I posted this in another thread but figured I'd create a new thread.
I'm including a pic of my "inline separator." It has about the lowest pressure drop you can achieve.
One thing you need to know... When a vacuum is pulled on a pipe, the air moves fastest through the center, slower at the edges. This means the pipe itself is acting like a separator. I can watch this with my clear pipe network, I see the bulk of the debris swirling around the outside and trying to stay out of the way of the faster moving air in the center of the pipe.
So in the pic I'm attaching, the gadget's left side is attached to my (or your) ductwork. On the right side is a smaller pipe to which suction is attached (I use a shop vac but this can be scaled-up to larger blowers). This extends through a sealed cap and past the end of the wye. This smaller suction pipe is the vortex finder.
So now what happens is that the suction forces the dust in the pipe into a spin. Once it arrives at the vortex finder, it can't reverse direction fast enough and passes the finder and continues into the Wye, where it is directed down into a discharge.
One downside to this method is that you have to size it so the largest expected bit of debris can fit between the inside, and outside pipes. That is true of systems using my baffle, too.
One other downside would be lower separation rates. I think. I'm not positive on that. There are quite a few variables like, how far should the finder tube extend past the wye's 45-degree leg? And on units designed to accommodate larger DC units, can offsetting the finder improve separation?
It works way better than anyone to which I've demonstrated it would have thought, though.
And there are numerous advantages. Takes up almost no additional space you haven't already lost. And limited ceiling height isn't a problem. Very little pressure drop, too. Not too difficult to build even in larger sizes.
I need more time to experiment with it.
(http://www.jpthien.com/smf/index.php?action=dlattach;topic=1194.0;attach=2997;image)
Anxious to see the results of further testing.
Where does the Wye discharge to? A trash can? Interesting I wonder if you could use this in series with your Thein separator to improve dust removal?
Quote from: clydeone on September 09, 2015, 12:28:40 PM
Where does the Wye discharge to? A trash can? Interesting I wonder if you could use this in series with your Thein separator to improve dust removal?
Yep, for that test I just had a long (36") straight tube with an end cap on it (makes fairly decent measurements possible). But ultimately, it would drop to a drum or box or whatever.
Pretty clever Phil, but I'm thinking about how that would scale up for use with a 1-1/2 hp DC. For example, I'm already plumbed with 5" pipe throughout for my Delta DC. So, if I installed a 5" wye just ahead of the blower I would only be able to use a 2-1/2" or 3" outlet. That would then transition to the 5" inlet size of the blower. I'm not sure how well that would work.
The other alternative is to use a 5" outlet pipe and replace all the plumbing with something like 8". Even if I were willing to do that, and I'm not, I don't think I could mantian enough velocity at that size to convey the waste.
Never the less, I'm anxious to see how your testing and fine tuning works out.
So if you're plumbed with 5", you'd need a short transition to 8" and you'd stick with a 5" vortex finder/outlet.
The transition from 5" to 8" could be fairly short and tapered.
I'm not sure when I'll have a chance to do more testing. Right now I'm pondering how to make some of the parts from off-the-shelf components.
Quote from: phil (admin) on September 09, 2015, 09:26:51 PM
So if you're plumbed with 5", you'd need a short transition to 8" and you'd stick with a 5" vortex finder/outlet.
The transition from 5" to 8" could be fairly short and tapered.
I'm not sure when I'll have a chance to do more testing. Right now I'm pondering how to make some of the parts from off-the-shelf components.
Oh, that might be possible for me to try without ripping out much plumbing, but I just don't have much time right now. Haven't even had a chance to install the Wynn cartridge filter yet.
Yep, join the club. I don't know how I can simultaneously be so busy yet getting nothing done! I think I'm working on all the wrong things.
Ditto here. Well at least I made a table...
very interesting idea.
I don't think it will work with a pipe that small.
I think you would end up with a clog
I will try this for my self soon and see.
im thinking that if you had the main line (say 4") in and out
then a 4" -6" reducer . then short length of 6" pipe . the dust will go between the 6" and 4".
then put a 6" to 10" reducer and a 10" T. then reduce back to the 4"
hopefully the 2" gap around the 6" pipe will be enough to allow the dust to freely fall down the T and into the barrel
i got a few hours today in the shop.
