Yet another newbie with questions

[jdon]

I'm in the process of planning/designing a top hat separator, and have found all the information on this site very helpful, esp. retired2's detailed assembly documentation.

I have a couple of questions:
1. I assume that a rectangular inlet is preferred, in that it provides for better laminar flow as air circulates in the separator. Also, the cross sectional area of the rectangle should be no less than that of the entering hose, to prevent air flow restriction.

That being the case, is there an optimal height to width ratio of the inlet? It seems to me that the narrower (and higher) rectangle would result in better laminar flow. So, for example, a 2 inch wide, 12 inch tall inlet would provide better laminar flow than a 4 by 6 inch inlet, and a 1 by 24 inch would be even better. Are there diminishing returns as the separator volume increases (to account for the taller rectangle)? This is all theoretical, since building 12 or 24 inch tall separator would get pretty involved. My gut instinct is to go with what appear to be practical dimensions (a la retired2's design).

2. I've been thinking about how to support the baffle, and rather than using threaded rods, I thought I might try using a couple of thin strips (3/8" to 1/2" wide) of sheet metal, hung from the top of the separator, with a right angle bend at the bottom of each strip to support the baffle. The wide dimension of the strip would be parallel to air flow, so I figure that it would produce less turbulence than a rod, but haven't seen anybody trying that. Any reason why it wouldn't work?

This is a great site with terrific information- all feedback will be appreciated!

[dabullseye]

i just used 1/4 hardboard with no supports and it works with no probs

[alan m]

your idea about the sheet metal brackets should work better than the threaded rod.

if you make the baffle plate thicker and chamfer the edges of the slot you wont have any supports in the way.

im not sure about the 1"x24" inlet. if its only fines it would be great but i would worry about it blocking up with shavings.
dont forget to add on about 10-15% extra for any losses going from round to rectangular

[retired2]

A sheet metal support might work great for fines, but shavings have a knack for hanging up on edges.  One user even reported stringy shavings hanging up at the end of his drop slot and creating a birds nest.

So, you might want to consider which type of support is most likely to snag shavings, a round one or thin edge? 

[jdon]

Snagging of shavings on the baffle support never occurred to me. Doh!  :-[  There's clearly a lot of experience abd wisdom on this site, and I've come to the conclusion that I should start with the tried and true designs, and tweak after I have a working system. Thanks to all- it's a great site!

[retired2]

if you make the baffle plate thicker and chamfer the edges of the slot you wont have any supports in the way.

I'm not sure what you mean when you say "edges", the emphasis being on the plural.  Someone might interpret that to mean top and bottom edges.  I would recommend chamfering the bottom edge, but not the top.  I believe chamfering the top edge would facilitate the escape of shavings and dust.

[alan m]

sorry retired . i only meant the bottom edge. adn the shallower the angle the better. leave the top ege alone.
what i meant was the bottom edge of the slot and the ends of the slot

[Bulldog8]

I just finished rebuilding my baffle and incorporated a round to rectangular inlet, tall chamber and a thin baffle. My original top hat, had a 3/4" thick baffle that I chamfered to thin the baffle.

In the new version, I narrowed the drop slot to 7/8". (Planer and jointer have Byrd heads so I get small chips instead of shavings)

The baffle plate is cut from 20 gauge steel and is supported by a smaller baffle under the steel baffle to eliminate the need for a support rod. (sandwich baffle idea like GAlawyer's)

Originally my chamber height was 6 1/2" with a 6" round inlet. After reading about the benefits of increasing chamber height, I now have a 14" chamber and get a really nice cyclonic action. I haven't attempted any empirical testing to measure CFM changes or level of separation. It was good with the old top hat and fine separation seems even better now. (based upon the drum sander)

Will get a couple of pictures posted this weekend.

Steve

[tvman44]

If by chamber height you mean the distance between the top on the barrel and the baffle with the slot I firmly believe efficiency would improve with more distance especially with fines even thou I have not tried to increase it yet.  I would like to try increasing it when ever I get time to play with it.  Curious to see your results.

[Bulldog8]

I was referring to the distance from the top to the baffle. In this build, I held the inlet to the very top of the chamber and then moved the drop slot/baffle a lot lower than I needed to. Previously, I had the baffle right at the bottom of the inlet pipe.

In this configuration, the outlet, or suction pipe is pulling the dust downward with seems to increase the cyclonic action. The dust travels twice around the chamber in a nice swirl before going through the drop slot.




Here is a video of it in action.

http://www.youtube.com/watch?v=RNl6sMVSVss&feature=em-upload_owner#action=share

[jdon]

I'm curious about Bulldog8's design- did you keep the same orientation of the baffle slot and the inlet- that is, did you leave the 120 degree solid section in front of the inlet?

My impression is that the orientation of the slot is important, at least when the inlet is just above the baffle.

In Bulldog8's video, it looks like the air path takes about two full 360's above the baffle. I can't tell whether: a) Bulldog8 carefully designed the geometry (inlet angle, height, etc.) to get good separation; b) he was just real lucky that the slot just happened to line up well with airflow by the time it hit the baffle; or c) maybe the orientation of the slot isn't all that important, at least with a tall separation chamber.

Still trying to sort out the complexities :P 

[alan m]

the 120 degree part of the baffel has to be where the air stream hits the level of the baffel.
in a single height seperater it is just infront of the inlet.
a multiple height seperater is harder to work out. it is usually just luck

[Bulldog8]

I'm curious about Bulldog8's design- did you keep the same orientation of the baffle slot and the inlet- that is, did you leave the 120 degree solid section in front of the inlet?

My impression is that the orientation of the slot is important, at least when the inlet is just above the baffle.

In Bulldog8's video, it looks like the air path takes about two full 360's above the baffle. I can't tell whether: a) Bulldog8 carefully designed the geometry (inlet angle, height, etc.) to get good separation; b) he was just real lucky that the slot just happened to line up well with airflow by the time it hit the baffle; or c) maybe the orientation of the slot isn't all that important, at least with a tall separation chamber.

Still trying to sort out the complexities :P

I left the 120 degree solid section right in front of the inlet as you normally would. Everything else was a SWAG. I made the rectangular section the width of some scrap wood in the shop. (scraps and clamps to do the forming) If I was ever to build it again, I'd increase the height and decrease the width, in an effort to keep the air stream close to the wall.

I didn't really have a plan as to how much I would increase the chamber height other than to double what I had before. So I went from 6" round inlet to a rectangular inlet. Previously it seemed like the dust dropped below the slot just before the end of the 240 degree portion. (in fact I chamfered the end of the slot in an attempt to help my 3/4" baffle drop debris below the plate instead of having to go around again.

I found that I usually get a very nice two times around swirl before the dust goes below the baffle. Different woods and different machines separate a little differently, but for the most part I get a nice thin stream twice around the chamber.