5", Rectangular Inlet, Bellmouth Outlet with Air Straightener, Top Hat Separator

[retired2]

TEST DATA


Well, the construction of my dust collector stand is far enough along that today I was able to do some ammeter testing.  I took meter readings for both a bellmouth outlet pipe and a straight outlet pipe.  I also took a set of readings with the separator removed.

The data table shows ammeter and sound level readings for three positions of each pipe.  The positions represent the distance down from top inside of the separator to the bottom of the outlet pipe.   I took readings at three locations just to see if the ammeter could pick up a difference, not to prove or disprove where the sweetspot of separation occurs.  I?m assuming that to be close to the 3? position.

At each position, I then took four readings for different conditions.  ?Deadhead? means the DC was running with no blast gates open.  The ?Floorsweep? is a moderate distance from my DC and has a 5? open port.  The ?Bandsaw? is my longest connection with a 4? and 2? port.  And ?Open? readings were taken with the 5? end cap shown in the photo removed.  That is a distance of about 3 feet of pipe and one 90 degree elbow from the separator.

Here?s the photo showing my test setup.  It's 5" taller than it was supposed to be - early-stage dementia!!! :):





And here's the data table with the ampere and sound level readings:





Here are the problems I found with ammeter readings:
- Meter readings fall in a tight range making data analysis difficult.
- The min reading (deadheaded) and max reading (open at DC) range is only 2.36 amps.
- The operational range (Floorsweep and Bandsaw) of this test is only 0.78 amps.
- Readings bounce by 0.10 during the test.
- Readings drift down slightly during a sustained test - not sure why.  Motor temperature?


So, what can we conclude from this data?  Well, it seems to provide more evidence that a bellmouth outlet pipe enhances the performance of a Thien separator.  It may look like a very small increase, but you need to keep in mind the small range of test readings.  But, without additional test equipment it is difficult to quantify the improvement in more meaningful numbers.

I did make some other observations from the data:

1)  I would have expected the "Deadheaded" readings to all be the same, but for some reason the bellmouth readings are higher which would suggest it is moving more air.  I doubt that is the case, since the only air could come from minute leaks.  I am thinking it might be the result of the separator being close coupled to the DC and the fan may be doing more work as a result of turbulence.

2)  Correction: The bellmouth pipe seemed to move less air when it was in the up (1-1/2") position, but it performed better in both of the other two positions.  Conversely, the straight pipe performed its best in the up position, and slightly weaker in the other two positions.  In no case did the straight pipe produce higher readings than the bellmouth.

3)  The normal operating load readings (floorsweep and bandsaw) are higher for the bellmouth pipe (middle and low positions only) than for no separator at all.  That makes no sense whatsover, unless there is a base-load resulting from separator turbulance as I suggested in item 1.

4)  The sound level readings mean nothing other than I was looking to see if some conditions might be noisier than others.  As it turned out they are all pretty close other than Deadheaded which is much quieter due to the lack of high velocity air.


And just to add to the confusion, here's a little gem from a Cincinnatti Fan engineering document: 

Duct Inlet Spin
"A major cause of reduced fan performance is an inlet duct connection that produces a spin or pre-rotation of the air entering the fan inlet.  Inlet spin in the same direction of the fan wheel will reduce air volume and pressure ratings.  Inlet spin in the opposite direction of the fan rotation will substantially increase the motor horsepower requirements.  An ideal inlet condition is one which allows the air to enter the fan axially and evenly without spin in either direction."


I wonder whether rotation might be a factor in a close-coupled top hat design with a bellmouthed pick-up?


Regards, and enjoy the data,

Retired2




 

[Don_Z]

Okay, just so that I am reading this correctly, you are saying that the inlet (bellmouth) pipe performed better being placed more than halfway inside the separator and that the bellmouth does perform better than a straight pipe?? By the way, a very solid build. I just finished mine up yesterday alas no time for painting it just yet. She isn't the prettiest girl at the dance but she has some nice moves. Pictures to come later today.

[retired2]

Okay, just so that I am reading this correctly, you are saying that the inlet (bellmouth) pipe performed better being placed more than halfway inside the separator and that the bellmouth does perform better than a straight pipe?? By the way, a very solid build. I just finished mine up yesterday alas no time for painting it just yet. She isn't the prettiest girl at the dance but she has some nice moves. Pictures to come later today.

