Optimum location for a bellmouth outlet

[retired2]

In a very recent thread Phil made the comment that the primary benefit of a 2H (double-high) separator is the reduction of waste by-pass that results from the outlet pipe being lower than the inlet pipe.  That jogged my memory about some interesting test data I collected using three different positions for the outlet pipe.

That data shows that a straight outlet pipe passed essentially the same amount of air regardless of position.  However, the bellmouth moved more air as it was moved closer to the baffle.  The increase is not huge, but it is large enough and consistent enough to suggest it is real rather than a test procedure aberration.

The question to be explored now is whether the optimum location for a bellmouth outlet pipe is lower than the conventional 2-1/2 times the pipe diameter in a 1H (single-high) separator?  And of course the follow-up question is where is the optimum location for a 2H unit?

Since my build now has the outlet pipe solidly anchored, it is very difficult for me to go back and do further testing.  So, this thread is for those of you who have a bellmouth build in progress, or have a bellmouth outlet pipe that is still adjustable.  Do some testing and share your results here!

Regards,
Retired2

[kayak]

Resurrecting an older post.  I am still in the drawing stages of my build.  Drawing, erasing, drawing again, etc.  I have a theory about the outlet height.  The numbers used in my theory are based on 6 inch diameter duct work, and on a baffle to top height twice the diameter of the duct, which in my case would make a 12 inch chamber height. 

First, imagine that you have your outlet pipe all the way down against the baffle so that no air flow could occur.  Now, imagine that you want to cut a rectangular window in the side of the bottom of the pipe to allow air flow.  My theory is that the area of this window needs to be at least the same, or larger, than the cross-sectional area of the duct.  With 6 inch duct, the cross-sectional area is 28.27 inches.  If, for example you cut a window 1.5 inches high and 9 inches long, the area of the window would be 13.5 square inches, which is less than the cross-sectional area of the duct, and would be choking the air flow.  Keeping the window at 1.5 inches high, but continuing to cut the window longer and longer, until you have completely cut the circumference of the duct, you will have effectively made a window 18.85 inches long (the circumference of the duct) by 1.5 inches high.  The area of this window, 18.85x1.5=28.275, is very close to the cross sectional area of the duct. 

So, my theory is that if you take the cross-sectional area of the duct (radius squared x pi), and divide it by the circumference of the duct (2 x pi x radius), that will give you the minimum dimension that you can put the duct to the baffle without choking the airflow.  That close to the baffle, I am pretty sure that you will need to be using a bellmouth outlet to minimize turbulence.  As far as airflow goes, I would think that you might see an improvement if you increase the height up to 10 percent, which in this case would be 1.65 inches high.  After a 10 percent increase, I don't know if you would see any more improvement in airflow.  All of this speaks to airflow only, not to turbulence, by-pass, or scrubbing.  But this is going to be my starting point.

Along these lines, does anyone have a suggestion as to how to make the outlet duct height adjustable, yet air tight, for testing purposes?

And, R2, as so many others have said, thank you for sharing what you have done on your build.  Of course, Phil is the beginning of this whole concept.  Everyone that has tried something new, and shared it, whether it worked or not, has been of great help.  Let me say that again in a different way.  Anyone who tries something new, EVEN IF IT DOESN'T WORK, please share it!  BOTH the successes AND the failures are a big help!

[retired2]

So, my theory is that if you take the cross-sectional area of the duct (radius squared x pi), and divide it by the circumference of the duct (2 x pi x radius), that will give you the minimum dimension that you can put the duct to the baffle without choking the airflow.  That close to the baffle, I am pretty sure that you will need to be using a bellmouth outlet to minimize turbulence.  As far as airflow goes, I would think that you might see an improvement if you increase the height up to 10 percent, which in this case would be 1.65 inches high.  After a 10 percent increase, I don't know if you would see any more improvement in airflow.  All of this speaks to airflow only, not to turbulence, by-pass, or scrubbing.  But this is going to be my starting point.

Since you are planning on an adjustable outlet height for testing purposes, I don't see any reason not to start with the height you calculated.  However, I have a feeling that the minimum distance from the bellmouth to the baffle will be more than you calculated because the efficiency is more dependent on airflow patterns than the volume of space, even with the added 10%.  I took a quick look on the internet, but could not find any information about minimum distances between bellmouth inlets and obstructions.  I'm sure it is there somewhere, but it is not worth the effort because you will be able to test it faster than you can find the scientific answer.

Along these lines, does anyone have a suggestion as to how to make the outlet duct height adjustable, yet air tight, for testing purposes?

Follow my design for attaching the bellmouth outlet to the separator top.  You can move the bellmouth with tailpiece up and down at will.  Testing positions can be quickly sealed airtight with simple rope caulk.  When you want to test another location, the rope caulk peels away easily, and can in fact be reused multiple times.  When you've found the optimum location, replace the rope caulk with a heavy bead of construction adhesive.  Well into the thread on my build there is a photo showing my air straightener.  That photo also shows the rope caulk sealing the bell mouth tailpipe.

