Exhaust drone with engine swaps - solved
04-08-2014, 11:00 AM
#1
Exhaust drone with engine swaps - solved Here is an accurate method for significantly reducing exhaust drone using simple math calculations and just a little fab work. The process applies to every gas and diesel engine. My example is for a V8, but the math is slightly different with a 4 or 6 or diesel cylinder engine. Drone can be a huge annoyance especially after completing your swap only to find that it has a horrible interior noise at cruise speeds.
Drone has been discussed to death on other forums. However most pople just guess on how to fix the drone and dont analyze it mathmatically. The problem can be explained and fixed using simple terms using proveable calculations.
First and foremost, I wanted to eliminate drone without introducing any flow restriction; so here is a little theory and my solution.
Exhaust gas consists of 3 major components; Heat, Pressure and Sound. I will look at each portion in turn.
First, Heat. We can partially ignore this as it has little effect on the exhaust sound. It does change tone slightly with temperature variations. Exhaust temperature comes into play with the math calculations shown later.
Exhaust pressure pulses can be thought of as rolling tennis ***** down a pipe, one at a time. Each ball represents a cylinder pulse (bang) as each cylinder is fired. Between each pressure pulses, there is less pressure but still higher presure than atmosphere. When the exaust valve opens and a high pressure pulse pushes the previous pulse (ball) down the pipe just like cars driving down a freeway, all following each other in line. Each pulse actually scavenges or helps pull the next pulse from the exhaust port.
Next is Sound. On a small block Chevy the firing order is 18436572. The rythm applies to LS1 too but with a slightly different firing order. So, if you would stand behind the car and listen to each pressure pulse in turn at the end of both tail pipes, you hear the driver and passenger side pulses in the the following order. D represents the Driver side pulse or bang and P is the Passnger side pulse. You would hear this: DPPD, and then PDDP. Note: the rythm is opposite of each other but are all part of the 8 cylinders firing. Think of it as two 4 cylinder connected together with opposite firing orders. So, what you hear is one Driver pulse followed by two Passender pulses and then another Driver pulse. Then, one Passenger pulse and two Driver pulses then lastly a Passenger pulse. Thats the rumpidy rump we hear and love. It all repeats the cycle beginning with cylinder number 1.
Now you understand the rythm and what it consists of, we can move on to the next part.
To reduce and hopefully eliminate the drone, we need to understand what causes it and then counteract it.
The drone or resonance you hear inside the vehicle varies with RPM (exhaust Frequency), load on the chassis and exhaust airflow. Other variables can increase drone like tail pipe length and size.
Generally at cruising speeds, the noise (drone) you hear occurs around 2000 RPM.
To understand how drone is created, think of this. If you blow at a 90 degree angle across a coke bottle or milk jug, it will resonate, both at a different frequencies based on the container volume and how fast you blow across it. This is how a pipe organ functions. Different length pipes resonate at different frequencies. So with little air flow, the organ is quiet but with large amounts of airflow the organ gets really loud. When you blow across the bottle the sound waves go into the bottle and hit the back side and bounce back at the opening and either amplify or cancel each other out. This concept is the same as you jumping in a swimming pool. The water wave moves away from you, bounces off the opposite side of the pool and is reflected back at you.
This youtube video shows how it works.
In the exhaust system the pressure pulse (bang) needs something for it to bounce off of and cancel the next pressure wave out. To do this we need to add a resonant chamber like the coke bottle. This is actually pretty easy to build. First the chamber needs to be attached to the exhaust system at a 90 degree angle and be capped at the other end to reflect the sound back at the next exhaust pulse just like the coke bottle.
Here is when math comes into play. We need to determine the frequenct of the exhaust pulses to then determine how long the resonator needs to be to cancel the drone. If the exhaust drones at 2000 RPM on a V8 the exhaust pulse frequency is about 133 pulses per second or 133hz. If you want to figure it out yourself at various RPM, the formula is in the attached spreadsheet.
Looking at the spreadsheet chart, you can see the length of the resoantor needs to be approx 26 inches long. Most cars dont have room to add a capped pipe that long under our cars especially since it needs to be attached at a 90 degree angle to the exhaust system. The solution is to use a mandrel bent 'J' pipe. Simply cut a hole in the side of your exhaust and weld the end of the 'J" to it and lay the rest of it nest to your muffler or whereever it fits.