Don, I don't want to make any bold claims because it has already been suggested by one member of this forum that my only qualificatons are having slept at a Holiday Inn Express and that I should be banned from posting. 

So, with that in mind, and considering the problems I've outlined above for ammeter readings, I can say the readings I took show the blower was doing more work when the bellmouth was at the mid and lower position (deeper in the chamber). and that in all cases the ammeter readings were slightly higher for the bellmouth than the straight pipe.  Normally, higher readings indicate the blower is moving more air.


p.s.  Don, don't paint anything with waterbased paint that you hope to get apart in the future.  The plug I made to get the bellmouth out of my chamber was only screwed in place one day and I had to pry it loose with a 1" butt chisel.

[Don_Z]

Well I have always been a person that thinks the proof is in the pudding. And whoever it was that said that regarding your work and approach I would invite to see their work and see what they can produce rather than hot air and derogatory remarks against someone when this is supposed to be an open forum to help each other. I bet that was what Phil had in mind when he made this forum but I am not going to talk for him.
Anyway, the reason I am asking is because I have ordered my bellmouth pipe and have built my system around a 5" pipe/ hose assembly. The interior of my chamber is 5.75", so I am asking your opinion on how far should I place the bottom of the bellmouth opening from the top of the chamber  (not trying to get your to refute or challenge anyone else's claims) since you are the only person I have seen use this kind of pipe?

And thanks for the warning on the paint. I never use waterbased products anyway. They just don't produce those fumes that make me feel all warm and fuzzy.  :o

[retired2]

Anyway, the reason I am asking is because I have ordered my bellmouth pipe and have built my system around a 5" pipe/ hose assembly. The interior of my chamber is 5.75", so I am asking your opinion on how far should I place the bottom of the bellmouth opening from the top of the chamber  (not trying to get your to refute or challenge anyone else's claims) since you are the only person I have seen use this kind of pipe?

And thanks for the warning on the paint. I never use waterbased products anyway. They just don't produce those fumes that make me feel all warm and fuzzy.  :o

Well Don, let me put it this way.  The bellmouth is staying in my separator until someone now provides evidence to the contrary.  I'm convinced there is some benefit - I just don't know how much. 

And by the way, I am equally convinced that the rectangular inlet is also beneficial, but again I can't say by how much.  What I'm not sure of is whether a custom fabricated transition from 5" round to rectangular like mine is necessary.  I think it is possible to fashion one from wood that would provide most of the benefits at a much lower cost.

[retired2]

Anyway, the reason I am asking is because I have ordered my bellmouth pipe and have built my system around a 5" pipe/ hose assembly. The interior of my chamber is 5.75", so I am asking your opinion on how far should I place the bottom of the bellmouth opening from the top of the chamber  (not trying to get your to refute or challenge anyone else's claims) since you are the only person I have seen use this kind of pipe?

And thanks for the warning on the paint. I never use waterbased products anyway. They just don't produce those fumes that make me feel all warm and fuzzy.  :o

Well, Don let me put it this way.  The bellmouth is staying in my separator until someone now provides evidence to the contrary.  I'm convinced there is some benefit - I just don't know how much. 

And by the way, I am equally convinced that the rectangular inlet is also beneficial, but again I can't say by how much.  What I'm not sure of is whether a custom fabricated transition from 5" round to rectangular like mine is necessary.  I think it is possible to fashion one from wood that would provide nearly all the benefit.

Don, I see I didn't answer your question about where I think the bellmouth should be positioned.  My chamber is 5-7/8" deep, very close to yours, and I intend to start out with it 3" down from the top.  I don't intend to do any formal testing at various positions, but I will be watching for signs that stuff is by-passing.  Then I'll try a different position to see if it works better.  I have a feeling the 3" location is very close, if not exactly at the sweetspot.

Before I started the ampere tests, I cleaned my filter bag.  Boy, did the DC get  loud after that.  So, now I am going to start working on noise abatement.

[Don_Z]

I think it would be rather easy to frame up a wall with a door and then soundproof it with insulation around that little nook where you have it placed now.

[retired2]

I think it would be rather easy to frame up a wall with a door and then soundproof it with insulation around that little nook where you have it placed now.