[BernardNaish]

If you are using a 6" diameter pipe and the height of your top hat is 12" then I presume you have not planned to have the inlet pipe rectangular? To my mind a rectangular inlet that encourages waste to hug the wall is more critical that the outlet bell mouth to baffle distance, or have I got it wrong?

Why are we thinking a double height top hat? This is double the height of the rectangular inlet "chute". With a 6" pipe the rectangle could be 2 " x 14" or 3" x 9" or 4" x 7". This would give 28" or 18" or 14" high top hats if they are double height. The ideal is to have the fans inlet fitted to a flow straightener directly above the top hat - all in a straight line. The height of the inlet to the top hat is set by the height of the dust (trash) bin and the raising/lowering mechanism. The ideal inlet main trunk from the collection network should be straight for the greatest distance possible.

Taking these all together we see that the bin height 28" (my guess), the raising/lowering mechanism perhaps 6", top hat 28", air straightener 12", fan say 12" all adds up to 86" or 7' 2" which is just below the height of a standard room. We have yet to add anything that may be needed to get the fans outlet into the HEPA filter/collection bag system. I conclude we need to keep the top hat height down as we are starting to run out of available workshop height. This is one reason why cyclones start to get impractical in amateur workshops.

I believe we worked out that the larger the internal volume of the top hat, the greater was the pressure (or was it flow?) loss, and this needed to be kept as small as possible. This also pointed towards a shorter top hat height.

Why has there been so much attention paid to a double height? It seemed to me that we were looking at two related factors that had been observed through plastic walls. The tendency to need more than one revolution of waste circulation for it to be forced down through the slot and the way circulating waste interfered with the flow entering the top hat. Hence giving some space below the inlet rectangle overcame the problem. I also suspect that hang ups on the end of the slot on the downstream side of the inlet may be caused by its proximity to the incoming air/waste stream.  I do not know if anyone has worked out how much is needed but it seems to me that it is probably half to a third the height of the inlet rectangle.

Of course there has also been discussion of a 6" diameter outlet pipe and a 5" diameter inlet pipe that will change all these calculations.

If I have got it wrong I know you chaps will not hesitate to tell me.

[kayak]

Actually, I am planning to use an approximate 3x9 inlet.  However, as pointed out, continually increasing the height of the top hat can become a problem in a home workshop.  My shop is in a basement with a 7 foot 4 inch ceiling.  My situation requires some compromises; rectangle dimension and chamber height is one of them, straight shot distance on the inlet is another.  However, I am trying to maximize the best choices I can in as many places as I can.  Hopefully, this will optimize the system enough to overcome the places where I have to make compromises

In a perfect world, my inlet rectangle would be 25-1/8 high by 1-1/8 wide (the same width as the drop slot).  But in my shop, that just ain't happenin'.  That said, I would really like to hear of someone who does have the luxury of headroom and a system powerful enough to try it! 

The quest for the perfect Thien, so much dust, so little time...

[BernardNaish]

With respect that somewhat misses my point that large top hat volume hits flow performance and probably isn't needed with a one and a half or even one and a quarter top hat height.

[kayak]

Below is a snip of advice I received on top hat design:
<snip> the key is to keep the outlet below the inlet but high enough from the baffle so as not to restrict flow <end snip>
I am not divulging the source of this advice as it was a private communication.

This, I believe, kind of leads us toward having the double height chamber.

The OCD part of me wants to design a variable chamber height/variable outlet height top hat to experiment with different configurations.  But life happens while you're busy making other plans.   :)

[nucww]

I struggled several months over a Thien baffle design. My system (described in "HF DC spiral collector with flow redirection") had to fit under a 74" ceiling shelf.  I only wanted to do it once so I put all the features that I thought would enhance flow with my cheap HF DC.  If you look at the computer generated flow patterns in a Thien baffle type collector linked in one of these posts, the velocity is high on the outside surface, slows down until it gets close to the outlet pipe.  The vortex speed in the center becomes huge which is why I think the flow straighteners work.  I am guessing that the high vortex is not productive in the center because the dust is collected on the outside and the flow inward negates the outward movement of fine dust. In contrast to a cyclone that can use that spin to separate the fines to continue to drop into the container.  To minimize the turbulent losses from the vortex, I added flow redirection between the baffle and the outlet pipe since flow doesn't like to hit a perpendicular wall.  My system works well for now but I would like to test the different outlet designs but its not easy to take apart.  In some respects, I regret not putting in some flexibility to test different configurations but on the other hand I have allergies and needed a  DC system.  So if you are inquisitive and enjoy the ride, go for it.  If you need a DC, keep it simple, wrap it up and start collecting dust; in my opinion the basic Thien design works very well.

[Clark Savage Jr]

Perhaps if you can't change but wish to experiment

[kayak]

Clark, I like that third option.  Reminds me of a setup I've seen to simulate different amounts of drag on the intakes of a DC system. 

I like the looks of it!

Of course, I like the looks of Ferrari's too...

[Clark Savage Jr]

It is kinda like what I alluded to in another post - limiting the exhaust charge to compensate for the dust load and raising the floor of the outlet same as you lower it - just not in the "middle" - seems to work for me - empirical data based only on observation and filter load.