I tried this.
I got a great deal on some 8" dc pipe and fittings, the type that clips together .
this is great for this type of brain storming . you can swap bits around easily and indefinitely
my main dc extractor is not connected in my ne shop yet so I tried this with my 4" record dx4000 . its more like a shop vac than a chip extractor
I got an 8" y with a 6" branch, reduced the end down to 4"
I got a blank cap and drilled a 4" hole in it and put a short pipe in it
I extended the pipe until it was inline with the start of the branch
the branch Is going towards the inlet side so that any dust goes down the pipe
I didn't empty the dc so I couldn't exactly check everything
I got a bucket of chip and dust (roughly 50:50)
I slowly fed it into the a 3 foot length of flexible ducting I had ( I found having a pipe before the separator worked better than just feeding it straight)
I fed it painfully slow
I worked out that about half the mixture bypassed the separator
probably 3/4 of the chips and 1/4 finer dust were separated
I will upload a pic or 2 tomorrow when I get home
overall it was a success . my gut tells me that it is more suited to chip separation rather than fine separation
I think you could hit 80-90% separation on larger dust but I feel you would be lucky to see 25% for fines
there is a lot of potential in this idea but it will need a lot of fine tuning.
I rearranged it a bit later trying to improve it and got almost total bypass.
Quote from: alan m on September 13, 2015, 05:31:55 PM
i got a few hours today in the shop.
I tried this.
I got a great deal on some 8" dc pipe and fittings, the type that clips together .
this is great for this type of brain storming . you can swap bits around easily and indefinitely
my main dc extractor is not connected in my ne shop yet so I tried this with my 4" record dx4000 . its more like a shop vac than a chip extractor
I got an 8" y with a 6" branch, reduced the end down to 4"
I got a blank cap and drilled a 4" hole in it and put a short pipe in it
I extended the pipe until it was inline with the start of the branch
the branch Is going towards the inlet side so that any dust goes down the pipe
I didn't empty the dc so I couldn't exactly check everything
I got a bucket of chip and dust (roughly 50:50)
I slowly fed it into the a 3 foot length of flexible ducting I had ( I found having a pipe before the separator worked better than just feeding it straight)
I fed it painfully slow
I worked out that about half the mixture bypassed the separator
probably 3/4 of the chips and 1/4 finer dust were separated
I will upload a pic or 2 tomorrow when I get home
overall it was a success . my gut tells me that it is more suited to chip separation rather than fine separation
I think you could hit 80-90% separation on larger dust but I feel you would be lucky to see 25% for fines
there is a lot of potential in this idea but it will need a lot of fine tuning.
I rearranged it a bit later trying to improve it and got almost total bypass.
Sure wish I had the time to jump in and help with this new concept from Phil. It is really intriguing and what attracts me most is Phil's comment about the lower pressure drop. I would gladly trade some fines by-pass for lower line loss.
I've started accumulating some parts for more experiments on my end.
I have a couple of ideas for maximizing fines separation but need to get my test rig built.
I think to fine tune this idea you really need clear sides.
one thing I noiticed is that you need a long straight pipe before the separator
if you feed straight into it there is no time for the dust to go to the outside
there is no centripetal force helping put the dust to the outside
Quote from: retired2 on September 13, 2015, 05:45:22 PM
Quote from: alan m on September 13, 2015, 05:31:55 PM
i got a few hours today in the shop.
I tried this.
I got a great deal on some 8" dc pipe and fittings, the type that clips together .
this is great for this type of brain storming . you can swap bits around easily and indefinitely
my main dc extractor is not connected in my ne shop yet so I tried this with my 4" record dx4000 . its more like a shop vac than a chip extractor
I got an 8" y with a 6" branch, reduced the end down to 4"
I got a blank cap and drilled a 4" hole in it and put a short pipe in it
I extended the pipe until it was inline with the start of the branch
the branch Is going towards the inlet side so that any dust goes down the pipe
I didn't empty the dc so I couldn't exactly check everything
I got a bucket of chip and dust (roughly 50:50)
I slowly fed it into the a 3 foot length of flexible ducting I had ( I found having a pipe before the separator worked better than just feeding it straight)
I fed it painfully slow
I worked out that about half the mixture bypassed the separator
probably 3/4 of the chips and 1/4 finer dust were separated
I will upload a pic or 2 tomorrow when I get home
overall it was a success . my gut tells me that it is more suited to chip separation rather than fine separation
I think you could hit 80-90% separation on larger dust but I feel you would be lucky to see 25% for fines
there is a lot of potential in this idea but it will need a lot of fine tuning.
