J. Phil Thien's Projects

Quote from: retired2 on July 18, 2017, 11:37:13 AM
I suggest you contact ??? and ask him how it is working.  Only people that have tried it know for sure, and I don't know of anyone who has tried it.

I'm guessing it doesn't work.  Good ideas spread fast, bad ones die quickly.

Unless you just love to experiment, I'd stick with a  design that works.

You mean Cosmas? I did contact him to ask where he got the design for the side opening in the outlet tube (and how it worked) but I didn't get a particularly clear response.

Quote from: nucww on July 18, 2017, 09:26:52 PM
Too much opening leads to too much flow in the lower collector.  I used a spiral design and only had about 30degrees closed. I was getting too much flow in the lower can.  The dust in the lower can was all in the center rather than the outside.  I closed it back to about 120 and it seems better.  I agree with R2 stay with the 120.  The turbulent entrance expands into the upper cavity before it has a chance to expand into the lower can.

That's a good point. I was thinking of a double height top hat design - meaning the inlet would be flush with the underside of the separator chamber ceiling. I assume then that airflow coming in from the inlet shouldn't immediately lift material from the lower can even if there wasn't a closed area underneath. However, an easily replaceable baffle would allow the addition of a 120 degree wider area.

Has anyone tried out different slot gap widths? I assume that a thin gap will plug with planer shavings, but is there a separation efficiency benefit of a thin gap for most normal cutting?

Quote from: dabullseye on July 19, 2017, 12:32:53 PM
i  just use a 1/4" hard board that is beveled back on the bottom side. the slot where it meets the 120 degree section has radius on the end of slot and is smooth. when r2 was building his top hat i had suggested maybe bolting a rod to bottom of trash can so the baffle could rest against it. so thinking of your idea  maybe you could make a sleeve that fits over the rod but has a magnet in it to hold baffle in place and then you could remove it to empty the can. but i believe in KISS method and would just go with the proven method and use the hardboard. ive never had a problem with mine

1/4" hard board would be easy to source, so that sounds fine.

I was thinking of a tapped hole in the top of the pipe for a bolt, which would take a matter of seconds to remove.

In my never ending quest to think about building a Thien Separator (but never actually getting round to making one)...

I noticed that Cosmas Bauer has a separator design that has a completely circular baffle (no 120 degree section that's connected to the side of the separator body), and instead the baffle is supported by the outlet tube (which comes all the way down from the top and touches the baffle). He's cut away a section of the side of the outlet tube, as obviously the end is sealed against the baffle.

I can see that would be useful in stopping larger material getting caught on the leading edge of the larger (120 degree) segment of the baffle (as there is no larger segment - just a gap all the way round). However, I assume that drawing air from a cutaway at the side of the outlet tube would cause a lot of unwanted turbulence.

The original Thien design (pictured at http://www.jpthien.com/cy.htm) uses three rods coming down from the separator's top plate (to support the baffle). I assume they will catch longer (e.g. planer) shavings, and also affect airflow.

So... what I'm wondering about is to create a baffle with a slot/gap all the way round (i.e. a perfect circle), but supported by a single rod or pipe from the center of the underside of the baffle, which would be attached to the bottom of the collection bucket; basically a bucket with a tall table (with a round table top) sitting inside.

I assume the air speed in the center column underneath the baffle should be low/zero (i.e. the supporting pipe shouldn't affect airflow), and with a 1 1/2" (38mm) diameter metal pipe the baffle should be fairly rigid. A tapped hole in the top of the pipe would allow the baffle to be easily unscrewed for tipping waste out of the bucket, and the air outlet above could be a nice open bell mouth.

Does that sound like a good idea, or have a missed something?

PS I understand that a thin material for the baffle is a good idea. With the baffle supported from the middle of the underside, it should sag less than when it's supported from the side by a 120 degree segment, so it would be interesting to find some really thin sheet materials (1/16" aluminium, or carbon fiber). With it attached by just a single screw from the top it'd be easy to test out different slot gaps and baffle thicknesses.

Quote from: phil (admin) on September 17, 2015, 08:50:06 AMIf you watch and of the youtube videos of clear-sided separators, you'll see that the debris actually falls quite quickly.

Indeed. I'm just assuming there must be reasons why Pentz added the protruding inlet. I did read it all some time ago, so I guess I should check again.

