First of all, I have not even built my separator yet, due being temporarily fiscally challenged, so everything in this article is pure theorizing on my part and open to discussion and criticism.
But before improvements can be designed/invented, it would be a good idea to understand exactly what is going on inside the chamber, how it works and why it works. What follows are my thoughts on how it works, based on observation of your videos and my engineering background.
How does it work - There are two air rotations going on at the same time inside the chamber. The first is the obvious clockwise rotation (viewed from above) that we all know about. But there is a second axial vortex going on that as far as I am aware, no one on this forum has discussed. This second vortex, I will refer to as the ?axial vortex? in future, is key to the whole operation of the separator.
Take a look at the excellent video that Pitbull posted:
http://www.youtube.com/watch?v=QZSK88Hjl3A
A definite anti clockwise axial vortex is visible as the debris enters the chamber (2? 10?), but the flow appears more laminar when the chips are rotating around, but with a downward trend. This is because the visible particles are too heavy to be carried by the axial vortex. Here is a snap shot from the video, the arrow shows the axial vortex direction. The picture does not show the axial vortex, but if you view the video, it will be clear:
Explanation for the axial vortex - The central exhaust port is forcibly drawing air out of the chamber and this action sets the axial vortex in motion, by drawing in air from the bottom of the chamber. When this drawn air is replaced, the axial vortex is set in motion. Here is a diagram showing how I think it works:
Axial vortex function ? from the above diagram, it can be seen that the axial vortex air at the outer wall of the chamber is moving in a downward direction. This is driving the debris downwards towards the slot. Without the axial vortex, the debris would just continue to rotate around the chamber until pure chance put the particle in the ?dead zone? above the slot. Gravity is totally overpowered by the strength of the centrifugal forces acting on the particles and the velocity of the main rotation.
Here is another excellent video by Bulldog8:
http://www.youtube.com/watch?v=SswUX_keN1M&feature=mfu_in_order&list=UL
One minute into the video, the flow can be seen to be moving in an angled downward direction. Here is a snap shot taken from the video, the arrow indicates the particle flow direction.
Large particle movement ? we know already that the particles get thrown to the outer wall of the chamber and eventually exit through the slot. The reason for this is that the larger particles have more mass and are strongly affected by the centrifugal forces. They also have more inertia and don?t change direction very easily. The axial vortex imparts a small downwards force on the larger particles, but can not carry the particles, hence they do not swirl with the axial vortex.
Small particle movement ? having a very small mass, the smaller particles are less affected by the centrifugal forces and affected more by airflow direction. As the particle mass reduces, there comes a transition point, when the effects of airflow overcome the effects of the centrifugal forces. The finest particles are carried in the axial vortex and eventually caught by the exhaust flow and end up in the filter bag or embedded in the filter itself.
Dead zones ? these are areas in the chamber that airflow is reduced or inefficient. There are four main ?dead zones? in the chamber: directly under the exhaust duct mouth, top centre adjacent to the exhaust duct, top outer corner of the chamber and above the exit slot. Here is a diagram labeling the ?dead zones?:
The only ?dead zone? that serves any purpose is above the slot. When the particles enter this zone, the reduced flow allows the particle to decelerate and to drop through the slot, by the gravitational force. This completes the function cycle.
Improvements ? just about everyone who has adapted this design to their needs, has asked themselves the question, ?Can the design be improved?? Phil Thien came up with this novel design of efficiently removing most of the debris before the filter. He no doubt spent countless hours and lots of money perfecting the design. Even Phil would probably admit that there are improvements, but if the improvement does not reap tangible, quantifiable results, then why complicate the design for such a small gain. Keep things simple is design rule No1.
Here follows my thoughts and ideas, again up for discussion and criticism:
Drop slot ? I have never really understood why the drop slot does not extend to 360 degrees. Phil obviously did a lot of testing on this feature and established the best working solution and I accept that solution. The only vague explanation that I can come up with, is that the flow settles before the slot is introduced. The last thing you want is air blowing through the slot and disturbing the already filtered fines.
