Under Floor Dust Collection

[Jon Snider]

My new shop will have a wood floor with probably joists over a small crawl space. Not something I’d ever want to go down into to fix ducting. I’m thinking I’ll lay OSB with screws so they can be removed if I need to get to them from above.

So, having read most of Bill Pentz’ site and gone through the last 250 or so pages here at SMC, I’m at a crossroads. There are literally hundreds of dust collection posts, far outnumbering any other category, but there doesn’t seem to be much consistency. Pentz pretty much says put your ducts overhead. Otherwise, expect clogs or bulky material damaging the impeller. Not to mention possible fire hazard from high amounts of residual material in ducts. That option isn’t my best for several design reasons. Plus, in building a clear span structure w no internal posts, there’s no way I’m going to go anywhere but below to my central saw and jointer cluster.

Whats the real world experience, assuming good duct layout with a powerful motor such as a 5hp ClearVu and 6” ducts as close as possible to each tool?

One of my main concerns is the rise from the subfloor up to my DC intake.

Have others seen these problems? If I have a clog I can probably get to it to fix it but not interested in anything more serious.

Thanks much for any advice.

Jon

[Jim Becker]

Outside of unusual circumstances...excellent air flow and proper duct design cures all. Dust collection is about moving air. The air, in turn, moves the material. If you make that air move as efficiently as possible and with a volume (measured in Cubic Feet Per Minute - SFM) that exceeds the requirements of the tool collection expectation, etc., you should be fine regardless if your duct work is overhead or down below.

[Jaromir Svoboda]

Never had clog in 7 years. I just finished building my new shop and I put duct in floor again.

[Jon Snider]

Never had clog in 7 years. I just finished building my new shop and I put duct in floor again.

Jaromir. 6” ducts?

[Chris Parks]

The short answer is it depends, if the dust extractor cannot transport the debris it will clog so don't buy a cheap DE and design the system so it can be got at if necessary.

[Tom Bender]

A couple of pointers;

Install quick disconnects at places you may need to access.

Sometimes vertical sections are designed smaller then the main run so stuff will be lifted by the velocity. As Jim says, "proper duct design"

[Joe Jensen]

I have overhead ducts in my current shop. I do get small cut offs that get sucked up above the blade guard but the don't get all the way up to the horizontal so when the collector shuts off they fall down. I suppose you might get small pieces of scrap falling down into the ducts and then fail to go vertical back out of the floor, I never considered that. I am sure that disconnecting duct under the floor would be a major pain in the butt.

[Andrew Seemann]

Probably the biggest advantage of having overhead/above ground ducts is that they can be moved.

I've had 3 different shops, and all 3 times I have installed dust collection, I have ended up not liking the initial result, despite weeks/months of planning. Same thing with the machines, I always seem to find a better layout after I get everything in there. I can't imagine being locked into a specific layout by underfloor ducts. I have 9' 8" ceilings with a 6" main and flex duct coming down from the ceiling. No problems with clogs or lack of pull with a 3HP Oneida.

Granted a crawl space is a little more flexible than burying them in concrete. I'd say go for regular plywood rather than OSB. It will probably wear better. Also look for screws that can have the gunk cleaned out of the heads easily.

[Chris Parks]

I think my DE ducting is on the fifth iteration at least.

[Jon Snider]

Probably the biggest advantage of having overhead/above ground ducts is that they can be moved.

I've had 3 different shops, and all 3 times I have installed dust collection, I have ended up not liking the initial result, despite weeks/months of planning. Same thing with the machines, I always seem to find a better layout after I get everything in there. I can't imagine being locked into a specific layout by underfloor ducts. I have 9' 8" ceilings with a 6" main and flex duct coming down from the ceiling. No problems with clogs or lack of pull with a 3HP Oneida.

Granted a crawl space is a little more flexible than burying them in concrete. I'd say go for regular plywood rather than OSB. It will probably wear better. Also look for screws that can have the gunk cleaned out of the heads easily.

Thanks for your reply. Very helpful info!

