PHYSICS to correct a poor previous (no gain) Muffler Mod in a OEM muffler.

[one.man.band]

Since i did it the wrong way......achieved no perceived gains, other than letting my neighbors (in a 1/4 mile radius) know that they should buy some ear plugs. to be fair, the saw did rev up quicker.

What i did the first time:
The OEM muffler did not have any baffles or catalytic converter. The OEM stock hole (round) diameter was 0.213 inches, and did looked very small once the screen was removed. Then i drilled two @ 3/16ths inch holes beside the original stock exit hole. Re-tuned the saw many times as i was cutting, using two tankfuls of fuel. Plastic from the saw case cover was melting, and so were my hands from the heat.

The muffler exit hole (now 3 holes), were on the top face of the muffler, and exit holes needed to be changed to side face or front face to redirect the hot fumes. The deflector could not redirect the additional flow away good enough. Was advised that my exit holes were too small to make a difference and the Basic rule of thumb, which was to enlarge muffler exit hole to 85%-110%(? can't remember the high end) of exhaust port area.

Got the welder out with some shim stock, to weld the holes shut. Then spent about an hour repairing burn-throughs. Enlarged the exit hole in muffler to 61% of ex. port area. Did not enlarge up to 85%, because I figured it was easier to take metal away, than to put it back. Could always enlarge the hole as necessary to work my way to 85%.

----------------------------------------------------------------------
If I had the time, initially, I would have worked the #'s and thought about this for a while, before drilling. This is what I have found:

The saw in question is a Husqvarna 340e with 18" bar .325 yellow Stihl chain.

the math:
Using stock muffler internal volume, muffler exit hole area/length using an EGT correction factor. Came up with a (wave) frequency of 1414.399. Worked out the frequencies of differing RPM from 9K to 12.5K. My results were that the OEM muffler and original stock exit hole size corresponded to the tuned 'peak' of 9500 RPM.

Working out the numbers for my current modded larger (61%) hole, yielded a frequency of 2749.954.

What this means.......
The larger hole (modded) muffler is out of phase with the motor. The power peak is now past the maximum recommended RPM of 12500 of my 340e. The peak is most likely in the 18000+ range.

The original OEM un-modded muffler running stock max RPM, was (and is) better for my 340e.

(Let the flaming begin).

Using the data I have, it would be a simple matter to calculate, and also to move torque peaks up or down in RPM depending on the need. The optimum frequency for a specific RPM could be achieved, using several different ways such as: changing or adding stinger pipe, altering muffler volume, or area of muffler hole.

Will post some graphs to illustrate the #'s as soon as i figure out how to post them.

---------------------------------------------------------------------------------
The EGT temps used were from a graph made by someone named "timberwolf". (Note: the EGT temps were curiously low, and do hope that they are correct, but they were the only numbers available on this site).

The dyno sheets (in German) for other husqvarna models were also used as a reference to check torque/hp peaks. I cannot remember the poster, sorry. The sheets were found on this site as well.

 

[Terry Syd]

Well now, this could be an interesting thread.

First off, what formula are you using? With all the different internal designs and the various surfaces the waves reflect off, I don't understand how a box muffler could be tuned.

Some background of Timberwolf's work, he was using a modified 026. That saw had a raised exhaust port (and thus a fair amount of blowdown) and the variable muffler opening was on the front of the muffler. He indicated at one point that engines with less blowdown could likely benefit from muffler openings somewhat larger than his emperical findings. I believe he was correct on that point.

The 85% rule (balance of power to noise) was thus established using a modified saw, it wasn't developed on a stock saw, let alone on a saw with very low blowdown.

Further, he ran into idle instability when he hit around 150% of port area. The key point to note on the idle problems was that the variable opening of the muffler was in a direct line with the exhaust port. A muffler with openings off to the side will not have the same wave characteristics as his test muffler.

EDIT: The lack of idle stability may have also been related to a drop in trapping efficiency in the cylinder with the raised exhaust port. An engine with low blowdown will have better trapping efficiency than one with greater blowdown. Thus an engine with lower blowdown can handle larger muffler openings without running into idle instability.

 

[one.man.band]

Two graphs attached. Axis' are not labeled yet. As soon as time permits, i will clean them up.

graph #1 = frequency in Hz. is the vertical axis. horizontal axis is of no significance (shows the # of plotted values only).

the 8 trend lines are as follows:
top line = 12500 RPM
7th line = 12000 RPM
6th line = 11500 RPM
5th line = 11000 RPM
4th line = 10500 RPM
3rd line = 10000 RPM
2nd line = 9500 RPM
lowest line = 9000 RPM

take note of the trend line 2nd from bottom.

-----------------------------------------------------------
graph #2

same information as graph #1. Frequency is shown for on one combined trend line.

frequency in Hz. is the vertical axis. horizontal axis is of no significance (shows the # of plotted values only).

