Better off with swivel (adjustable) elbows or with flexible duct on 4 inch turns?

[Roger Rayburn]

I'm getting ready to move to a 4 inch piped (sheet metal snap lock) duct system with drops to replace my single duct moved machine to machine approach. Contemplating it, I have some challenges that require some of the connections to be really flexible (I have to physically rotate my on wheels collector about 30 degrees to get at my clamp rack and I have to move it about a foot to change my air cleaner filter). I can use 4 inch flex duct for those, but it put me to wondering about the comparative disadvantages of flexible duct over adjustable elbows for the rest of the run. Using flex duct I can get a r-e-a-l-l-y long bend which I understand helps with losses. On the other hand an adjustable elbow has smoother sides but I'd have to use two of them to get a long bend. I presume the sheet metal approach delivers better performance, BUT IS IT A HUGE ADVANTAGE OR DOES IT MATTER ENOUGH TO WORRY ABOUT IT on a run that has only three bends in it and one of them is only going to be about 45 degrees? Total run length across the ceiling is about 20 feet and then add another 10 feet for going up and then down the wall for a total run of about 30 feet of duct. Thanks for your thoughts and experiences.

[Paul Wunder]

Someone with practical experience might chime in, but if you go to Bill Pentz's web site he has a worksheet called static calculator (excel spreadsheet). Just pop in your duct type, diameter and length for each method you are considering and out will pop the resistance or loss factor.

[Anthony Whitesell]

First question: What type and size DC?

I have a 1HP single bag unit and piped my basement with the semi-rigid black ABS. That was before I found Sawmill Creek and Bill Pentz website. I would never do it again. If I was to go with PVC, it would be all 6". The machines would be 4, 5 or 6" ports depending on the machine and modifications required. Opting for sheet metal, it would be piped with a 6" trunk line with most drops as 5".

Please carefully consider the information at Bill Pentz website and scan through a few of the dust collection sizing posts before deciding to install the 4" duct work.

[Jim Neeley]

Roger,

Specific to your question, I went to Bill Pentz's web site and played with his staticalc.xls and was surprised at what i found.

For a 4" line, at 600cfm, the pressure drop in a single flex 90 equals about 2.24" of static pressure. This is equal to three long-radius 90*'s (2.5 R/D) or 3.8' of flex hose.

With 2 flexes it all doubles so if you could use less than 6 long-radius 90's or less than 7.6' of flex hose, you'd be ahead.

What this calculation doesn't talk about however is what angle you have the flex flexed to for the pressure given. I'd *guess* it to be at 90*. If so and if you wouldn't "normally" have them flexed, the flexes may look far better.

FWIW..

Jim

[Michael W. Clark]

Need to know the flow to get the loss.
Jim, I know your making an assumption, but 600 ACFM in a 4" is singing. The hood loss will kill it inless the motor and fan are capable.
Not sure about the calc sheet referenced but most duct losses are based on straight duct for flex.

[Anthony Whitesell]

Roger,

Specific to your question, I went to Bill Pentz's web site and played with his staticalc.xls and was surprised at what i found.

For a 4" line, at 600cfm...

How did you arrive at 600 CFM in a 4" duct? Can your DC/fan draw 600CFM in a 4" duct or at the total pressure drop in your spreadsheet?

[Jim Neeley]

My response was to a "which option should I choose" type question. The key part of the message was the relative comparison of the fitting approaches, not the specifics of the SP drop. With the three options I presented (2 flexes, 6 LR ells or 7.6' of hose):

At 600 CFM each option drops 2.24" of SP

At 400 CFM each option drops 2.08" of SP

At 300 CFM each option drops 1.20" of SP

At 200 CFM each option drops 0.56" of SP

My point was that 2 flexes = 6 LR ells = 7.8' of hose according to the calculator, not the specifics of the pressure drop.

My caveat on the bends in the hose (and the likelihood the value for the flexes is for when they are making a 90* bend) still stand.

Jim

[Anthony Whitesell]

My bad. I got lost, and thought I was responding to an update from the OP. Sorry for the weedwacking.

[Michael W. Clark]

The OP is looking for relative difference. The magnitude of this difference is dependant on the duct velocity. Losses increase with the square of the velocity change. Velocity change is proportional to the flow change since the area is constant. IE, the difference in loss will be more obvious at higher flows. It may not be significant at a low flow.