The framing is piled up on my workbench and the solid core door is sitting in a corner.  Even the wall and door locations are marked out on the floor.  I'd like to start immediately, but I want to be sure I'm done messing with the stand. 

I really don't want to disconnect hoses to empty the waste drum, so I'm planning to suspend the top hat from the framing above.  It will be locked in place.  Then the drum will be lowered to empty it and then raised up against the separator to seal it.  Sounds great, but I haven't figured out an easy way to do it.  I've been thinking about either a wedge or cam lifting mechanizm.

ANOTHER INFORMAL TEST
This evening I threw a bunch of mixed dirt at the separator via my floor sweep.  It was a combination of fines from my radial arm saw and shavings from my thickness planer.  I don't have any sanding dirt to try, but everything went into the drum and nothing bypassed to the DC plastic bag.  It was a thing of beauty to watch.  The only thing I noticed is the finer dirt seems to make more revolutions before it all drops out.  The shavings seem to drop much faster. 

[retired2]

THE CASE FOR A "TWEAKED" RECTANGULAR INLET

When I was framing my rectangular inlet port, I made a small construction tweak that I failed to point out.  The amount of the tweak is dimensionally small enough that it is difficult to see in the construction photos I posted.

As Phil has often pointed out, the Thien separator depends on a fast, smooth, centrifigal air stream for optimum performance.  Since air in the separation chamber makes more than one complete revolution, it must converge with incoming air.  The design of the inlet port will determine how smoothly the two air streams converge.

Below is a sketch that illustrates the benefit of a rectangular inlet and the tweak I made.  The sketch is not to scale and it is slightly exaggerated to illustrate the differences between the three inlet possibilities.




The green lines (A) show a wider, typically round inlet port.  The green tangent line (A) shows that the recirculating air will converge with incoming air, and the far wall of the separator, at a more obtuse angle.  This will create more turbulance.

The red lines (B) represent a narrower rectangular inlet.  As you can see the tangent line, and the angle of convergence, is becoming more acute and therefore smoother.

The black lines (C) represent the small tweak I made.  When I faced glued the maple to frame my inlet port and capture the ends of the plexiglass (see my construction photos), I brought the curvature of the inlet framing slightly inside the circular circumference.  It was only about 1/8", but it has the effect shown by the tangent line (C).  It is slightly more acute and hopefully, the smoothest of the three examples.  In some very small way, this tweak may accomplish some of the benefits of the "scroll" design.

Caution:  More is not always better, so overdoing my tweak could cause a "ski lift" turbulance.

[Don_Z]

You need to get that Wynn filter on that bad boy!

[plauale]

Nice test write up. Concerning the blower inlet spin issue (since it is not really a problem), you could steal a handful of straws from McD's, cut them in half, pack them into your bell mouth, and set them in place with superglue. And if the blower doesn't immediately rip them out and shred them (probably not covered in the warranty) then you can see if there is more air resistance due to the straws that you lose to the inlet spin. I doubt there would be a significant gain.

Oh and to other readers who are wondering about my sanity, I did not pull the straws out of my @55, test wind tunnels usually have a honeycomb panel before the test section to linearize the flow. Straws are cheaper! All may doubt my sanity when I post pictures for my 3gal bucket prototype I just built for 99 cents. I'l save that for a different thread...

[RonS]

Here's what Bill Pentz has to say about the spinning air inside the cyclone vrs the spin of the impeller. This may not make a big differance in a basic Thien Seperator but could have some effect in a TopHat design, where the spin inside the tophat is counter to the spin of the impeller.

From Bill Pentz's website. (link below)
If the spinning air inside the cyclone hits the blower impeller going the wrong way, then the motor must do lots of extra work to reverse that spin and we lose considerable airflow. We need to make sure our cyclone, motor and impeller spins match! I prefer to build either a left or right hand cyclone to best match the shop in which it gets installed. We can easily figure out which way the air spins in our cyclone. The airflow coming into the cyclone will spin in the same direction that the inlet starts it. The air in my cyclone design spirals down then goes up without changing direction of spin. So a left hand cyclone will have its inlet coming out the back left of the cyclone which means we get a clockwise airflow. Most build left hand cyclones because that airflow matches most available impellers and reversed impellers are expensive and hard to find. If the inlet is located on the back right of the cyclone then the airflow inside is counterclockwise and we have a right hand cyclone. Both right and left separate equally well, but only if you match the motor, impeller and cyclone airflows.

http://billpentz.com/woodworking/cyclone/BuildCyclone.cfm

Ron

[retired2]

Here's what Bill Pentz has to say about the spinning air inside the cyclone vrs the spin of the impeller. This may not make a big differance in a basic Thien Seperator but could have some effect in a TopHat design, where the spin inside the tophat is counter to the spin of the impeller.