I will attach a picture/chart of the resonant RPM and length of resonator needed to greatly reduce the exhaust drone.
It really does work and cancels tons of resonance.
I forgot the chart at home but will edit and add it tonight.
Drone has been discussed to death on other forums. However most pople just guess on how to fix the drone and dont analyze it mathmatically. The problem can be explained and fixed using simple terms using proveable calculations.
First and foremost, I wanted to eliminate drone without introducing any flow restriction; so here is a little theory and my solution.
Exhaust gas consists of 3 major components; Heat, Pressure and Sound. I will look at each portion in turn.
First, Heat. We can partially ignore this as it has little effect on the exhaust sound. It does change tone slightly with temperature variations. Exhaust temperature comes into play with the math calculations shown later.
Exhaust pressure pulses can be thought of as rolling tennis ***** down a pipe, one at a time. Each ball represents a cylinder pulse (bang) as each cylinder is fired. Between each pressure pulses, there is less pressure but still higher presure than atmosphere. When the exaust valve opens and a high pressure pulse pushes the previous pulse (ball) down the pipe just like cars driving down a freeway, all following each other in line. Each pulse actually scavenges or helps pull the next pulse from the exhaust port.
Next is Sound. On a small block Chevy the firing order is 18436572. The rythm applies to LS1 too but with a slightly different firing order. So, if you would stand behind the car and listen to each pressure pulse in turn at the end of both tail pipes, you hear the driver and passenger side pulses in the the following order. D represents the Driver side pulse or bang and P is the Passnger side pulse. You would hear this: DPPD, and then PDDP. Note: the rythm is opposite of each other but are all part of the 8 cylinders firing. Think of it as two 4 cylinder connected together with opposite firing orders. So, what you hear is one Driver pulse followed by two Passender pulses and then another Driver pulse. Then, one Passenger pulse and two Driver pulses then lastly a Passenger pulse. Thats the rumpidy rump we hear and love. It all repeats the cycle beginning with cylinder number 1.
Now you understand the rythm and what it consists of, we can move on to the next part.
To reduce and hopefully eliminate the drone, we need to understand what causes it and then counteract it.
The drone or resonance you hear inside the vehicle varies with RPM (exhaust Frequency), load on the chassis and exhaust airflow. Other variables can increase drone like tail pipe length and size.
Generally at cruising speeds, the noise (drone) you hear occurs around 2000 RPM.
To understand how drone is created, think of this. If you blow at a 90 degree angle across a coke bottle or milk jug, it will resonate, both at a different frequencies based on the container volume and how fast you blow across it. This is how a pipe organ functions. Different length pipes resonate at different frequencies. So with little air flow, the organ is quiet but with large amounts of airflow the organ gets really loud. When you blow across the bottle the sound waves go into the bottle and hit the back side and bounce back at the opening and either amplify or cancel each other out. This concept is the same as you jumping in a swimming pool. The water wave moves away from you, bounces off the opposite side of the pool and is reflected back at you.
This youtube video shows how it works.
In the exhaust system the pressure pulse (bang) needs something for it to bounce off of and cancel the next pressure wave out. To do this we need to add a resonant chamber like the coke bottle. This is actually pretty easy to build. First the chamber needs to be attached to the exhaust system at a 90 degree angle and be capped at the other end to reflect the sound back at the next exhaust pulse just like the coke bottle.
Here is when math comes into play. We need to determine the frequenct of the exhaust pulses to then determine how long the resonator needs to be to cancel the drone. If the exhaust drones at 2000 RPM on a V8 the exhaust pulse frequency is about 133 pulses per second or 133hz. If you want to figure it out yourself at various RPM, the formula is in the attached spreadsheet.
Looking at the spreadsheet chart, you can see the length of the resoantor needs to be approx 26 inches long. Most cars dont have room to add a capped pipe that long under our cars especially since it needs to be attached at a 90 degree angle to the exhaust system. The solution is to use a mandrel bent 'J' pipe. Simply cut a hole in the side of your exhaust and weld the end of the 'J" to it and lay the rest of it nest to your muffler or whereever it fits.