 

[one.man.band]

terry...

the only info that i needed from "timberwolf's" data graph was accurate EGT temp range of a saw. the EGT temps on his graph were low, compared to most EGT temps i have seen in other types of motors. most EGT's are a few hundred degrees F higher. curious if the temps were lower because of saw modification, overly rich condition or other factor's. there were about a hundred posts mentioning EGT on this site, but i found only two references that listed actual temp data #'s. the other post which measured EGT with data #'s listed was measured with temp probe outside of exhaust flow, (was mounted outside above EXH port, not actually in exhaust gas flow stream). if the EGT i used is not a representation of the actual true temp, then the calculations will be skewed and formula's will be off.

 

[Terry Syd]

I'm still trying to understand what you are measuring and why. As near as I can tell you are looking at the exhaust gas temps and then computing the change in the speed of sound. If that is correct, how is that relevant to how the muffler flows?

You refer to a tuning of the stock muffler at 9,500 rpm. How did you determine that the muffler was 'tuned' (in the sense that there was a formula for construction of the can muffler)?

 

[one.man.band]

explanation of data graphs:

where the original un-modded muffler frequency = 1414.399 Hz. Any area of the graph, horizontally below the 1414.399 Hz value listed on the vertical axis, show the frequencies based on harmonics that are readily reproducible by this muffler.

Until I fix the graphs for more clarity (which may be awhile). Place a sheet of paper across graph #1. Bottom edge of paper lined up horizontally with 1414.399Hz value on the left axis. All the frequencies from 150 Hz to 1414.399 Hz that the muffler can reproduce will be viewed.

The trend lines are frequency harmonics related to a specific engine RPM.

The modded muffler, with a frequency of 2749.954 Hz, does not correspond linearly with any trend lines shown. If the trend line were plotted for this frequency, it would be at about 45 degree angle.

The point of this is to match muffler frequency, to cover the highest RPM range necessary (or wanted), during operation of the saw.

Hope I am explaining this clearly enough at this point. I will add RPM ranges from 13000 on up.

 

[mtngun]

Your graph is meaningless -- no units, no reference to how you came up with the data.

Did you do a before and after cutting speed test ? If not, then we have nothing to go by.

It sounds like you didn't PERCEIVE a difference in cutting speed and you don't like the increased noise. Fine and dandy. But lots of luck telling us that the muffler isn't tuned right. Box mufflers are not tuned like a pipe.

 

[mtngun]

where the original un-modded muffler frequency = 1414.399 Hz. Any area of the graph, horizontally below the 1414.399 Hz value listed on the vertical axis, show the frequencies based on harmonics that are readily reproducible by this muffler.

Until I fix the graphs for more clarity (which may be awhile). Place a sheet of paper across graph #1. Bottom edge of paper lined up horizontally with 1414.399Hz value on the left axis. All the frequencies from 150 Hz to 1414.399 Hz that the muffler can reproduce will be viewed.

The trend lines are frequency harmonics related to a specific engine RPM.

The modded muffler, with a frequency of 2749.954 Hz, does not correspond linearly with any trend lines shown. If the trend line were plotted for this frequency, it would be at about 45 degree angle.

The point of this is to match muffler frequency, to cover the highest RPM range necessary (or wanted), during operation of the saw.

I'm not aware of any formula for "muffler frequency," nor have you provided one. I think you are full of it.

 

[one.man.band]

to find the frequency of a volume other than a sphere, elaborate calculus is necessary. there are a few research papers out there that prove it can be done mathematically. but why?

take a speaker cabinet. rectangular. ported. tuned. throw in a temp correction, port end corrections, basic wave equation to get frequency, wavelength, period (time in milliseconds), then muffler volume, etc. on motor side of things, rpm as to frequency, taking into account engine displacement volume (effective bore and stroke).

approximation using math is what the goal is, to get you close. even the tuned pipe calculations, no matter which one's you use still require fooling around with lengths before welding up for good.

 

[one.man.band]

full of it yes.

can you offer, or have a better or any mathematical explanation?

all ears.

 

[Terry Syd]

OK, I think I see where you're going with this, you are considering the can muffler as some sort of Helmholz resonator. A few members on this forum might recognise the concept as some sort of 'boost bottle'.

Never worked with a can muffler as a resonator, but I've done intake work. I gave up on the differential calculus equations (the math was correct, but there were too many variables). Instead, I went to a straight hose that I could block off at different points to change the volume. Worked a treat and I could tune it exactly the way I wanted.

For the exhaust on a saw, if you're looking for power, the less backpressure the better. The noise can be a problem with a poorly designed muffler mod. I am running in excess of 150% of port area, but I have brazed an internal baffle into the muffler so that the sonic wave has to bounce off of at least two surfaces before exiting.

Maybe you will come up with something, heck, I can't count the number of cul-de-sacs I've been down when it comes to tuning.

 

[one.man.band]

...standard tuned pipe calculations find the reflective point wave length of a cylindrical pipe. can this calculation work for a rectangular muffler? i for one, do not think it can apply.

waves are measured in frequency, amplitude and length the last i knew.

 

[Log Hogger]

Mtngun nailed it.

If your X axis doesn't represent anything, then your data needs to be presented in some other way if you want others to interpret it. It would also help if you gave the equations you're using.

