Measured loss in air flow?

[Rick T]

Has anyone measured the actual loss/difference in air flow resulting by inserting the separator into the loop for the various configurations?

[phil (admin)]

Has anyone measured the actual loss/difference in air flow resulting by inserting the separator into the loop for the various configurations?

This question comes up from time to time.  But with no two units being alike (side inlet/top inlet, tight/sorta leaky, various diameters, etc.) it would be a pretty large undertaking to offer any #'s that would be widely applicable.

I've taken some measurements of not only "my" units, but also conventional cyclones.  At one point I thought it would be interesting to make comparisons.  But without holding everything else constant (same blower, same ducting, etc.), it just wasn't possible.

[Rick T]

Has anyone measured the actual loss/difference in air flow resulting by inserting the separator into the loop for the various configurations?

This question comes up from time to time.  But with no two units being alike (side inlet/top inlet, tight/sorta leaky, various diameters, etc.) it would be a pretty large undertaking to offer any #'s that would be widely applicable.

I've taken some measurements of not only "my" units, but also conventional cyclones.  At one point I thought it would be interesting to make comparisons.  But without holding everything else constant (same blower, same ducting, etc.), it just wasn't possible.

I hear you.
My experience anecdotally is that it's negligible for mine.. (Delta 50-760 with top hat, 5" in and out, side entry, Wynn poly pleated filter).
I seem to recall Retired2 and some of those guys doing some semi-scientific measurements when they tested the Bell-end and 30% lose sticks in my head.. which seems bogus and I must surely be off-cue on that recollection.
This came up in a round-about way on a discussion about the Oneida Super Cyclone. It escapes me why the air flow would be more efficient than a top hat like mine. It might be more efficient at separating dust out.. but mines batting in the high 90's% range consistently so why would I care? Maybe i have to clean the pleated filter 1 more time/year. No big deal.

[phil (admin)]

I hear you.
My experience anecdotally is that it's negligible for mine.. (Delta 50-760 with top hat, 5" in and out, side entry, Wynn poly pleated filter).
I seem to recall Retired2 and some of those guys doing some semi-scientific measurements when they tested the Bell-end and 30% lose sticks in my head.. which seems bogus and I must surely be off-cue on that recollection.
This came up in a round-about way on a discussion about the Oneida Super Cyclone. It escapes me why the air flow would be more efficient than a top hat like mine. It might be more efficient at separating dust out.. but mines batting in the high 90's% range consistently so why would I care? Maybe i have to clean the pleated filter 1 more time/year. No big deal.

There is undeniably a hit, and when you combine the loss with what a typical shop's ductwork ads, the overall reduction in CFM can be disappointing for some.

As we all know, the losses come from the friction that occurs as the air revolves inside the unit.  And cyclones aren't exempt.  While a conventional cyclone doesn't have a baffle (or the frictional losses associated with the baffle), it can unwrap to a substantially longer length.  That is, the air inside a cyclone travels a greater distance before it exits the outlet tube.  If you were to make a conventional cyclone taller and taller, and extend the outlet tube further and further, you would by so doing also increase the losses introduced by the cyclone.  At some point, the air would just stop spinning and travel in a more direct route to the outlet tube (and separation would cease).

One of these days I should get a DD and use my anemometer to make some comparisons between it and my daily driver (a 30-gallon baffled unit with elbow inlet in the top).  The DD is a pretty tight radius, and therefor unwraps to not that great a length (but the tight radius is a hit).

[retired2]

I hear you.
My experience anecdotally is that it's negligible for mine.. (Delta 50-760 with top hat, 5" in and out, side entry, Wynn poly pleated filter).
I seem to recall Retired2 and some of those guys doing some semi-scientific measurements when they tested the Bell-end and 30% lose sticks in my head.. which seems bogus...

You can't see CFM's and FPM's.  So you will be deceived if you are making judgments about separator losses based on how much of the waste ends up in your drum.

My DC system was in place for a while before I built my separator.  I went out of my way to build it to be as efficient as possible because I knew my longest run would be taxing the performance of my Delta 50-760.  However, it all worked fine.  Then I built my separator.  Again, I made every effort to build it in a way that would minimize losses.  When I installed it, I could see very little difference in waste pick-up.  So, "anecdotally" I could say the difference was negligble.  However, my longest run is to my bandsaw, which also produces the finest waste, and I noticed there was a little more dust escaping the pull of the DC system then before.

So, I decided to purchase an anemometer of the same brand Phil uses in order to find out what what my system was really doing.  I did extensive testing at a midpoint in my system where I measured CFM's and FPM's with and without the separator installed.  Without the separator, the CFM's were about 700, considerably less than the stated rating of the Delta 50-760.  Of course the stated rating does not include any piping and probabaly not even a filter.  When I installed the separator, the CFM's dropped to about 430.  That is about a 38% reduction in CFM's.

I repeated these tests several times because the results vary slightly from one test to the next.  The numbers I published are averages of those test readings.

So, should everyone expect to see a 38% loss of CFM due to the addition of a top hat separator?  Well, if their DC is exactly the same model, and their top hat is built to the same design, and the piping size, material, and confiuration is exactly the same as mine, then the answer is they should expect something very close to 38%.  If there are any differences, then the answer is the results will surely be different.  How much different?  The only way to know that for sure is to do what I did, build it then measure it - with instruments!

[Rick T]

Were/are you using 5" flex hose and if so how long to the BS?
I've been using the 5" on my 50-760 (@ 10' length) and moving it from machine to machine with the exception of my jointer where I add a piece of 4" to extend it another 8 or so feet. The tapper increases the velocity of course. As you say it's hard to judge the flow rate without equipment.
My setup might benefit from a run of solid 5" and reduce the length of flex hose.

[retired2]

Were/are you using 5" flex hose and if so how long to the BS?
I've been using the 5" on my 50-760 (@ 10' length) and moving it from machine to machine with the exception of my jointer where I add a piece of 4" to extend it another 8 or so feet. The tapper increases the velocity of course. As you say it's hard to judge the flow rate without equipment.
My setup might benefit from a run of solid 5" and reduce the length of flex hose.

If all you are running is the 10' of flex hose that came with the DC, you have enough surplus SP that you probably don't see much difference, and I suspect the separator doesn't impose as much of a hit under those circumstances.  However, without test instruments you can't be sure, and that is why I qualified my comments to say that my measurements reflect a somewhat typical fixed DC installation.

The details of my build and photos of my test setup are shown in the following thread:

http://www.jpthien.com/smf/index.php?topic=563.0