I will attach a picture/chart of the resonant RPM and length of resonator needed to greatly reduce the exhaust drone.
It really does work and cancels tons of resonance.
I forgot the chart at home but will edit and add it tonight.
Last edited by aknovaman; 04-09-2014 at 09:23 AM.
04-08-2014, 11:13 AM
#2
I did the calculations about a year ago, and got the same answer (26") for a 2000RPM drone on my car. I will make the (Helmholtz) chamber the next time I drop my motor/trans. Your spreadsheet would be helpful when you post it.
Thanks,
Andy1
Thanks,
Andy1
04-08-2014, 01:22 PM
#4
Won't that negate the whole purpose of long tube headers? You want the resonate scavenging effect of the alternate pulses.
Also wouldn't the length need to vary with the diameter of the pipe? Plus the speed of sound varies with the pressure, density and humidity of the medium, so do you account for that in your calculations?
Also wouldn't the length need to vary with the diameter of the pipe? Plus the speed of sound varies with the pressure, density and humidity of the medium, so do you account for that in your calculations?
04-08-2014, 01:56 PM
#5
Won't that negate the whole purpose of long tube headers? You want the resonate scavenging effect of the alternate pulses.
Also wouldn't the length need to vary with the diameter of the pipe? Plus the speed of sound varies with the pressure, density and humidity of the medium, so do you account for that in your calculations?
Also wouldn't the length need to vary with the diameter of the pipe? Plus the speed of sound varies with the pressure, density and humidity of the medium, so do you account for that in your calculations?
Great research AKnova, thank you for sharing. So if drone elimination is a function of cancelling the pressure waves, is that why the addition of an H or X pipe can help to eliminate dorn? By bringing the left and right bank together you are causing opposing pressure waves to collide and dissapate before exiting the exhaust?
04-08-2014, 02:02 PM
#6
Resonation and scavenging are two different things. What you want is the initial exhaust pulse to move down the pipe and pull the next in firing order out of the header.
Yes the pressure, density and humidity of the medium are present but negligible in this exercise. My chart and graph shows what is needed at 120 degree pipe. I did enter other extreme temperature figures but it did not significantly change the resonant frequency.
The j-pipes are added after the pulses are inline in a single pipe like just before the muffler. The resonator chamber needs placed as shown here.
http://upload.wikimedia.org/wikipedi...HR_analogy.JPG
Here is crude drawing I found.
http://www.f150online.com/forums/mem...onantpipes.jpg
X and H pipes work a little differently as the pulses are not occurring (present) at the same time within the crossover pipe. For an x -pipe config, think of it as figure 8 racing with school busses. Each cylinder pulse (bus) crosses through the x-pipe one after the other, in line. The exhaust pressure pulse is shared by both mufflers instead of only one muffler. Contrary to popular belief and simple if you think of it. Exhaust does not stop, turn 90 degrees and again another 90 degrees and then exit toward the muffler in a h-pipe config. The pressure pulse 'sharing' is what smoothes it all out. Thats why you dont have the popp popp out of each tailpipe and it shounds smoother. Mustangs in the '90's had x-pipes on them from the factory. If guys would put true duals on them they would sound like poop. For some reason people think 'true duals' sound cool but actually they do not capitalize on the exhaust scavenging. Honda's sound smoother as the pulses are inline and not erratic (uneven) like a GM V8.
I will try to draw up some pics of a h-pipe and expand on that too. Its just difficult to explain in words.
Yes the pressure, density and humidity of the medium are present but negligible in this exercise. My chart and graph shows what is needed at 120 degree pipe. I did enter other extreme temperature figures but it did not significantly change the resonant frequency.
The j-pipes are added after the pulses are inline in a single pipe like just before the muffler. The resonator chamber needs placed as shown here.
http://upload.wikimedia.org/wikipedi...HR_analogy.JPG
Here is crude drawing I found.
http://www.f150online.com/forums/mem...onantpipes.jpg
X and H pipes work a little differently as the pulses are not occurring (present) at the same time within the crossover pipe. For an x -pipe config, think of it as figure 8 racing with school busses. Each cylinder pulse (bus) crosses through the x-pipe one after the other, in line. The exhaust pressure pulse is shared by both mufflers instead of only one muffler. Contrary to popular belief and simple if you think of it. Exhaust does not stop, turn 90 degrees and again another 90 degrees and then exit toward the muffler in a h-pipe config. The pressure pulse 'sharing' is what smoothes it all out. Thats why you dont have the popp popp out of each tailpipe and it shounds smoother. Mustangs in the '90's had x-pipes on them from the factory. If guys would put true duals on them they would sound like poop. For some reason people think 'true duals' sound cool but actually they do not capitalize on the exhaust scavenging. Honda's sound smoother as the pulses are inline and not erratic (uneven) like a GM V8.