I've never found - and I have searched extensively - a single report on AS or any other chainsaw mod site of an enlarged muffler opening causing a decrease in power. The simple can mufflers on most chainsaws are simply designed to reduce sound and increase back pressure to pass emissions standards, with a screen added to keep sparks in and bugs out. On some stratocharged saws, opening them doesn't give much if any increase in performance, but that's probably due to their scavenging design.

The 555/562XP muffler is intriguing in being so large and one AS member said it weighed nearly a pound. Husqvarna must have had a good reason to use a big, heavy, and costlier muffler when when simpler designs work just as well. Maybe there's some sort of pressure wave tuning going on there, but we won't know until someone tests one. It'd be interesting to make a few 562XP test cut comparisons with the stock muff and with a muff off the 357XP.

 

[one.man.band]

wave frequency is where i think it's at. opinions will and do vary. so what.

dia. of muffler hole: larger opening gives higher frequency, since air can rush in and out faster.

volume of muffler: larger muffler gives lower frequency, since more air must move out to relieve a given pressure excess.

stinger length: longer length gives lower frequency, since there is more resistance to air moving in and out.

opposite of all above is true.

------------------------------------------------------------

theoretically: in my muffler (whose volume is 368.71cc), it takes 9.04 piston revolutions displacing 40.8cc to fill the muffler. therefore, 9.04 frequency pulses of exhaust gases. the 9.04 frequencies in relation to the piston speed is what is important in my thinking. rpm can be converted into frequency. frequency to period which = a time event.

example: at 9500rpm
the frequency is 150Hz.
since it takes 9.04 piston revolutions to fill the muffler......

158.33Hz
316.66Hz
474.99Hz
633.32Hz
791.65Hz
949.98Hz
1108.31Hz
1266.64Hz
1424.97Hz

using EGT temp correction for speed of sound......muffler volume frequency = 1414.399Hz

thats how.

 

[wyk]

Computational fluid dynamics - Wikipedia, the free encyclopedia

 

[one.man.band]

the x-axis is 9 data frequencies for each RPM: 9 for 9k, 9 for 9.5k, 9 for 10k, 9 for 10.5 k, 9 for 11k, 9 for 11.5k, 9 for 12k and 9 for 12.5k. = total 72 points.

if there was not a x-axis value assigned, (in this case #1 through #72), the graph would be one big dot.
-----------------------------------------------------------------------------------------------------------------------
proof is in the pudding.

results from actual saw cutting would tell the tale, one way or the other.

if i am wrong in my theory, then so be it.

 

[naturelover]

opcorn:

 

[Terry Syd]

"Husqvarna must have had a good reason to use a big, heavy, and costlier muffler when when simpler designs work just as well"

I expect it is a way of dropping the pressure in the muffler without increasing the size of the outlet - double the volume and you have half the pressure.

I looked at the pictures of the Husky muffler and thought that someone should make an accessory muffler for that saw and reduce the PHO weight. If the new 550 is 10.8lbs with a humungous muffler, I wonder what it would weigh with a smaller muffler.

 

[rbmopar]

I comend you for your effort, but I am afraid you are chasing your tail with this line of thinking. Even if you hit your magical frequency, what is going to happen? It is not going to reflect gases back into the port like a pipe. If you want to get exhaust gases out of a box muffler faster, up to a certain point you make a bigger hole. When the hole is too big and your saw wont hold a stable idle, and you are deaf, you will know. Maybe you are on to something, but I don't see it. Good luck anyhow.

 

[one.man.band]

I comend you for your effort, but I am afraid you are chasing your tail with this line of thinking. Even if you hit your magical frequency, what is going to happen? It is not going to reflect gases back into the port like a pipe. If you want to get exhaust gases out of a box muffler faster, up to a certain point you make a bigger hole. When the hole is too big and your saw wont hold a stable idle, and you are deaf, you will know. Maybe you are on to something, but I don't see it. Good luck anyhow.

i too realize that realize that a box muffler will never reflect the a single wave frequency back into the exh port with the authority of tuned pipe. along with that, the box muffler will not have the nearly the same effect as 'being on the pipe'. there is not a current method for a way to calculate effectively the period of the natural frequency of the box muffler to be made equal to the period of the scavenging process that i know of.

my goal is trying to maximize the effect of the box muffler has, to reflect back to the port the as many proper harmonic frequencies wave lengths that can be had.

a rectangular box reflects all different wavelengths. (speaker cabinets that produce freq from 20 to 20k Hz are never shaped like a true cube, or a sphere). but with some calculations, the driver wave can be synced with the internal volume of the cabinet, making it more efficient.

there are many variables that play a role in solving this tuning a box muffler completely. trying to pick out the major solvable players and leaving things unresolved such as: crankcase pressure effect, lack of blow-down event time at very high speed piston speeds, varying exhaust pressure (160 or more psi down to ?psi), rate of exhaust gas cooling, mass of a combusted gas/oil/air mixture, and many more....... due to my inability to obtain or even estimate these factors.

hope there is another way to skin the cat.