[sean meltvedt]

6” pvc duct under concrete and steel. ClearVue 1800 sucking on it. No problems so far (3 years). Only issue is the permanence. So the entire run is not underfloor to each machine. Instead, I put them up along the far wall, and localized clusters of machines, and run along the walls for distribution. Not the best layout, but beats having it hang below the finished ceiling level above.
This partial underfloor has accommodated me rearranging the shop a couple of times.
So, yes if I were building again, I would run underfloor to key areas and above floor to the machine.
Good luck, and enjoy
Sean

[Bill Dufour]

Unless it is a deep trench the up turns will not be long smooth arcs so they will reduce flow.
Bill D

[Chris Parks]

There has been flow testing done in Australia that shows long curves flow worse than short curves contrary to what we all believed and to what seems logical.

[Jon Snider]

Mine will be in a crawl space under joists so plenty of room for long radii.

[Peter Christensen]

There has been flow testing done in Australia that shows long curves flow worse than short curves contrary to what we all believed and to what seems logical.

Chris would you pm me the link if you can find it? I know two forty-five degree corners are worse than a large radius ninety but had seen nothing about a long sweep being worse than a short one. Thanks

[Morey St. Denis]

"Flow testing done in Australia that shows long curves flow worse than short curves"?..

Agreed, that statement is bound to be highly controversial. Chris, could you kindly provide some science and further documentation behind that assertion? Not only does it go against what "seems logical"; you're also implying that nearly two centuries of scientific research into fluid mechanics and our resulting textbooks on fluid mechanics and fluid dynamics that we learned from at university were wrong... Surely there has got to be another explanation to whatever might have been empirically observed. The flow coefficient of two 45 degree fittings back-to-back and added together always is better than a single 90 of equivalent radius and the longer the curve radius getting from vector A to right angle vector B, the better the flow coefficient, remains my thinking until proven otherwise. Simple conservation of momentum and energy, even within a fluid media of mixed densities .

[Jon Snider]

I think my underfloor duct question is about to get derailed....

[Peter Christensen]

Morey here is a link with engineering data (http://www.lorencook.com/PDFs/Catalogs/Cookbook_Catalog.pdf) for air handling systems. On page 54 the duct losses for ninety degree and three piece ninety (i.e. two forty-five's) degree elbows are shown. The 3 piece ninety is almost double that of the smooth ninety degree elbows. Does it challenge your thinking? Even if you take in the value of the two forty-fives added together and not counting the straight section between they only best the smooth ninety with a .75R.

[Peter Christensen]

I think my underfloor duct question is about to get derailed....

Maybe a little. Sorry.

If I had a crawlspace and a strong floor I'd put the ducts below. Any future changes would be only a little harder in sitting room rather than from a ladder. You would just have to plug the old floor holes.

[Carlos Alvarez]

"Flow testing done in Australia that shows long curves flow worse than short curves"?..

The curve was upside down.

[Chris Parks]

Let me clarify what I mean before everyone starts jumping to conclusions because maybe I have jumped to a conclusion. The conventional method of using two 45's instead of a single 90 bend is what I am referring to and not a manufactured bend in a pipe. Also the tests were done in PVC, we in Oz do very little metal ducting simply because it basically does not exist due to no HVAC industry as you guys have. Go to the Ubeaut woodworking forum and have a search through the dust extraction sub forum for the tests. There is a lot of work done in that forum looking for ways to improve flow and also looking at dust monitoring, more than anywhere else I know of. There is hours of reading there for anyone who is interested.

[Joe Jensen]

Chris would you pm me the link if you can find it? I know two forty-five degree corners are worse than a large radius ninety but had seen nothing about a long sweep being worse than a short one. Thanks

I too would love to see the link, pretty hard to believe

[Chris Parks]

I can't link here but do the search, it isn't hard to find the forum or the sub forum. Sending a private link is just as restrictive as only one person gets it and everyone really should read some of the stuff on that site. Don't blame me I would love to put the link up.

[Peter Christensen]

No sweat Chris. I’ve been a member there and the DC forum is usually the first I check. QC Inspector ;)

[Chris Parks]

I think it has the most up to date stuff available for anyone looking at DE in a serious way. BobL did the tests sometime in the last twelve months (I think) with full graphical results as he always does.

[Robert Engel]

I think my underfloor duct question is about to get derailed....

