Duct Size Question-Yes Another but short and quick

[Anthony Whitesell]

I finally noticed in the email from Oneida, the blower for my acquired cyclone is spec'ed at 900 CFM and 8"SP. Will this blower handle a 7" main trunk? Or do I need to drop to 6"?

[Dave MacArthur]

800 CFM flows through a 7" duct at 2994 FPM

3.5" radius
area =pi * r*r, so 3.14*3.5*3.5 = 38.465 sq in. Divide by 144 sq in/ sq ft to get this in sq feet, so we can use it with cubic feet per minute and get feet per minute, = .267 sq ft.
800 cubic ft / min divided by sq ft (cross sectional area of pipe) will give the flow velocity in feet per minute, so 800 / .267 = 2994 fpm

In a horizontal run, 3000 fpm is usually listed as totally acceptable to keep chips entrained in the airflow. In VERTICAL DROPS, I have often seen "3000-4000 fpm" listed. If you wanted to get 4000 fpm, you'd have to reduce to 6" drops, which would be fine. While it's often said to keep the mainline the same as your drops, in fact from this calc you can see that 7" mains will flow acceptably and will give you less static pressure loss from wall friction and turbulence. I'd run 7" and drop to 6" on the verticals down to the machines. I would split into two 4" if the machine had dual dust ports I could feed.

[Ole Anderson]

I agree with Dave, and you will get less friction loss in the 7" at any flow, and it will allow you to run more than 1000 cfm if needed, which your DC is cabable of if it can flow 900 cfm at 8".

[Dave MacArthur]

Yes, I forgot to say, you may not have 8" SP so it may flow more, depending on what port you open and it's distance/angles.

[Anthony Whitesell]

I was hoping for a yes or sure. That's more than I expected. Thank you. And a second thanks for the double check on my logic. It seemed too easy so I thought I missed something.

Hopefully you will see the fruits of all the information provided. I am sketchup-ing (like cading only cheaper) my shop to post proposed duct layouts for critique, commons, and suggestions. I have peacemealed my questions out based on specific issues as it's been to difficult to explain without pictures and thanks to Gabriel I've been able to get the models required to build a presentable representation of my shop. Thanks again.

[Alan Schaffter]

Dave's calculations are correct, but they do not tell the whole story.

i don't recall the details of your DC or shop, but that question can't be answered without knowing a lot more info- (1) what does the entire fan curve look like? (2) what is the size of the intake now? (3) how was the 900/8 data point determined- at the intake? at the end of a 10' long 7" pipe, some other diameter pipe? (4) how long is your main, total runs, and what kind of SP resistance will they have with 6" duct? with 7" duct? (5) what are the CFM reqmts for your machines? That is where you should start your calculations to see if you have the CFM. I think too many folks get hung up on velocity, and I know many do not have 3000-4000 fpm flow, yet I don't think I have ever seen a post by anyone who has experienced a blockage due to dust falling out of the stream.

Remember also, on a fan curve higher SP means less CFM. You may not be able to get 900 CFM. I suspect if you do the calculations, you will indeed have 8" or more of SP, but again I don't know your shop or duct plan.

Now for a real WAG- if the blower motor is less than 3 hp, the impeller less than 14" in diam, and your runs greater are than 20' and not perfectly straight, I would say probably not, go with 6"

Remember, as you design your runs- forget about nice and neat runs tucked tightly along the walls and making 90 deg. bends. Locate your DC, orient the intake, and design your runs so the air takes the most direct path (shortest and straightest) horizontally AND vertically from each machine to the DC. That often means a diagonal with multiple branches coming off at a 45 rather than longer, serpentine branches that must negotiate 90 bends- a diagonal (hypotenuse) is shorter than the sum of the other two sides of a (right) triangle. The best layout is usually XMAS tree shaped. Unused drops/branches (those with no flow/closed by a blast gate at the end) have no affect on flow in an operating branch.