I rearranged it a bit later trying to improve it and got almost total bypass.
Sure wish I had the time to jump in and help with this new concept from Phil. It is really intriguing and what attracts me most is Phil's comment about the lower pressure drop. I would gladly trade some fines by-pass for lower line loss.
I presume there is less pressure drop because the air stays straight and doesn't have to spin around like in a normal separator
the reason for putting the baffle separator before the dc is to stop the larger chips hitting the impeller
if this separated them out then we could put a baffle separator that is designed for fines and maybe increase the overall efficiency of the system
Quote from: phil (admin) on September 13, 2015, 08:38:07 PM
I've started accumulating some parts for more experiments on my end.
I have a couple of ideas for maximizing fines separation but need to get my test rig built.
im looking into getting parts so I can try this with 8" lines
im thinking of getting 12" y and reducers but they are expensive here in Ireland.
what ideas have you got.
This idea is exciting. I am too ill to get to my work shop just now or I would be there playing and measuring but it does give me time to think about it.
Some of us use DC systems fitted with very fine filters to keep dangerous wood dust out of our lungs. The important but secondary function is to keep the shop and machines clear of shavings, chips and dust. A Thien baffle or a cyclone is used to remove the dust to keep filters as clean as possible. This keeps the air flowing at effective levels for longer before having to clean the filters.
Separating out fine dust takes precedence over coarser debris.
With the in-line separator (ILS) is the space between the vortex finder and the outer "Y" fitting equivalent to the width of the slot in the Thien baffle? If so the narrower the gap in an ILS the more efficient the separation of fines!
I do not think the debris is hiding from the air so much , it is more likely being elbowed out of the way by the greedy air slamming down the pipe, pushing the "weaker" particles out of the way – the bullies.
My second question is: - do the fines get spun closer to the pipe wall than the coarse? Clearly large lumps will be prevented from getting as close to the wall as fine dust because of the curve of the wall. Secondly each particle of dust weighs less than a larger piece of debris so presumably larger particles take more energy to move them? Hence are larger bits more likely to stay towards the centre of the pipe? If this happens does it mean that a cross section of the vortex would show an outermost layer composed of fine dust getting progressively coarser towards the centre?
It is not as simple as that as there are many other things going on. Friction, surface areas,etcetera, etcetera. It may be that the only way to find out is to "look" at some mixed dust and shavings spinning down a pipe. I cannot get to my shop to do it myself so if anyone knows for sure I would be gladdened to hear from them.
Quote from: alan m on September 14, 2015, 05:01:51 AM
what ideas have you got.
Bunches of them. I've been pondering some and realize how terrible they are. By the weekend I'll hopefully be able to try a few out.
Quote from: BernardNaish on September 15, 2015, 09:50:51 AM
I do not think the debris is hiding from the air so much , it is more likely being elbowed out of the way by the greedy air slamming down the pipe, pushing the “weaker” particles out of the way – the bullies.
My second question is: - do the fines get spun closer to the pipe wall than the coarse? Clearly large lumps will be prevented from getting as close to the wall as fine dust because of the curve of the wall. Secondly each particle of dust weighs less than a larger piece of debris so presumably larger particles take more energy to move them? Hence are larger bits more likely to stay towards the centre of the pipe? If this happens does it mean that a cross section of the vortex would show an outermost layer composed of fine dust getting progressively coarser towards the centre?
It is not as simple as that as there are many other things going on. Friction, surface areas,etcetera, etcetera. It may be that the only way to find out is to “look” at some mixed dust and shavings spinning down a pipe. I cannot get to my shop to do it myself so if anyone knows for sure I would be gladdened to hear from them.
some good insights Bernard.
im not sure what the relationship between slot size and the difference in diameters of the 2 pipes is yet.
