Top Hat construction question

[sneezy]

I am thinking of attaching the sheet metal, shown in the photo, to the top and bottom retaining rings with finishing nails spaced approx. 1" apart. Even though they can be countersunk they will still protrude slightly and the wall won't be completely smooth at the very top and bottom. Do you think that will affect the seperator's performance?
Thanks

[Bulldog8]

I used Lexan not metal so I am not experienced with your situation, but in any case nails every 1" sounds like too many to me. If each end is held captive either with nails or structure and a couple of nails in the top edge to prevent drooping, I don't see how the metal could move.

Steve.

[galerdude]

Nice construction, Sneezy! I too, as Steve (Bulldog) did, used lexan. On the bottom of the lexan I countersunk screws to keep it taught to the framework. Turbulence ?, I'm sure there is. Does my Tophat work? Yes! If the nail attachment becomes a performance concern, possibly you could use a glue or silicone between the metal and framework and then clamp to the framework with a temporary circular jig, thus fewer fastners. I don't know if that would work, just throwing ideas out there.

Best,
Gale

[retired2]

I am thinking of attaching the sheet metal, shown in the photo, to the top and bottom retaining rings with finishing nails spaced approx. 1" apart. Even though they can be countersunk they will still protrude slightly and the wall won't be completely smooth at the very top and bottom. Do you think that will affect the seperator's performance?
Thanks

I used "truss" head screws to attach mine.  They look like a pan head sheet metal screw, but they have an even lower head profile.  Bought mine online from McFeeleys.  They are anodized black.  They look very neat and the head has absolutely not impact on performance of the separator.  Nothing gets hung up on them and if you watch my test videos you will see there is no way they are affecting the circulation of the air or dust.

Retired2

[phil (admin)]

Nails every 1" may present too many chances for debris to snag a nail head.

Can't you use some sort of adhesive instead?  If you clamp both ends of the steel, does it sit nice and flat against the form?  If it does, what would you think about using silicone?

You could still fasten both ends of the steel, but I'd imagine that silicone would hold it amazingly well.

Anyone else have any thoughts?

[retired2]

I missed the proposal for 1" spacing on the nails.  I agree that is much closer than necessary.

Other than material differences, this build presents the same problems as my build.  It has two retaining rings that secure the side wall material.  My sidewall is plexiglass while this one is sheet metal.  I used dadoed stops to retain the ends of my plexiglass.  Then I cut the plexiglass to such a length that it had to be muscled into place.  It literally was jammed against the retaining rings once it was in place.   To be sure it was sealed, I used rope caulk in a rabbit that was cut circumferencially around the retaining rings.  Then I anchored the plexiglass it in place with the truss screws.  When everything was done, I smoke tested the separator and discoverered in spite of my efforts I still had one or two small sidewall leaks.  That forced me to silicone caulk all around the outside of the plexiglass, which was one of the things I was trying to avoid in the first place.  So there is the first problem with sidewalls that are not set into a top and bottom dado.  The separator runs at a sufficient negative pressure that even a good seal on the face of the sidewall retaining rings can be caused to leak without a mechanical fastener of some kind.

The second problem is applying any kind of sealant or adhesive to the faces of the retaining rings before the sidewall is installed.  Whatever material is used it needs to be thick enough that is does not run or sag, because there are two rings that need to be buttered with the adhesive and there is no position that they can be put in that gravity is not going to work against you.  So, keeping the gook in place until the sidewall is installed is another challenge.

And then the next problem is installing the sidewall once the retaining rings are surfaced with adhesive.  It almost takes six hands.  It is one thing to do it when the surfaces are dry, but it quite another task to do it when there is adhesive on the retaining rings and you are trying to maintain it in a uniform thinkness as the sidewall is installed, and of course the sidewall is fighting you the whole time because it would prefer to stay straight rather than be bent into a circular shape.

Those issues are the reason I came up with the idea that I presented in the following thread.  I was too far along in my build to give it a try, but it really would not have added very much work to what I did at all.  However, as a more practical forum member pointed out, this solution might be over-engineering the problem.  But, here's the link anyway.

http://www.jpthien.com/smf/index.php?topic=573.0


So, accepting that the above solution is overkill, here is what I would do in this case.  First I would not attempt to put any glue or sealant on the faces of the retaining rings.  I would force the sheet metal into the circular shape under pressure, just as I did with the plexiglass.  I would retain the ends temporarily with clamps or blocks, whatever seems to work best for keeping the sheet metal pushing outward against the retaining rings.  Then I would start fastening the sheet metal with the truss head screws, maybe every 4" to 6".   I'd pre-drill each hole before installing the screw.  That way the truss head will seat very snuggly with almost nothing for debris to snag on.  In fact, if you overdrive the screw slightly, it will create its own dimple (counter sink) in the sheetmetal.  I have seen no evidence that the truss head screws in my build are snagging any circulating material.  I think my screws are #6. 

Once the sheetmetal is retained fully, I would silicone caulk along all the outside edges where the sheetmetal touches wood.  Caulking the outside edges allows you to apply the caulk in a position that does not cause runs, drips or sags.  Also, by caulking along all these outside contact points it is much easier to insure that the separator will not leak under vacuum.

Retired2

[sneezy]

Thanks for the replies. Do you think contact cement would work? The ends will be retained securely.

[sneezy]

By the way Merry Christmas everyone.

[galerdude]

Thanks for the replies. Do you think contact cement would work? The ends will be retained securely.

I think it would. I'd sure do a test piece first if scrap was available. I would think you'd have to do the ol' dowel  thing like laying countertop laminate or a version there of. Might be tricky getting the sheet metal where you want it. I only say that 'cause I'm not that masterful with contact cement and I know others use it with never a problem  ;).

And a Merry Christmas to you and yours also!

Best,
Gale

[phil (admin)]

The problem w/ contact cement is you get one shot to position it perfectly.

Silicone caulk won't droop, it will bead.  And I wouldn't use enough to provide much squeeze-out anyhow.

[retired2]

The problem w/ contact cement is you get one shot to position it perfectly.

The only time I use contact cement is if I've learned some new profanity that I want to use ten times in a sentence!

[Peter]

I used aluminum flashing and beads of silicone sealant on my first separator. No problems (with that, anyway). If you begin with a coil of metal and install it "against" its already coiled direction, the metal will want to open up against your plywood stop. A couple beads of ordinary clear silicone caulk/adhesive will be plenty good to hold it in place. After all, you're not depending on the metal as a structural element; the plywood enclosure shown in your photos does that.

+1 on the construction, too, btw. Very nice.

[Pop Pop]

I built a top hat on a 5 gal can and used PL construction adhesive to secure flashing to 1/2" OSB rings. The PL holds it well. The only issue was holding it in place while it cured. A clamp on each end of the flashing did the trick.