Maybe I should suggest turning it off to the ceo. Actually, it must run all the time. Its demand cannot wait for a sound control system to turn off. At anyrate, quiet would be if outside in an open field if a person stands more than 50 feet from it they should not be able to hear it.
A lot of effort is justifiable? Are vast quantities of space and money also available? Both will be needed.
Constant speed? 2 exhaust channels with lengths calculated to cancel each other out would be useful to kill some of the sting. Engines make a lot of intake "honk" so that area requires attention too. Look at the complicated air boxes on quiet but far-from-silent snow mobiles and jet skis.
A BIG box with stiff walls, lined with 12 inch thick fluff, with twisty paths for intake and exhaust, also lined with fluff, could make a big reduction.
C) buy some bricks and cement. Build a series of enclosures around the engine.
50 feet is say 20 m
limit of audibility is say 20 dBA (it isn't)
typical SPL of an engine is what 90? dBA at 1 m
so at 20m it'll be around 65 dBA
So you have a 45 dBA task ahead of you. The first 10 is easy. The next 10 requires great skill and experience and money. The last 25 dBA require infrastructure.
I think you need a much better definition of your 'inaudibility' requirement.
I don't know what brand of engines Sterling uses in their trucks, but they don't seem particularly quiet.
The Stirling engine is fairly quiet, but nobody sells them commercially. Okay, you can buy a nice toy one that uses air as the working fluid. One of my friends did. It uses fuel at about the same rate as a small model airplane engine, and produces enough power to run ... itself, provided you use really thin oil on the bearings. Not ready for prime time now, and probably never.
Fuel cells have a better chance of someday being really available.
In the meantime, it is possible to make a two- stroke pretty quiet, using techniques that have been outlined here. There may be more effective strategies used in certain military outboard motors, or maybe they just sweat the details more.
Greg, he didn't really say it was air-cooled!
And of course 50 feet is almost exactly 15m as anyone who'd done passby noise testing should know - most current test procedures being based on either adaptation of, or defiance of, SAE J986!
"Tmoose" pretty well has it pegged, concerning the total enclosure with heavy barrier walls and lined with deep absorber, but if it's a constant-speed engine there are also Helmholtz and/or quarter-wave tuners that will cancel a big part of intake and exhaust noise. After that, well: does it have to be mobile?
Because if you're talking about say an 18hp 2-stroke for any conceivable vehicle, you're out of luck. It can't be done in any practical fashion.
If you're stuck with one engine, I think you need to address two issues -vibration transmission and noise absorption. Vibration isolation could reduce the transmission of vibration to the surroundings and an effective barrier could reduce the airborne noise.
If this is a stationary application, low frequency springs could be affective. An inertia base might be helpful. If it's a mobile application, low frequency springs may not work, but focalized elastomer mounts may be beneficial. In either case, an acoustic barrier would also be necessary to handle the airborne noise.
You need to speak to IAC industrial acoustics ltd USA.
They make acoustic "kennels". First you need aa reference sound to determine the noise attenuation. Is a a residential area or a rural countryside type area. What is the existing sound level?
You could get some very useful info from TRANE chillers ad well. They publish sound attenuation data and calculations for their range of chillers.
In ther UK we would either use the Trane guide to assess the existing sound level or ASHRAE. Then you need the SWL sound power level of the engine. From this you can assess the sound reduction index needed for the walls. I presume you will also need combustion air, so acoustic louvres will be needed??