I guess that you want the dust traveling as close to the outlet pipe as possible so that the dust just slips by . in the tests I did I used a 4" inlet and outlet with a 8" y. that 2" all around is probably too big for fines. most of my fines were bypassing.
Quote from: phil (admin) on September 15, 2015, 02:43:29 PM
Quote from: alan m on September 14, 2015, 05:01:51 AM
what ideas have you got.
Bunches of them. I've been pondering some and realize how terrible they are. By the weekend I'll hopefully be able to try a few out.
come on phil . your teasing us
at least give us some of the bad ideas.
there might be some value in them ,
or at least tell us why you think its not worth trying
Don't get your hopes up, this entire idea may turn out to be one of my worst.
Quote from: phil (admin) on September 17, 2015, 06:55:22 PM
Don't get your hopes up, this entire idea may turn out to be one of my worst.
keep your hopes up. all visionary's think their ideas are rubbish.
my gut tells me that this separator wont ever be as good as a top hat separator but from my very rough playing with parts I had there is definitely merit to it . I got 50 % separation on the first decent try.
when all the brains come together and test the variables we should see a major increase in the efficiency.
Many thanks to all the contributors on this site. I have been planning a top hat and some thoughts have crossed my mind on this idea. I think the wall friction is why the dust tends to collect there. The dust has more frictional losses than the air. To enhance this design a spiral vane (propeller of sorts) inside the feed pipe will tend to spin the dust to the outside before it gets to the smaller pipe separator. Depending on the extreme of the spiral, flow straighteners may be needed in the smaller receiving pipe to minimize pressure loss. The other thought would be that the collector Y be much larger but retain the narrow clearance where the small end of the pipe is. The larger volume will take the wind out of the dust and the back pressure may be less. Also, the Y could be tilted so that the cavity of the Y uses gravity to keep the dust flowing down the Y.
anyone get time to try it yet
I was planning to do a top hat, but this has me intrigued. I do not have much spare time and may not have much time to experiment. I may try a prototype that's a little different than whats proposed. The question I have is if the air speed is reduced below ~45 mph, the theory says that the dust will not be picked up by the air flow. So if the dust is on the wall traveling a speeds ~20 mph and the air flow is diverted from the wall to the center at speeds of ~20 mph, will the dust get pulled away from the wall or does it stay close to the wall if the pipes are vertical? I also have concerns that with internal baffling (the spiral vanes) in the pipes that clogs may occur which means the design must come apart easily. Any thoughts appreciated.
what prototype changes are you thinking about
A modification that I have been considering is to use a 5" base pipe and expand to a 10" wye. Before the transition to the 10" pipe, place a 3" pipe inside the 5" pipe after the spiral vane. The 3" pipe goes into the 10" wye. At the end of the 3" pipe is a 5" or 6" pipe that goes out of the other end of the wye. At the transition from the 5" to 10" pipe add an internal transition on top of the 3" pipe that maintains a 1" channel between the outer transition and the inner transition. When it is about 8" in diameter, collapse it back down to the 3" pipe inside the 10" wye. If the dust stays on the outside surface, the dust flows outside the 3" pipe expands to 10", hopefully stays on the outside of the pipe and falls outside the 5" or 6" pipe with very low air flow and air flow without the dust is always moving forward and goes to the outside area between the 3" and the 5" to 6" pipe and hopefully has minimal pressure drop. Without the transitions, I do not think there will be enough airflow to get the dust to the outside of the 10" pipe. With these clearances, my main concern is dust hangups (on the spiral vane and the 1" channel) requiring frequent internal cleaning which means more design features to take it apart easily.