Quote from: BernardNaish on September 17, 2015, 08:57:17 AM
Here in UK it is possible to buy second hand steel bins that have been used to ship fruit juices. Sometimes they have small holes in the bottom to allow the plastic liners to be installed but these are easily sealed.

http://stores.ebay.co.uk/J-S-Simcox-Ltd?_trksid=p2047675.l2563

I'm worried that 60cm might be a bit too big, and, for various logistics reasons, I'm going to have to lift the drum/can I use so I don't want something too large - otherwise I'm going to struggle with it when it gets a bit full. I've found a 120 litre plastic container  that's claimed to be around 50cm dia and 74cm tall, so I'm making enquiries about that.

Quote from: BernardNaish on September 17, 2015, 07:13:09 AM
sploo, acryllic was called Plexiglass by one manufacturer.

I believe that retired2 came to the conclusion that the threaded rod he used to support the baffle was probably not needed. I do not like it being in the airstream, only because some long shavings might hang up on it. Your idea of a sort of scaffolding sounds OTT. Glueing a thicker piece of material to the underside while making sure there is at least 2" of clearance will certainly make it certain that it does not vibrate.

I do not think that angling the inlet duct downwards would improve performance very much and it just makes construction more complex.

Let us know how you get on.

Yep - plexy is just a brand name, like Vivek or Makrolon. The "glass" bit is indeed appropriate though  ;)

I worried that the supporting rod above the baffle face would catch shavings too; hence my thought of putting it underneath, but a supporting thicker sheet underneath would be much simpler.

I'm currently hunting for a suitably sized drum or trashcan, which is strangely proving to be slightly harder than I anticipated, but I may have a lead.

Quote from: BernardNaish on September 17, 2015, 05:50:16 AM
sploo,

Perhaps you have not seen this:

http://www.jpthien.com/smf/index.php?topic=563.msg3034#msg3034

You could leave the outlet of your fan rectangular and attach it to the inlet rectangle of the Top Hat as in retired2's design. If I have understood you correctly this means:

     20" overall diameter Thats fine
    14" tall (2x height of the 7" tall inlet) 1 1/2 x would do because gains not huge but may stop hang up of big shavings
    Side inlet mounted just under the top lid (called a "Top hat" design?)  your fan outlet tight to the top
    Inlet maybe protruding into the separator like a Pentz cyclone.  No. It will cause turbulence and degrade separation
    6" round outlet pipe, protruding ~10.5" down from the underside of the lid (inlet height + 1/2 of inlet height)  OK
    1/4" thick material for the baffle - preferably something very smooth               OK
    1.25" wide slot in the baffle         OK

I hope this helps and if so I would love to hear how well it performs......or otherwise.

That's a great looking build. I had flicked through it a while back when I was still considering cyclone vs Thien (and top inlet vs side) so other than being impressed I probably hadn't taken in enough of the fine details. I'm not surprised he had problems with plexiglass* though - polycarbonate or PETG would be much preferred.

By "Side inlet mounted just under the top lid" I was either thinking of tight to the top, or angled down a bit. Certainly it'll be easier to build it tight to the top, and as it's already rectangular there's no need for a round-to-rectangle transition, so that makes life a bit easier.

For supporting the thin baffle, I was thinking that you could have a rod protruding from the centre of the underside of the baffle (maybe 12" long), and another rod coming down from the middle of the solid part of the baffle (near the inside of the drum/trashcan) and then join them with a horizontal piece. Whilst it'd get buried in debris when the can gets full, it would keep the support well out of the way of the air flow above the baffle.

Otherwise, I was looking at his support (reply #6 http://www.jpthien.com/smf/index.php?topic=563.msg3034#msg3034, after the text "I had a circular piece of MDF waste, so I decided to make a stiffener plate for the baffle"). Would it make sense to extend that to the side wall (right hand side of the picture), but still stepped back from the baffle slot? I assume that some small amount of disruption to the air flow in that area on the underside of the baffle wouldn't be too detrimental, and may even reduce scrubbing.

* Also known as "acrylic", or "bl**dy h*ll the bl**dy stuff has bl**dy cracked again"

Quote from: BernardNaish on September 17, 2015, 01:42:15 AM
It has been noted many times that the baffle needs to be as thin as possible. Frequently 1/4" oil tempered hardboard has been used.

Thanks. The reason I asked was that I've seen a lot of builds using 3/4" sheet material for both the lid and baffle (more so than a thin baffle).