I will try to draw up some pics of a h-pipe and expand on that too. Its just difficult to explain in words.
Last edited by aknovaman; 04-08-2014 at 08:29 PM.
Trending Topics
04-08-2014, 02:52 PM
#8
OK, so why does drone only occur near specific RPM ranges, like that magic 2000RPM that everyone complains about? Why don't we hear it from idle to max RPM? In your spreadsheet, depending upon RPM, the Hz vary from 60Hz to 170Hz, all well within the human hearing range (approx 20Hz to 20kHz). Why is it only an issue at SOME RPM?
I'm not trying to punch holes in your theory, I am just simply curious to learn more.
Also, do you plan to implement or test your resonate pipe design? I would be VERY interested in your results!
I'm not trying to punch holes in your theory, I am just simply curious to learn more.
Also, do you plan to implement or test your resonate pipe design? I would be VERY interested in your results!
04-08-2014, 03:05 PM
#9
Go back to the pipe organ analogy; a whisle is tuned for a particular tone or pitch. The flow and rate the exhaust pulses exit the tail pipe resonate at different frequencies. The rpm (freq) and size of the pipes make it sound like poop.
Last edited by aknovaman; 04-08-2014 at 03:12 PM.
04-08-2014, 05:04 PM
#10
FYI if you look at a cutaway of a stock 2010+ Camaro muffler, it actually incorporates a J-pipe resonator. The same principles are applied to reduce intake noise - factory intake ducting almost always includes a few resonators tuned to reduce noise at certain frequencies. You can actually download some basic frequency analyzers on your smartphone to determine where drone (resonant frequencies) is occurring so you can design the appropriate resonator.
04-08-2014, 05:55 PM
#11
TECH Enthusiast
This is a subject many of us would benefit to understand better. Nothing new, but it needs more discussion just to maybe one day get to the level of the average person that has an exhaust shop. Mentioned it to several local shops before, all were clueless & I hate having a person I may want to pay for a service knowing less than me.
Camaro 5 forums have a lot of threads on this & I think it was Solo Performance that was mentioned as an aftermarket source for premade J-pipe kits with an adjustable end cap for tuning.
For those skeptical of having a performance loss,keep in mind that sound ONLY goes in the j-tube & bounces back, not air flow. Soooo, the only performance loss is a few pounds more weight. Also as I understand it , the actual j shape of the pipe isn't important, just the length & coming off the exhaust pipe at a 90 degree angle.
The part I get lost with right now is the figuring formulas. When I get my next exhaust, I may just pick a similar ls1 used length like 26" with adjustable cap, test & go from there.
UPDATE::: see pic in link
http://cfster.smugmug.com/Cars/Resto...1060753-X2.jpg
Camaro 5 forums have a lot of threads on this & I think it was Solo Performance that was mentioned as an aftermarket source for premade J-pipe kits with an adjustable end cap for tuning.
For those skeptical of having a performance loss,keep in mind that sound ONLY goes in the j-tube & bounces back, not air flow. Soooo, the only performance loss is a few pounds more weight. Also as I understand it , the actual j shape of the pipe isn't important, just the length & coming off the exhaust pipe at a 90 degree angle.
The part I get lost with right now is the figuring formulas. When I get my next exhaust, I may just pick a similar ls1 used length like 26" with adjustable cap, test & go from there.
UPDATE::: see pic in link
http://cfster.smugmug.com/Cars/Resto...1060753-X2.jpg
Last edited by jlcustomz; 04-09-2014 at 06:14 PM.
04-08-2014, 06:37 PM
#13
This is very interesting.
So I assume this additional chamber is of the same diameter of pipe as the main exhaust. Does running a single 3" pipe change anything? My swap has resonance right around the 2000 rpm mark with a single exhaust.