[Anthony Whitesell]

1) Don't have a fan curve. Not available from Oneida for that blower. It pre-dates having that information available.
2) Intake to the cyclone is 7" round, intake to the blower is 8" round.
3) I have no idea. The spec sheet just said 900CFM at 8" SP. As I understand it based on reviewing the specs versus the fan curves at Grizzly and Penn State, most often the CFM and SP are given at opposite ends of the curve. ie., Penn States DC200B specs 1500CFM and 9.5", but that is 1500CFM at 0SP and 9.5"SP at 0CFM. So I should be better than that.
4) I took a rough pass with the staticcalc.xls, and the difference is about 1"SP between 6" and 7" ductwork. But the layout isn't finalized
5) I don't know the exact requirements for my machines. With my current system using a 1 1/2HP Rigid DC and 4" black ABS it is OK (no option of 6" inlet, or I would have gone that route). Most everything needs a bit more, so my assumption is that the cyclone system will be able to provice that bit more.

As for the layout, I will hold comments until I have the layout completed and a new thread started. In short, it is a basement shop spread across two rooms, tight on floor space and headroom. So tightly packed runs is the first design requirement. There isn't space for them not to be. The Xmas design will not be the most effective layout. The video of your layout has given me some ideas to "buy back" some space by running some of the ductwork on the other side of the wall.

But you've also given me an idea to reduce some 90s to 45s.

[Alan Schaffter]

Basements, especially those with low headroom can be a real ducting challenge for sure. I guess the biggest thing, work with what you have, but never compromise CFM by choosing what looks good over what works best. Always think in 3D. It sounds like you know what needs to be done. It just kills me to see a layout that snakes along three walls and has a lot of 90's!! Unfortunately it is too common.

If cost is a concern and if your decision on metal vs PVC is based on a 7" or 6" main, go with 6" and PVC S&D. I have worked with both and just find PVC much easier and certainly less expensive, especially when dealing with fittings. Some folks luck out and find a local fabricator who will sell metal duct and fittings at a price that is competitive with PVC. I find PVC ducting easier to reconfigure. Metal sure makes a shop look more professional, but . . .

I am lucky to be able to run my ducting along the garage ceiling below my second story shop. My 6" main splits into two to cover all my tools (12 ports in all.) The run from my farthest tool to the blower is about 38', but by careful layout, using diagonal runs, wyes oriented correctly, etc. it is almost a straight run as are most of my runs, and I don't have any 90's in my entire system. By carefully locating, orienting, and elevating my blower, the duct runs straight into it!! I oriented the wyes in the main so they both branch off at a 45, and most angle up through the ceiling/floor to my shop at 45 to the machines and short secondary branches.

Some old shots showing my mains, use of wyes, branches that angle up from below, etc.:

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/P1010088a1.JPG

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

[Aaron Rowland]

After you read on the opinions on this you are going to be really confused, so good luck. I've been there and done that and finally figured out that the best advice comes from people who have paid their dues like Allan.

You did not list the biggie and most important item on a Cyclone. The impeller diameter is every thing for CFM. Forget HP because marketing can say anything they want and will. If a supplier will not give you the impeller diameter look elsewhere.

To get 900 CFM @8" SP in a typical system you need a 15" impeller. A 14" impeller at 8 " will flow about 500 CFM. A 12" is way to small for a cyclone. It varies a bit and depends on the cyclones resistance.

The only cyclone I know of that could handle 7" duct would be the ClearVue . They have a 16" impeller. In the real world with leaks, rough duct work etc there are a lot of phantom losses. The finall killer is the machine port. A lot have 4" ports and that really cuts off air flow. Some machines even have 3" ports. With a 14" impeller you would be lucky to see 200 CFM on this set up.

Allan uses 6" PVC duct work from the cyclone to the machine. Don't consider anything else.

How to I know all this? I bought a set of Dyer gauges and ran tests. It was an eye opener.