So... I'd have to make a "push" system due to the space I have, and the exhaust from my dust collector is roughly 7" tall x 5" wide (allowing for losses due to the rectangular shape, I understand it's roughly the equivalent of a 6" round pipe). This would form the inlet for the Thien separator. Assuming I can obtain a ~20" diameter barrel, I'd be heading for something on the lines of:

• 20" overall diameter
• 14" tall (2x height of the 7" tall inlet)
• Side inlet mounted just under the top lid (called a "Top hat" design?)
• Inlet maybe protruding into the separator like a Pentz cyclone
• 6" round outlet pipe, protruding ~10.5" down from the underside of the lid (inlet height + 1/2 of inlet height)
• 1/4" thick material for the baffle - preferably something very smooth
• 1.25" wide slot in the baffle

Does that sound sensible, and is there anything else I'm missing (such as the angle/positioning of the baffle slot vs the inlet location)?

I can vent outside so I don't need to run the outlet to filters. Does it matter then how far the outlet tube extends above the top of the lid? Could I just have a hole in the lid that I round over, or would it be better to have some amount of pipe protruding?

Quote from: phil (admin) on September 16, 2015, 04:54:39 PM
I haven't tried a helical ramp, the Pentz version really seems like a complete knock-off of the Donaldson-Torit and I've studied their approach a little.

I think we like to imagine that the helical baffle "trains" the debris to the outside wall where it is forced lower until the air reverses and the debris drops.

The reality is, though, that having that helical baffle means we've now basically made a winding sort of rectangular duct.  And if you've ever watched sims of airflow through a rectangular duct, you know there is a lot of weird turbulence within them, I'm not really sure how much they force debris to the outside wall.  Maybe the much heavier stuff but the much heavier stuff isn't the problem anyway.

Don't let me dissuade you, by all means give it a shot.

I think the idea was that it, also with the inlet being angled down slightly, encourages the dust to spiral down (i.e. more than just gravity) and so less of it ends up getting pulled into the outlet. Easy to say, time consuming to test, and probably very hard to perfect (assuming it's actually a benefit at all).

Reading through the "Here's the plans" thread, I see that a 1.25" wide slot in the baffle was mentioned for a 20" diameter separator, as well as some thoughts that a smooth surface on the baffle helps. I was wondering if a thin sheet for the baffle would also be beneficial - i.e. perhaps a 1/4" thick sheet of smooth plastic, rather than a 3/4" sheet of MDF or ply?

Quote from: phil (admin) on September 16, 2015, 10:10:39 AMThe 2x high design has increase separation a small amount and reduced pressure drop a small amount as well.

Thanks.

That gives me an idea (which may well have already been done)... Pentz's cyclones have the inlet protruding into the cylinder of the cyclone, and a spiral ramp to help direct air/dust down. With a 2x height separator it strikes me that you'd have the potential to do the same, as air/dust that makes it all the way around should then be below the level of the inlet (so you're not fouling the path of the air/dust, which you would with a 1x height design with a protruding inlet).

I guess the key is also the width of the slot in the baffle, and the placement of the slot vs the position of the side inlet. Are there any best practices for that? (e.g. 20" separator => 1" wide slot, 2x height => slot opening at 60 degrees after the inlet)

Quote from: phil (admin) on September 15, 2015, 06:43:46 PM
No, the cone wouldn't work, you can't really put a larger separator on top of a small drum.

The outlet tube should expend roughly 1/2 to 1D of the tube, so a 6" tube should extend 3" to 6" into the separator.  Exception is if you're doing a 2x high separator, in which case the inlet should be held to the top and outlet tube should extend past the bottom of the inlet by 1/2 to 1D of the tube.

Thanks Phil. I'll see if I can get a larger diameter can/drum then; I guess it would also make the build simpler. Whilst I'm struggling for height with a cyclone, that's obviously not an issue with your separator, so is there benefit to going for a 2x high design? I can see it might be useful in that you've got material hugging the outside wall long before it's dropped enough to get near the opening of the outlet tube, so might be beneficial to separation.

Quote from: phil (admin) on September 15, 2015, 02:54:54 PM
Side inlet has the least pressure drop, so that is superior.  With side inlet, distance between bottom of lid and top of baffle only has to be large enough to accommodate the height of the inlet.

The tighter the radius, the greater the separation, but also the greater the pressure drop.  I'd go with as large a D as your drum will allow.

