My problem is how to figure exactly how to balance a 12" rod 1/8" dia rotating at 20K-30K RPM. Andy "M. Adam Davis" on 04/12/2000 09:18:11 AM Please respond to pic microcontroller discussion list To: PICLIST@MITVMA.MIT.EDU cc: (bcc: Andrew Kunz/TDI_NOTES) Subject: Re: [OT] INVERTE THE WAVE Just had another thought. One thing you can do to null out the sound is to track everyone on the factory floor within earshot of the machine. Once you know their location to within a few inches (particularily the location of each head, or even each ear) then you can solve a series of simultaneous equations which will determine the phase of sound you should produce from a speaker near the machine. This equation will give the best noise dampening for each person, given that a few aren't in optimal positions. It occured to me as I was thinking about the program I am currently working on. A long rotating shaft will have different unbalances along its length. Balancing one point on the shaft affects the balance of all other points on the shaft, so you need to find out how much each point affects the others, and then solve a whole slew of simultaneous equatons to find the 'best' balance for every point on the shaft. Some people in the factory will be in optimal positions, but others will not. You can find the best phase to output which will place each person as close as possible to a nulled area of sound. Of course, this will help a teeny bit, but sound bouncing around the factory will also have to be nulled seperately. In a completely deadened room this would work beautifully. Make it a bit more complex: Space speakers above the factory floor every 20 feet or 4 meters. then solve a series of simultaneous equations which deadens sound even better. You could almost create a sound curtain, such that particular area in the factory is completely quiet, almost like being in an office, but it is open air. This solution would require several magnitudes greater computer power than the suggestion above, but you may not need to deaden the sound reflections as much, this system could handle them as well. Not for the faint of heart, the equations alone are mind bending, not to mention trying to fit them in a computer, nevermind a DSP... -Adam > WF wrote: > > Hi, > > A student is studyng how to invert some wave... > > He intends to NULL some produced sound...in the factory that he works have > some TEXTIL machines and intends to reduce the noise produced by > them...applyng a inverted wave... > > Does someone have some suggestions about this? Formula, site, etc... > > I think that is too hard! > > Miguel Wisintainer >