[Ole Anderson]

Aaron, Wow. Another CV fan obviously (no pun intended). Let me just say that I just installed a 2 hp Oneida and it works just fine, even with a 7" horizontal main run, using 6" and 5" vertical. Metal duct. Works fine too. Just collected 2 barrels of planer chips, almost nothing in the filter catch pan. American Woodworker did some independent tests and flowed the 2 hp Oneida at 800 cfm at 8" SP.

http://www.oneida-air.com/documents/AWW%20article%20jan%202006.pdf

Anthony, yes you will find a lot of opinions here and will have to sort thru the chaff, mine included. Your system will likely run fine with either a 6" or 7" horizontal main (OMHO).

[Alan Schaffter]

Let me just say that I just installed a 2 hp Oneida and it works just fine, even with a 7" horizontal main run, using 6" and 5" vertical. Metal duct. Works fine too.

almost nothing in the filter catch pan

Anthony, yes you will find a lot of opinions here and will have to sort thru the chaff, mine included. Your system will likely run fine with either a 6" or 7" horizontal main (OMHO).

Ole, not to be contrary, but for your info to be useful, you need to define "works fine."

You seem to be talking about what is going on in the duct and the filter/filter pan. That is of much less importance, the real test is what is the CFM at each machine. That is where the rubber meets the road, and where you are most likely to inhale fine dust. Every thing else is nice, but is not the true measure of dust collection.

If your cyclone is working properly (and you are operating it properly*), no chips or large dust should get through to the filter or filter catch pan. Fine dust which is the lightest and hardest to separate out in the cyclone, is the only thing that should be able to make it through to the filter, but you should not see it in the pan unless you have a filter shaker or scraper or tap on the filter because this stuff is so fine it usually lodges in the pleats of the filter.

* With a cyclone, you need to ensure the blower is up to speed and air flowing at max before creating dust. Likewise, you should keep the blast gate open and DC running for a few seconds after you stop creating dust. If the velocity drops in the cyclone, centrifugal separation will suffer. Chips and dust may not separate out and may pass through to the filter.

[Anthony Whitesell]

According to the Onieda Specs, it is a 2HP Marathon Motor attached to an 8" round inlet and 7" square exhaust producing 900CFM at 8"SP. I don't have a any way to contradict their numbers. I don't have the gauges or the electric supply to set it up.

[Anthony Whitesell]

I'm not worried a thing about looking good. My top issues are headroom and space. Cost is proportionate to performance. I would like to run the 6" PVC as it would be easiest to ensure that my homebrew ('creeker team) built PVC gates will fit on the drops. The mesaure will be if the longest run in 6" puts me over the line and forces me up to the 7" metal.

[Ole Anderson]

Alan, When I say works fine, my definition is that the duct is large enough so as to not choke the system and not so large that there is insufficient velocity with the largest upstream port open to move the larger material down the ducts and into the cyclone. In sizing the ducts accordingly, you will hit the sweet spot which will provide the appropriate flow at each machine to collect and move the chips (sawdust management) and to pull the fine dust (health management). This is provided that you have an impeller sized and with enough horsepower such that you are able to pull the needed air (cfm) while overcoming the friction losses (inches of suction) in the entire duct run. 900 cfm is enough air for most machines, including a table saw with top and bottom collection, so if his cyclone/duct/bend/flex/inlet and exit losses are less than 8", Anthony should be good to go. Agreed, that is a big IF with a lot of variables. I read that his question was limited to the appropriate MAIN run sizing for 900 cfm. Whether or not you get little or some dust to the filter is immaterial, and just goes to the level of maintenance needed on a particular system, with little maintenance being the ideal goal. Remember pre-cyclone when we just had bags? That was maintenance. Now you just clean the filter every couple of barrels of chips and dust.

[Aaron Rowland]

Anthony, I think the thread got a bit messy, sorry. I'll start over. I'm confused by your original question. You now have an old Oneida cyclone with out a blower? Your looking for a blower for it from Oneida? Is this correct or a dumb question on my part? I wonder if you have an Oneida 650 or one of their cyclones from the distance past like the one with the internal filter.