Quote from: nucww on October 19, 2015, 10:12:48 PM
A modification that I have been considering is to use a 5" base pipe and expand to a 10" wye. Before the transition to the 10" pipe, place a 3" pipe inside the 5" pipe after the spiral vane. The 3" pipe goes into the 10" wye. At the end of the 3" pipe is a 5" or 6" pipe that goes out of the other end of the wye. At the transition from the 5" to 10" pipe add an internal transition on top of the 3" pipe that maintains a 1" channel between the outer transition and the inner transition. When it is about 8" in diameter, collapse it back down to the 3" pipe inside the 10" wye. If the dust stays on the outside surface, the dust flows outside the 3" pipe expands to 10", hopefully stays on the outside of the pipe and falls outside the 5" or 6" pipe with very low air flow and air flow without the dust is always moving forward and goes to the outside area between the 3" and the 5" to 6" pipe and hopefully has minimal pressure drop. Without the transitions, I do not think there will be enough airflow to get the dust to the outside of the 10" pipe. With these clearances, my main concern is dust hangups (on the spiral vane and the 1" channel) requiring frequent internal cleaning which means more design features to take it apart easily.
We're on the same page, someone just needs to do some testing.
For me right now I believe the top hat is going to be the solution for me and and don't have time to make two systems. I'll continue to feed ideas for this for now. Maybe later during the holidays when I'm off work I may experiment.
This type of technique has been used for well over 50 years in some chemical processes. It is called turbannular stripping. It is well documented in patents. Normally it is used to remove liquid and solid particles from a liquid/solid/vapor mixture. Provided the velocity if the mixture is high enough at the inlet of the turbannular stripper the liquid/solid phase will stick to the wall of the tube and the vapor will retain it's high velocity. The turbannular stripper then feeds a cyclone separator, the liquid/solid phase stays together and goes out the bottom while the vapor phase goes out the top of the cyclone. The geometry of the tube is critical with the diameter small enough to keep velocity high enough and the length of the tube sufficient to establish and maintain a donut of liquid/solid on the wall before the entrance to the cyclone.
Now that is interesting! I love it when we re-discover something when we come at it from a different direction.
Are there any on line sources of information about this?
See page 452 figure a
https://books.google.com/books?id=1W64gOffM7EC&pg=PA452&lpg=PA452&dq=turbannular+stripper&source=bl&ots=7W8Jf9sE9h&sig=wvWEqOuK0OEUfRE8YZKRyc8CUpM&hl=en&sa=X&ved=0CCMQ6AEwAmoVChMI8PK3k6-DyQIVxBseCh0AIQTk#v=onepage&q=turbannular%20stripper&f=false
The turbannular stripper is also capable of doing organic extraction when the vapor medium is something like steam. The steam will boil off a volatile material in the stripper and allow an easier separation of a water and organic mixture.
Thank you. Bit above my pay grade but I will struggle through.
It occurs to me that if the inline separator can be developed to remove a significant quantity of the finer waste without wasting very much of the DC suck, then we could put one in front of a Thien top hat.
I think this may result in an overall increase in the removal of fines with the bonus of being able to increase the Thien slot width so deducing blockages there.
I am doing less and less machine work but this whole waste/dust collection theme fascinates me.
This is summary of my dust collector that I spoke of in an earlier response. Most of the design and credit comes from the contributors of the topic:5", Rectangular Inlet, Bellmouth Outlet with Air Straightener, Top Hat Separator.
This is being used with a HF DC so the design was trying to minimize pressure drop to maximize flow.
Some of the significant differences are:
- spiral chamber with the spiral starting ~6" before chamber entry. The spiral lost ~2 1/4" per revolution.
- a high narrow rectangular inlet (10"x2.25") so that entry dust is already close to the outer surface
- Plexiglas on the top and formica for the spiral sides
- Baffle slot was extended to ~320 degrees since the spiral design should minimize turbulent entrance.
- a 3/4" MDF baffle with the slot edges tapered to ~1/4".
- 5" outlet pipe with a X flow straightener
- a commercial bellmouth outlet pipe with the flat part of the rim cut off resulting in ~7.25" OD mouth for a 5" base pipe
- flow diverters at the bottom of the wide mouth to divert the flow up rather than spinning around. Some CFD simulations show that vortex in the center is huge and is likely causing significant pressure losses.
I had a anemometer to measure airspeed to test some of the differences.
