Generator Compartment Build

[buswarrior]

It is all in the space you build into the corners. Every dimension of every corner has to be chosen respecting the cross section/area that is required to allow enough air to pass by.

You don't want to reduce your 6 lane road to a single lane going around the corner...

The whole thing, fans included, have to be inside of the switchbacks. Fan noise is a shocking amount of the noise coming from a generator.

Happy coaching!
Buswarrior

[Jcparmley]

As far as the design, what are your thoughts?  Should the radiator fan pull from the below or from the side?

[Jim Blackwood]

JC if you are going to the trouble of those switchbacks it isn't much more trouble to add tuned quarter wave cavities and those can have a profound effect on sound levels. Lincoln uses them in their engine intakes. All you need to know is how big to make them, and to nail it down you should make friends with an audiophile. Often the sound man at a church is such a person. He may have a meter which samples the sound spectrum and can tell you the frequencies that are loudest. From the center frequency you can determine the quarter wave length and from that include trap(s) in your design. If you cancel out just the loudest couple of bands the rest should be fairly easy.

Jim

[chessie4905]

Line an intake box with lead or use a car muffler. Go look at a Prevost xl Marathon coach and see how they did it. The compartment intake goes through some sort of baffle on end of compartment and exhaust is directed into a cast iron circular chamber.

[Jcparmley]

Hi Jim

I'm not sure what a "tuned quarter wave cavities" are.  Could you be more specific?    The suggestion on using a meter is a great one.  Thanks

JC if you are going to the trouble of those switchbacks it isn't much more trouble to add tuned quarter wave cavities and those can have a profound effect on sound levels. Lincoln uses them in their engine intakes. All you need to know is how big to make them, and to nail it down you should make friends with an audiophile. Often the sound man at a church is such a person. He may have a meter which samples the sound spectrum and can tell you the frequencies that are loudest. From the center frequency you can determine the quarter wave length and from that include trap(s) in your design. If you cancel out just the loudest couple of bands the rest should be fairly easy.

Jim

[Jim Blackwood]

JC, I think I explained it in an earlier post maybe in this thread but  it's the old idea that opposing waves will cancell each other out. So what you do is make a dead end that is 1/4 wave in length at the point where the air path makes a bend. In other words it is an extension beyond the air path. The dead end reflects the wave back and by the time it gets back to the junction it is 180 degrees out of phase and cancels the incoming wave.

Now if it worked perfectly it would completely cancel every other wave in theory at least. Even that is going to do a lot more than just square corners. But I suspect that with some damping it might do much better.

Jim

[Jim Blackwood]

So let's think about this. If you have a pulse and you use it to cancel the next one what does that do to the first pulse? Doesn't it cancel it too? Seems it would. Probably depends on how the exit feed was constructed. I don't know all the answers here but certainly someone does and I'm sure you can find it on the web. I do personally know at least one guy who built wave traps to get rid of a drone in his exhaust and said it worked very well. He simply attached it using a "T"and a 90 so it ran along the pipe. Personally I think I would use a tall narrow slot right at the cancellation point with enough throat area to handle the airflow without it having to speed up too much. It might be an advantage to angle it towards the reflecting wall or it might not.

Now one thing you can do is play different frequency tones from a google search and it should be fairly easy to match the sounds from your genset. My guess, it'll be somewhere not too far from 200 hz that you hear the loudest noise. The wavelength of 200 hz is 150 cm or 59", so 1/4 wave for 200 hz would be 14-3/4". Well within reach for this construction.

BTW, all this info is an easy google search away.

Jim

[Jcparmley]

So I think I understand better.  Basically I would measure the generators frequency and find out at what hertz it's the loudest.  Then I run the equation to see how many inches is the quarter wave.  Then I build the box with the switchback however many inches the quarter wave equation dictates.  Line the switchback with some acoustical foam and in theory it should cancel out most to the sound but not restrict air flow.  Am I understanding the concept, or am I way off?

I found this pic that I thought was interesting how the do the air intake and exit.  What are your thoughts?

So let's think about this. If you have a pulse and you use it to cancel the next one what does that do to the first pulse? Doesn't it cancel it too? Seems it would. Probably depends on how the exit feed was constructed. I don't know all the answers here but certainly someone does and I'm sure you can find it on the web. I do personally know at least one guy who built wave traps to get rid of a drone in his exhaust and said it worked very well. He simply attached it using a "T"and a 90 so it ran along the pipe. Personally I think I would use a tall narrow slot right at the cancellation point with enough throat area to handle the airflow without it having to speed up too much. It might be an advantage to angle it towards the reflecting wall or it might not.

Now one thing you can do is play different frequency tones from a google search and it should be fairly easy to match the sounds from your genset. My guess, it'll be somewhere not too far from 200 hz that you hear the loudest noise. The wavelength of 200 hz is 150 cm or 59", so 1/4 wave for 200 hz would be 14-3/4". Well within reach for this construction.

BTW, all this info is an easy google search away.

Jim

[Jcparmley]

Is that a correct description?

So I think I understand better.  Basically I would measure the generators frequency and find out at what hertz it's the loudest.  Then I run the equation to see how many inches is the quarter wave.  Then I build the box with the switchback however many inches the quarter wave equation dictates.  Line the switchback with some acoustical foam and in theory it should cancel out most to the sound but not restrict air flow.  Am I understanding the concept, or am I way off?

I found this pic that I thought was interesting how the do the air intake and exit.  What are your thoughts?

[Jim Blackwood]

Sounds to me like it ought to work.

Jim

[daveola]

Does your setup kill most of the noise?  What do you mean when you suggest you would do something different with the exhaust?  You also mention the intake.  Is it too small? Not enough air flow?  Does your gender overheat?

Sorry that I didn't see this until now, so let me answer so it's at least saved for perpuitity.

As far as my gender overheating or being too small, I think you'd have to ask my wife about that. 

The doghouse does kill quite a bit of the noise, though I've never measured it, and/or compared it to closing the bay door without the doghouse, since I've never had it installed as such.

I will say that it's considerably quieter, and is comparable to a car engine in terms of noise when it's all buttoned up.

As far as the exhaust - I would put on a silencing muffler, and probably run it all the way to the back of the bus (to get the exhaust away from the side windows).

I'd also move the air intake and air flow exhaust further apart.  I was making some compromises to make it fit, and it was a mistake.  I deal with it now by putting a flap between the two every time I go park out in the desert, which is every year, to make sure it doesn't suck in too much of the hot air flowing out.  Climbing under the bus twice a year isn't a huge cost, but it would be nice if it had been done better.  The genset also comes sitting on two small airbags courtesy Wrico.