If you don't have a blower, yet, there are a lot of blower choices other then Oneida that can save you a lot of money.

[Anthony Whitesell]

It did but that's OK. There useful information in there, especially all the different things that people considered while putting their system together.

I acquired a Oneida C950 Cyclone body and an Oneida Marathon 2HP 12" impeller 8" round inlet 7" square outlet blower. The spec on the blower states 900CFM at 8"SP. I was double checking with the more knowledgeable people that this is achievable with 7" duct work (metal). If not, is it with 6" duct work (PVC). It would appear that as a starting point it can and I must now determine the duct work for further evaluation.

[Aaron Rowland]

I'm familiar with the 950. Very nice and will work well. You need to up date and add a Neutral vane and this is important. The idea was first published by Wood Magazine ,Ap.02 issue.
John Dillbeck had built the Wood Magazine Cyclone and was searching the net for ideas. He found an obscure article that described the Neutral Vane and copied it. It boosted the CFM from 350CFM to 525CFM. Their ahd also been posts on the net about this discovery several months before Wood Published.

To modify your 950 just slide a close fitting tube inside the inlet until it hits something and stops. Back off a couple of inches and that it. You can cut the end of the slider at a 45 degree angle to work better. Seal the gap between the new and old inlet tube. To fine tune for max CFM some people claim they have used a ammeter and slide the tube in and out until you find a point of maximum amps. I have tried this with out much luck. The ammeter is not sensitive enough to nail it. A slack tube manometer you can make your self is much more sensitive. That is a simple clear 1/4" plastic tube bent in the shape of a U. Fill it with 4-6" of water and observe the center point where both levels of water are equal. Next insert one end of the tube into the inlet line going into the cyclone about 12-18" distance away. Slide the Neutral vane tube in and out until then manometer water level is maxed out. One level will be high the other level will be low. You can measure the difference between the two levels and that is your SP . A typical system installed usually runs about 6-7".

The 950 cyclone used a 14" impeller. You will not be happy with a 12" one. The impeller in a cyclone runs in a partial vacuum and does not get a good "bite" of air. Take a leaf from what Allan did and place the impeller ahead of the cyclone. This will bump the CFM up about 15-20%. Make sure you have an ammeter. Harbor Freight sells one for about $20.00. Very important to see if the motor is loaded properly. The Marathon 2HP you have is typically specked at 8.5 Amps FLA,on the label. No so good. A 2Hp Baldor has a FLA of about 11.5A. Watch craigs list for a good deal on a plain 14", 3HP DC blower. They show up occasionally. The Blower does not have to be placed on top of the cyclone. To reduce headroom place it on the floor and connect with PVC.

[Chris Padilla]

Are you planning metal or PVC for your ducting? I ask because PVC will lock you into the even numbers for pipe diameter: 4", 6", 8"....

Generally speaking, 6" diameter ducting will serve most folks very, very well.

[Anthony Whitesell]

I'm not familiar with the neutral vane. Do you have pictures?

The C950 is just a cyclone body. Oneida spec'ed and provided the 2HP motor/12" blower for the system. I plan on integrated an ammeter in the mag starter I need to build for the system.

The two things I'm tightest on is floor space and headroom. I don't have the floor space to have the motor anywhere but on top. The C950 with the motor on top is one tall sucker (forgive the pun). I'm going to have to build a support frame and custom dust bin to deal with the reduced space for the dust bin. I'm working as hard as I can on getting the shop layout completed so you guys can see the reasoning behind all my trade-offs.

[Chris Padilla]

My DC build (http://www.sawmillcreek.org/showthread.php?9933-cyclone-blower-fan-question-excuse-ignorance-please) shows the neutral vane around post 32, Anthony. It is a spiral of metal within the cyclone body. It helps with air and material flow.