Dead zones ? the dead zones that do not serve any purpose could be ?faired? out. This would improve the flow and efficiency. Whether the improvement would be large enough to justify the not inconsiderable extra work, is in doubt, but in theory, it should provide improvements. Here is a diagram of what I had in mind:
Angular rather than curved fairings, not as good, but would still improve the flow, but still a very complex and time consuming modification. Input from a fluids engineer would be valuable here.
Exhaust height adjustment ? this idea could reduce the amount of fines that make it to the filter bag, but it will not eliminate completely. I could be wrong on this one, but I believe the rotation of the axial vortex could be slowed down by raising the height of the exhaust duct mouth. By slowing down the axial vortex, the particle size threshold will be reduced and smaller particles will make it to the slot dead zone and be collected. Here is a diagram of what I had in mind:
Again, I am sure Phil experimented with this adjustment and I will gladly bow to practical proven experience. I am not convinced that raising the exhaust mouth to the roof of the chamber would give the desired result, but maybe there is an optimum height that works best. I throw it out for discussion anyway.
Rotation speed ? rotation speed can be increased by reducing the cross sectional area of the chamber. This increase in speed would increase the centrifugal force and affect finer particles, lowering the threshold of escaping fines. This would also reduce airflow efficiency, but I am sure there is an optimum cross section for any given system. An adjustable false top would reveal the answer to this one, requiring very little extra work.
Filter removal ? seeing as the debris that makes it through the exhaust and into the filter bag are so few and so fine. Could these not be exhausted to the outside atmosphere? They are so fine and light that they would be diffused in the slightest of breeze and cause no problem. Maybe this is a definite NO, but just thought I would mention the idea. By removing the filter bag, airflow efficiency would be greatly improved, due to lack of back pressure.
Any more ideas? ? these are not the only ideas that I have had, but they have been discarded either because of reduced efficiency of airflow or they did not address the fines problem, which is the only problem that exists.
I invite you now to submit your own ideas, no matter how wacky they are, for discussion.
Dave
But before improvements can be designed/invented, it would be a good idea to understand exactly what is going on inside the chamber, how it works and why it works. What follows are my thoughts on how it works, based on observation of your videos and my engineering background.
How does it work - There are two air rotations going on at the same time inside the chamber. The first is the obvious clockwise rotation (viewed from above) that we all know about. But there is a second axial vortex going on that as far as I am aware, no one on this forum has discussed. This second vortex, I will refer to as the ?axial vortex? in future, is key to the whole operation of the separator.
Take a look at the excellent video that Pitbull posted:
http://www.youtube.com/watch?v=QZSK88Hjl3A
A definite anti clockwise axial vortex is visible as the debris enters the chamber (2? 10?), but the flow appears more laminar when the chips are rotating around, but with a downward trend. This is because the visible particles are too heavy to be carried by the axial vortex. Here is a snap shot from the video, the arrow shows the axial vortex direction. The picture does not show the axial vortex, but if you view the video, it will be clear:
Explanation for the axial vortex - The central exhaust port is forcibly drawing air out of the chamber and this action sets the axial vortex in motion, by drawing in air from the bottom of the chamber. When this drawn air is replaced, the axial vortex is set in motion. Here is a diagram showing how I think it works:
Axial vortex function ? from the above diagram, it can be seen that the axial vortex air at the outer wall of the chamber is moving in a downward direction. This is driving the debris downwards towards the slot. Without the axial vortex, the debris would just continue to rotate around the chamber until pure chance put the particle in the ?dead zone? above the slot. Gravity is totally overpowered by the strength of the centrifugal forces acting on the particles and the velocity of the main rotation.
Here is another excellent video by Bulldog8:
http://www.youtube.com/watch?v=SswUX_keN1M&feature=mfu_in_order&list=UL
One minute into the video, the flow can be seen to be moving in an angled downward direction. Here is a snap shot taken from the video, the arrow indicates the particle flow direction.
