', cuz even a single "cycle" actually includes both a positive-side along-with it's negative-side !
Although it's frequency-range could possibly spread-out it's two halves sufficiently far-apart for it's neg.side to perform the prime task,, if it doesn't ever include both it's halves, then it certainly can't be considered as a complete sound-wave,, and if not complete, then it's simply NOT 'sound' !! _ And-so then if it's not real sound, then we really shouldn't refer to the neg.pressure-wave as a 'sound-wave' !



" I see Bobs added more, "

____ That should always be expected whenever my posting is-not yet signed-off with '-Bob' !
__ When I began that prior post, I had awakened from a good long nap and then found nothing on TV worthy of distracting me for the time I had presumed it would take me to work-out the math-figuring which I had before mentioned I would get-around to tackling.
But after working on step-numbers 3 thru 6 (longer than expected, [as I then had-to settle a choice of cam.model for the figuring-example]),, getting past step-7 then began to really bog-down my posting, since TV.programing starting competing for my mind's attention by that time !
__ Sorry for not being ready for your posting,
(you must get-up pretty-early over on your side of the globe, [as your two consecutive posts were placed at 5:59 & 7:17pm EST.time, and Kev mentioned you're 13-hours ahead down-there]) !


Done-Cheers,
-Bob

[ecurbruce]

Not meaning to but-in on your conversation, but...
Harvey, I read that book some 20+ years ago, and it's all you say it is!
Back then I used it's principals to tune exhaust on racing go-cart with some success. Messed with pipe lengths and megaphone tapers. These well known principals have been around for a long time, this book just gives a good explanation.

Bruce