[Aaron Rowland]

I think this is whats called paying your dues. The expert gets the money and you get the experience. Tha'ts a hard road and I've been down it. With the net and various sites some times you can beat Murphy. Problem is Murphy post all sorts of bad info and is a very good salesman. I can't believe this, a huge 950 with a 12 " impeller. And some are talking here about 7" duct and almost 4000 fpm velocity.

I knew the 950 was very tall and there is a good reason for that. Tall cyclones have better separation of fines. I also suggested that you place the blower on the floor. There is nothing wrong with that layout. Suppliers should do that instead of reducing cone or barrel height for shops with low ceilings. Not to sure the Oneida blower will lend itself to that however.

On the Neutral vane google Neutral Vane pictures. Lots of stuff. Todd Crow has a site Crows Nest which shows what he did. I think he posts here once in a while.

I would be on the phone with Oneida and see if you can exchange the blower for a 14" one. I hate to guess what you paid for the 12" one. If they recommend 7" duct with a 12" impeller someone doesn't have a clue.

[Anthony Whitesell]

Here's a kicker about the cyclone. I picked it up from an auction including all of Oneida's design paperwork and bill of lading. I know what the system cost and I know what I paid. Let's just say that I don't think I could find a motor or impeller for less than I paid for the whole thing. All the pieces correspond. The bill matches the design which matches the parts in the boxes. So I know the system was designed to use a 5x shaker bag filter in a shop that was 16x24. The blower was designed to sit on the floor between the cyclone and filter. The main trunk was run offset to one side of the shop. The 7" inlet to the cyclone was to be reduced to 6" immediately and ran half way down the length of the shop. Three spurs ran from the main to the other side of the shop, each 5" and the main was reduced to 5" to complete the run down the shop to a single drop. The near side of the shop had three 5" branches from the main trunk.

It's a faily nice tree design. From what I've read and back calculated, there's too much pressure loss with sizes they specify.

[Alan Schaffter]

Still not crazy after all these years! Everything you didn't want to know about cyclone "neutral vanes" and other stuff! :)

Back in the early days of the Wood magazine cyclone and Bill Pentz's research and website, most all cyclones were based on various industry designs for separating out chaff, cotton seeds, etc. They were very turbulent, reduced CFM and didn't separate out fine dust very well.

They generally did not have a neutral vane. Neither did the Wood mag design. Without a neutral vane, there was considerable turbulence as some of the incoming air struck the central outlet tube while the rest of it traveled around the central tube, where it joined the incoming air at a right angle. Since the spiraling air had already started to decelerate there was a mis-match in velocities of the two air masses which also contributed turbulence. (see drawing)

Extending the (circular) inlet into the cyclone body created a "neutral vane" which reduced turbulence and improved performance some. (see drawing)

At about the same time Bill Pentz was researching and experimenting with other improvements, I built a Wood mag cyclone. Bill and I reasoned that excessive turbulence still existed because, even with a neutral vane, some spiraling air was able to strike the side of the neutral vane. A math professor at Cornell sent Bill an equation for a helix- to make a spiral ramp that would cause the air to pass under the inlet/neutral vane (as it turned out a helix can't be made from a flat section of sheet metal- it has to be stretched and warped. The equation on Bill's site is only a mathematical approximation of the helix). (see drawing)

I made a cardboard template after spending a lot of time making it fit, I used it to make a metal spiral inlet ramp for my Wood mag cyclone- I believe I was the first to do that as well as configure my cyclone as push-through (air is forced into the cyclone as opposed to being pulled through.) It worked very well but still had a few problems which Bill addressed in his advanced design (see his website). Bill's design included a rectangular inlet to match the geometry of the cyclone walls much better than a circular inlet. He slanted the inlet at the same angle as the spiral ramp to further reducing turbulence. Instead of a basic neutral vane he incorporated the spiral inlet ramp so the spiraling air would not hit the side of the neutral vane and would join the incoming air tangentially.