Ron

Ron,

My separator rotation is opposite the blower rotation, and I worried about that a little, especially since no one had experience with a bellmouth outlet.  I suppose if I were to build a mirror image separator, the ammeter might tell me if air rotation was affecting performance.  However, I don't have the energy or patience to do that right now.  Sorry, "plauale", I'm not up to trying the McD's straw test either.

One of the observations I noted from my ampere test was that the readings were lower with the separator removed than with the bellmouth installed.  I expected them to be higher.  I don't have an explanation for that, but it makes an argument that air rotation is not an issue.   

[Don_Z]

Due to the air coming in from the separator chamber to the outlet pipe and to the impeller being such a short distance I feel there is not enough of a span to allow it to spin at all. Also, with regards to the difference between a cyclone vs a top hat separator, is that the cyclone uses the entire chamber and impeller to achieve its separation vortex while the top hat uses the separator chamber to achieve its separation vortex BEFORE it reaches the impeller and WITHOUT direction of the impeller.  I thought of that issue before I began my build and designed accordingly, but the video of Retired2's build shows it works just as well as if it were built the other way. I feel that the utterances of bill pentz should be regarded as extreme overkill and a shameless point of his self promotion/ego/marketing campaign that unless we do just as he says and buy the products (clearvue, ect.) that he endorses, then we are all going to die a horrible and painful death with our lungs jammed full of sawdust. If you want to sell me a product, fine. But I do not give into scare tactic sales pitches.    ;D

[retired2]

Separator Update:

Noise:
I recall at least one person commenting that their DC didn't seem as noisy after they installed their separator.  That is not the case with mine.  The increased noise is a low frequency rumble.  The source seems to be my baffle and stiffener plate, and possibly the Brute drum.  When I turn the separator over and tap on the center of the baffle, it really seems to be "tuned" like a drum.

If I were rebuilding my separator, I would do what Don_Z did, i.e. a full thick bottom that is routed back from the baffle edge.  That would accomplish several things.  It maintains a thin drop slot edge, it eliminates the need for a support rod in the separator cavity, and it stiffens and possibly dampens the rumble from the baffle.

I think the plastic walls of the Brute drum may also contribute to the noise.  At the very least the light weight of the walls does not do a very good job of containing the noise.  I suspect the plexiglass walls also allow a lot more noise to escape than the heavy-walled designs.  Separator noise makes a good argument for using heavier stock for construction.  I think even a thin top plate could add to the low frequency noise.

I have mis-diagnosed the source of this noise.  See the update to this discussion in Post #92.

Paint:

I can't emphasize enough how bad the decision to use water-based paint was.  I have a very dry shop and I allowed the paint to dry for days before putting it into service, but it never seems to harden the way an oil based finish does.  The plug top that allows me to remove the bellmouth for testing has to be removed with a butt-chisel everytime because the painted surfaces continue to stick together weeks after the paint was applied.

Plexiglass:
Sexy, but a really bad choice.  It cracks too easily and this makes it difficult to secure and seal.  The coefficient of expansion must be very high because when I first cut it to length, it fit perfectly.  After I heated it slightly with a hair dryer to ease it into shape the length grew by at least 1/8" so that it would no longer fit.  After a short cool-down time, it fit perfectly again.  So, I am not sure what the temperature swings in my shop are going to do to the seal.

And with regard to seal, I did a smoke test of an area that looked a little suspicious and sure enough there was a small leak.  I have a feeling this is the result of the flexing that occurs from the on/off cycling of the DC, and possibly thermal expansion.  So, now I am going to do what I didn't want to do in the first place, i.e. caulk all the exterior edges everywhere plexiglass meets wood.