Surprised? DC threads are like sharpening and dovetail threads......LOL

I do not have an underfloor system, but I've never once clogged a DC duct (mine are 6" w/ 4" drops).

There won't be an issue going overhead I think you will find the vertical drops not an issue.

To your question, with a unit that big, I can't see a vertical rise will be an issue.

I don't know about Clear Vue but Oneida will help you plan your system if you purchase a unit.

[Chris Parks]

Go underfloor if you can and want too, part of mine is underfloor. The only provisos I would put on doing it is a very good dust extractor such as Oneida or Clearview, nothing less, a good design that can be accessed for any blockages and I have outlined that above where the main ducts can be accessed from the boundaries of the slab or floor. I certainly have never had any clogging in my overhead or underfloor system using a CV1800 and would not expect any to occur in a well designed system. I hate overhead drops unless they are directly over the machine as they just interfere with me getting around the workshop so I went to extreme lengths to design the ducting to drop straight onto the machines. The problem I see with going underfloor is no matter how well I plan it always seems to need design changes and most people find the same thing. I mentioned above my system is on its fifth iteration at least, new machines, new cabinets and benches etc. Even the way we work changes as we gain more experience etc.

[Carlos Alvarez]

WHAT?? Why can't the link be posted?

[Alan Schaffter]

The big advantage of underfloor is that you can make the most direct runs, avoid 90° bends, and not worry about overhead beams, pipes, lights, etc. Remember, however, that unless your DC is under the floor too, the duct will need to go up and you'll need a horizontal section before it enters the cyclone (if using a cyclone.) My shop is on the second floor of my garage so most ducting runs along the garage ceiling. My "push-through" cyclone is at garage level also. The duct penetrates the floor for machines in the center of my shop and for machines along the wall it runs up to and behind the knee walls. It has worked great for over 14 years!

http://www.ncwoodworker.net/pp/data/1329/medium/P10100382.JPG

http://www.ncwoodworker.net/pp/data/1329/medium/P10100902.jpg

http://www.ncwoodworker.net/pp/data/1329/medium/P1010091a1.JPG

http://www.ncwoodworker.net/pp/data/1329/medium/P3030143.JPG

http://www.ncwoodworker.net/pp/data/1329/medium/P1010088a1.JPG

http://www.ncwoodworker.net/pp/data/1329/medium/P3070165.JPG

[Carlos Alvarez]

I am SOOO jealous.

[Peter Christensen]

Jealous no. Envious yes. Did you buy the long cone cyclone or make it?

[Alan Schaffter]

Jealous no. Envious yes. Did you buy the long cone cyclone or make it?

I built it- it is a version of Bill Pentz's design. It is a so called 3D cyclone which supposedly does a better job separating out the fines than the standard 1.64D cyclone. These are old pics showing manual blast gates. All gates are now fully automatic and each one opens whenever its machine is turned on (current sensor switches at the wall outlets, solenoid air valves, low pressure air, and pneumatic actuators). There are various write-ups and pics on this and other WW forums.

[Jaromir Svoboda]

Yes, green pipe from Menards

[Jack Frederick]

Just because you do not "want" to have to go into the crawl space does not mean you are not going to "have" to go into the hole. I would suggest that you build the crawl space to allow you to get under there with a minimum of fuss. Minimum 30" deep, good vapor barrier & a good stone base that doesn't tear you up when crawling on it and good lighting. My shop has concrete floors but when we re-modelled the house I put plenty of lights down in the hole. I don't recall the brand but the electrician used 6" fluorescent hanging lights with a protective plastic lens. If you do this you will be able to access the DC system as well as move any electrical connections you may need as your machinery lay-out changes. I would not clutter that space up with HVAC ducting. Pay attention to how you seal that area up from the outside. You can't keep spiders out but you want to know that nothing else is going down there. I like your idea of a framed floor and wish I had done a wooden floor in my shop rather than the concrete.

[glenn bradley]

I think my underfloor duct question is about to get derailed....

As you've noticed, several people here have run under floor with good success. Dust collection threads, like electrical threads, run the gamut since everyone's needs are different; excessively so in a home shop. No problem with under or over. I have overhead and some waist-high runs between machines. I put in clean-outs at strategic locations but, when I open them everything looks like bone china. Sawdust does a great job at polishing PVC.