Large particle movement ? we know already that the particles get thrown to the outer wall of the chamber and eventually exit through the slot. The reason for this is that the larger particles have more mass and are strongly affected by the centrifugal forces. They also have more inertia and don?t change direction very easily. The axial vortex imparts a small downwards force on the larger particles, but can not carry the particles, hence they do not swirl with the axial vortex.
Small particle movement ? having a very small mass, the smaller particles are less affected by the centrifugal forces and affected more by airflow direction. As the particle mass reduces, there comes a transition point, when the effects of airflow overcome the effects of the centrifugal forces. The finest particles are carried in the axial vortex and eventually caught by the exhaust flow and end up in the filter bag or embedded in the filter itself.
Dead zones ? these are areas in the chamber that airflow is reduced or inefficient. There are four main ?dead zones? in the chamber: directly under the exhaust duct mouth, top centre adjacent to the exhaust duct, top outer corner of the chamber and above the exit slot. Here is a diagram labeling the ?dead zones?:
The only ?dead zone? that serves any purpose is above the slot. When the particles enter this zone, the reduced flow allows the particle to decelerate and to drop through the slot, by the gravitational force. This completes the function cycle.
Improvements ? just about everyone who has adapted this design to their needs, has asked themselves the question, ?Can the design be improved?? Phil Thien came up with this novel design of efficiently removing most of the debris before the filter. He no doubt spent countless hours and lots of money perfecting the design. Even Phil would probably admit that there are improvements, but if the improvement does not reap tangible, quantifiable results, then why complicate the design for such a small gain. Keep things simple is design rule No1.
Here follows my thoughts and ideas, again up for discussion and criticism:
Drop slot ? I have never really understood why the drop slot does not extend to 360 degrees. Phil obviously did a lot of testing on this feature and established the best working solution and I accept that solution. The only vague explanation that I can come up with, is that the flow settles before the slot is introduced. The last thing you want is air blowing through the slot and disturbing the already filtered fines.
Dead zones ? the dead zones that do not serve any purpose could be ?faired? out. This would improve the flow and efficiency. Whether the improvement would be large enough to justify the not inconsiderable extra work, is in doubt, but in theory, it should provide improvements. Here is a diagram of what I had in mind:
Angular rather than curved fairings, not as good, but would still improve the flow, but still a very complex and time consuming modification. Input from a fluids engineer would be valuable here.
Exhaust height adjustment ? this idea could reduce the amount of fines that make it to the filter bag, but it will not eliminate completely. I could be wrong on this one, but I believe the rotation of the axial vortex could be slowed down by raising the height of the exhaust duct mouth. By slowing down the axial vortex, the particle size threshold will be reduced and smaller particles will make it to the slot dead zone and be collected. Here is a diagram of what I had in mind:
Again, I am sure Phil experimented with this adjustment and I will gladly bow to practical proven experience. I am not convinced that raising the exhaust mouth to the roof of the chamber would give the desired result, but maybe there is an optimum height that works best. I throw it out for discussion anyway.
Rotation speed ? rotation speed can be increased by reducing the cross sectional area of the chamber. This increase in speed would increase the centrifugal force and affect finer particles, lowering the threshold of escaping fines. This would also reduce airflow efficiency, but I am sure there is an optimum cross section for any given system. An adjustable false top would reveal the answer to this one, requiring very little extra work.
Filter removal ? seeing as the debris that makes it through the exhaust and into the filter bag are so few and so fine. Could these not be exhausted to the outside atmosphere? They are so fine and light that they would be diffused in the slightest of breeze and cause no problem. Maybe this is a definite NO, but just thought I would mention the idea. By removing the filter bag, airflow efficiency would be greatly improved, due to lack of back pressure.
Any more ideas? ? these are not the only ideas that I have had, but they have been discarded either because of reduced efficiency of airflow or they did not address the fines problem, which is the only problem that exists.
I invite you now to submit your own ideas, no matter how wacky they are, for discussion.
Dave