Thats the reason for the question on flow, not lost in the weeds....yet

[Jim Neeley]

Michael,

Unless I'm misinterpreting the original request or what you've said, I believe I disagree with you. I believe it was asking the relative merits of flex ells vs flex hose. My (surprising) findings were that according to Pentz's static pressure calculator, the pressure drop of 2 flex ells equals 7.8' of flex pipe whether the flow is 200cfm, 300cfm, 400cfm or 600cfm. The pressure drop changes with flow but the comparative pressure drop for the different options remains the same.

Maybe I'm missing something here?

Jim

[Michael W. Clark]

I may be missing it Jim, I thought he wanted a magnituded to see if it mattered much or not. One is a clear winner, but the difference is magnified as the flow increases. Losses are based on velocity pressure which is a square function of the duct velocity. V=4005 x SQRT(VP), so VP = (V/4005)^2, assuming air and standard density.

Based on ACGIH:
I assumed a "r-e-a-l-l-y" long sweep is 24" radius and on long sweeps there is no elbow loss, it is treated as straight duct. Two 90s plus a 45 equals 2.5 90s or 7.8 linear feet of flex resulting in 2.49"wg SP loss at 600 CFM. The elbows in galvenized steel, 2.0D radius (0.19VP each) gives 1.49"wg loss at 600 CFM + 7.8 feet of duct since they are figured to the centerline intersections for a total 2.96"wg of loss.

So, theoretically, the flex is better by ~0.5"wg at 600 CFM based on the assumptions. If the fan can only develop 400 CFM when you consider the other losses in the system, the flex is only better by ~0.25"wg. The flex will always be better based on the assumptions I made, but how much it is better is dependant on the volume flow.

Now, this assumes that the radius of the flex elbows are long enough not to require additional (measurable) energy to change the flow direction. If they are shorter and sharper, all bets are off.

There are some downsides to using the flex besides the pressure drop consideration. Elbows are more prone to wear than the straight duct, and the flex is likely less abrasion resistant, but this may not be noticeable in a home/hobby shop environment. Industrial systems use elbows and fittings that are often a couple of gauges heavier than the straight duct.

Another downside that will affect the OP is mounting. The flex will likely not support the ductwork and maintain the correct bend radius/shape. The duct on both sides of the flex will need to be supported. Supporting using rigid elbows is simpler.

Personally, I would use a single adjustable metal 90 where the duct is fixed. Use the flex for connections to the machines and collector to minimize vibration being transmitted to the ductwork. I would also use it if I had machines that moved or did not stay connected all the time. Bottom line, I would use metal/rigid where possible and flex only where I had to.

[Jim Neeley]

I see your logic and it is highly dependent upon our assumptions. <g>

Back to the OP's question, he indicated the DC must be moved to get at clamps or to maintain the DC filter but doesn't say what the angle is when the system is running; it could be a straight-shot when in use for all we know. If that were the case he might be better off using another approach such as a quick-disconnect on a straight run and dodge the flexing entirely.

[Michael W. Clark]

Yes Jim, lots of assumptions! (Typical)

I agree, conventional wisdom is to minimize the use of flex where possible.

[Roger Rayburn]

I want to thank everyone for their analysis and their comments. It is absolutely everything I hoped for to have people who actually understand this stuff discuss it and arrive at data and recommendations that apply in my particular situation. You are examples of why this forum works so well. To answer your questions, my DC is a HF "2HP" which we all sort of know isn't. But, I have replaced the impeller with a Jet 11.75 inch impeller bored out to fit the motor shaft. It is really loud, but I wear hearing protection anyway so that isn't a concern. Significantly, my static pressure improved 50% with this mod ( from 6 to 9 ). I have no idea what the effect was on flow. I live in Colorado at 6900 feet above sea level. Air is significantly less dense here so while all of the rules apply, many are surprised at the differences in my numbers compared to their own. What I think I have learned from all of you is to limit flex to machine connections and use adjustable elbows for duct turns, preferably two of them as 45 degree bends to add up to 90. I think I have laid out my route to get away with one 45 degree bend instead of a 90 against the ceiling and simply arrive at a different location at the wall. Thanks for your help, my friends. You have debated numbers I cannot comprehend and made me comfortable with my plan. May you always have high flows, enviable static pressure and my your fines be trapped in your separators and never clog your filters.

[Michael W. Clark]

Hi Roger,
The statice pressure improvement is indicative of about a 22% improvement in flow. Based on the info you gave in the original post and 9" SP, you are flowing about 550 ACFM at your elevation. If you were a flatlander, you would be running about 440 ACFM.

I like the DC/Roy Underhill sign-off.