I guess you could add a section of pipe that is about 3-5" shorter, then use a v-band or bolted flange for the tube end and try different lengths until you don't have the drone at cruise speeds.
I think when the exhaust frequency matches the chassis frequency it causes the whole chassis to vibrate which crates the "drone" sound.
So I assume this additional chamber is of the same diameter of pipe as the main exhaust. Does running a single 3" pipe change anything? My swap has resonance right around the 2000 rpm mark with a single exhaust.
I guess you could add a section of pipe that is about 3-5" shorter, then use a v-band or bolted flange for the tube end and try different lengths until you don't have the drone at cruise speeds.
I think when the exhaust frequency matches the chassis frequency it causes the whole chassis to vibrate which crates the "drone" sound.
04-08-2014, 09:04 PM
#14
Here is my favority example of how the resonant chamber reduces or amplifies the noise.
The larger the resonator pipe, relative to the exhaust pipe, the more it can attenuate the drone.
This math is just to prove my experiences and give some solid evidence to support what works in reality.
The best way I have found is to actually measure the frequency using one of the free smart phone apps and build the tube based upon my calculations to that specific frequency.
I have attached my spreadsheet in zip format. you will need to download it and unzip it then open it in Excel.
Here is an in depth article with all the details.
http://www.enoisecontrol.com/related...rrier_wall.pdf
One last link to prove my data. This is a real world analysis and reduction of exhaust dronecomplete with charts and pretty graphs before and after.
http://mkiv.supras.org.nz/articles/exhaust_drone.htm
The larger the resonator pipe, relative to the exhaust pipe, the more it can attenuate the drone.
This math is just to prove my experiences and give some solid evidence to support what works in reality.
The best way I have found is to actually measure the frequency using one of the free smart phone apps and build the tube based upon my calculations to that specific frequency.
I have attached my spreadsheet in zip format. you will need to download it and unzip it then open it in Excel.
Here is an in depth article with all the details.
http://www.enoisecontrol.com/related...rrier_wall.pdf
One last link to prove my data. This is a real world analysis and reduction of exhaust dronecomplete with charts and pretty graphs before and after.
http://mkiv.supras.org.nz/articles/exhaust_drone.htm
Last edited by aknovaman; 04-09-2014 at 09:13 AM.
04-09-2014, 06:01 AM
#16
i was playing with you a bit earlier but i think the question above is valid.
Won't an active muffler affect this sort of thing? I know Mitsubishi used them on Evo's, I think a few other manufactures did also. Do they work on drone?
04-09-2014, 09:27 AM
#18
Wood,
Thats correct it is about cancelling noise not cancelling pressure. The reflected sound wave would be 180 degrees out of phase and thus cancel the wave. The pressure pulse still gets scavenged out of the header/manifold/cylinders in turn.
Plz watch the vids and it will assit you in understanding.
Thats correct it is about cancelling noise not cancelling pressure. The reflected sound wave would be 180 degrees out of phase and thus cancel the wave. The pressure pulse still gets scavenged out of the header/manifold/cylinders in turn.
Plz watch the vids and it will assit you in understanding.
04-09-2014, 10:49 AM
#19
When I investigated the drone issue about a year ago, Google was my friend, and
I found lots of information, discussion, and photos of J pipes installed (doesn't have to be a J pipe, but it does provide for neat packaging by parallelling you exhaust system). It's common to confuse pressure wave (or pulses) and scavenging with resonance or resonant frequency.
Nice spreadsheet, aknovaman!
Andys
I found lots of information, discussion, and photos of J pipes installed (doesn't have to be a J pipe, but it does provide for neat packaging by parallelling you exhaust system). It's common to confuse pressure wave (or pulses) and scavenging with resonance or resonant frequency.
Nice spreadsheet, aknovaman!
Andys
04-09-2014, 11:04 AM
#20
That is very kind of you, but I have been implementing adaptive sidelobe cancellation algorithms in radars for years. What I don't have much experience with is header pulses. It is my understanding that you can tune headers to work at different power bands by adjusting the length of the tubes. If your J pipe hooks near the collector then why wouldn't damping the resonance to take away the low pressure pulse affect the scavenging?