With the standard HF setup, 10' HF flex pipe and a 4" to 5" transition, the airspeed at the surface of the 5" transition was 48 mph. (airspeed with a 4" pipe was off scale max of 67 mph) With the same piping and with the dust separator installed, the speed was 45mph. So this design only lost 3 mph.
Installed a .5micron wynn filter, put the motor on top of the collector, patched some leaks in the impeller housing, and put a foam insulation strip between the bottom bag and the housing. With the dust separator installed, the speed improved to 51mph with the same piping as the previous test. So getting rid of the 5" flex pipe and adding the extra area of the wynn filter gained about 6 mph as noted by others on this web site and elsewhere giving me a net gain with the dust collector of ~3 mph .
I used hvac 5" ducts for the main line and all drops from the ceiling are 4" pipe. I made my own wyes since the commerical ones were expensive and the cheaper hvac ones had the wrong end crimped. The longest run of 5" was about 16'. For all 90 degree turns, I used 2 elbows at 45 to make more gradual turns. After installing the pipe I tested the longest run of 5" pipe and the speed was only 38 mph. Taping all the joints made a huge difference. The speed was now 54mph. A gain of ~3 mph changing the 10' - 4" flex pipe with the metal duct.
I allowed a cup of flour to be sucked into the inlet to the collector and ~ 2/3 of it ended up in the trash can. There was some visible in the chamber and the edges of the trash can. Some made it to the bag after the impeller.
Attached is a picture that better shows what I am describing. Evidently, I can only send one of these at a time.
I agree. I'll copy it so it doesn't clutter this one. This was the forth time I time to post it and I did not know why it was not working so I added it to an existing thread. While doing this I got a message saying that I could only load one picture of this size at a time.
Trying to post an image. If the swirl created by the impeller is great enough [TPI] then perhaps you have enough of a vortex to cast particulate to the outside of the tube. So wondering if a 6" or so tube, long, with the fan entrance at the top per this setup may cause the particulate to first flow outside due to the elbow inlet, continue to swirl enough to rise and exit the slot cut into an outer tube and fall down to a container. If the container is airtight, then there is no pressure differential or flow and same as your Thien baffle, will cause separation. You may have simplified the Thien baffle as a pseudo reverse cyclone arrangement inside-out! It would be easy to support the tube lower than the exit, therefore nothing in the way to create any friction or other turbulence.
Just throwing it out there for you, could not sleep last night.
Just realized - not clear - the [intake] inner tube is 6" diameter minimum, the length is much greater in order to allow centrifuge separation.
Good night :)
Well, you are making me crazy - I may have to build this. It would be SO simple.
A few more images, supports for the center intake tube, flange for the lower can, removable top hat to the fan.
Easy to check, easy if needed cleaned [can't actually see a reason anything would need to come apart]. I like it. I may just abandon my more complicated build and do this first. Thanks for the prod.
It should not require air straighteners because we would want the impeller to impart the vortex. Straighteners at the entrance - BEFORE the elbow may help in hoses or pie runs to curtail some circular flow losses, streamlining the flow to the device. Once past the elbow, particulate not thrown out from the elbow and now spinning vortex imparted by the impeller, would then be encouraged to the outer inner pipe wall and then thrown out at the NOW 360 degree slot with no re-entrainment to fall through the outer pipe wall to the bin! Could it be possible? The center of the vortex is somewhat negative, perhaps like the eye of a hurricane. So if the very light fines are not washed out, they may make it to the impeller and thus the filter. But it may work better than any other alternative?
Perhaps I am just too tired and will try to stay away until I sleep.
We know the effects of the bell-mouth, and how akin to a distorted bubble the air flows out between two opposed bell-mouths as two mirrored black holes if in a constant state [unit on and vacuum established]. So I am wondering if the spinning vortex of this may yield the separation I seek, allowing the dust to escape into the outer chamber, if not all, at least I am believing it would be a very low loss first stage? I really think I need to try this. Maybe I have become not a bellmouth, but a blabbermouth due to the fact I have not slept today. Like all good things in life, I may just end up regretting it in the morning :o Go ahead, trash me :)
Just wondering if any one had any luck with this? I'm looking for a pre separator for planer shavings and had hopes for this design.