Thanks Phil. How far down from the top should the exhaust pipe protrude (as a percentage of the lid-to-baffle gap)?

I have a fairly standard UK size trash can to hand, but I think it's only about 17" diameter at the opening. I assume that, within reason, once the dust has dropped through the slot in the baffle, what happens below isn't a big deal? I.e. if I made an 18 or 20" separator and had what was effectively a stubby cone underneath the baffle (simply to interface the larger separator to a smaller trash can) that would be OK?

Quote from: phil (admin) on September 09, 2015, 07:59:51 AMLOL, there is a shortage of friendly experts with access to simulation software.

Indeed!

Phil - returning to my original question about a separator with a 6" inlet, are there any best practice guidelines: i.e. for a 6" would it be better to go for a top entry vs side entry? Roughly what diameter of separator would be advisable? What depth/height should I use between the top plate and the baffle? How far below the top plate should the exhaust pipe protrude?

I'm not looking for a detailed drawing (though that would be great), I was more just hoping there might be some info/evidence on the lines of "18" dia, side entry more efficient, 8" gap between plates, 4" protrusion of exhaust pipe etc".

Quote from: phil (admin) on September 10, 2015, 04:32:33 PM
I think the problem is that CFD is time consuming, too, and there are errors involved.  You can find studies comparing CFD predictions to real-world (line wind-tunnel) tests.  Pretty interesting stuff.  Can take months of work to get close, and the CFD results improve because of feedback learned from empirical tests.

I suppose if CFD sims were accurate, outfits like Oneida and ClearVue would be beating one another over the head with it.

Yea, it's not really an accessible subject. Even if there were budget software tools available for the layman I suspect that few would be able to use them well enough to replicate reality.

Quote from: tomservo on September 08, 2015, 08:15:56 PM
I need to clean up my garage so I can make some proper sawdust, all I have right now is an enormous pile of hardboard sawdust made with a 120 tooth melamine blade. It's all I have right now and I hardly think it's a fair test to put a pile of super fines in front of it, but the result was quite a big puff of dust out the blower, no idea how much got separated but most of it, certainly.   I also don't have a hose for the thing either.

Bear in mind that, with pretty much any kind of separator (cyclone, Thien baffle etc.) if you bury a hose into a big pile of dust you're likely to overload the system and get some dust out of it. That seems to be the case of pretty much any one I've seen tested anyway. However, in normal conditions (e.g. a stream of dust from cutting or routing) they tend to work pretty well.

The killer is sometimes planer shavings, as the volume of material they that can produce can cause issues.

Quote from: phil (admin) on September 08, 2015, 09:05:34 PM
You're correct.  The key would be a vortex finder that matches the mains, and an outer pipe that is upsized for a length.  So if your mains are 6", the outer pipe would be 10" for at least some length.  I don't know what that length is, you obviously don't want settling before the wye.

Obviously need a lot more testing.

My ducting is rectangular, but now I'm just being an awkward troll  ;)

It sounds to me like you've got some really good ideas but are doing a lot of empirical testing. Would it be worth trying to find a friendly fluid flow expert with access to the relevant simulation software? It occurs to me that you could probably move things forward quite quickly with visualisation of the actual flows. I'm making significant assumptions that you don't already have that, given I don't know your background, I'll admit.

Quote from: phil (admin) on September 08, 2015, 03:52:03 PM
I'm including a pic of my "inline separator."  It has about the lowest pressure drop you can achieve.

Very clever. For a low pressure system such as an impeller based extractor, would the smaller pipe not kill the airflow though? I understand that low pressure air behaves more like water - i.e. incompressible - so forcing it through a smaller gap doesn't really work (unlike with a high pressure shop vac, which works fine).

Quote from: phil (admin) on September 05, 2015, 06:06:27 PM
Do we know pressure loss is much less?

The pressure loss should be somewhat proportional to the D of the separator.  Some of those newer cyclones are pretty tiny.

Mind you, I'm not saying you're wrong and I'm not trying to limit discussion, this is a free and wide-open forum.  And someone may have posted evidence that the pressure drop is lower somewhere (even here).

Just asking.

Based on the info at Bill Pentz's site it looks at though smaller diameter cyclones do suffer from significant pressure loss; quoting from his site "Instead of building my recommended 18" diameter cyclone, those with 3 hp motors should make 20" diameter cyclones and those with 2 and even 1.5 hp motors should use 22" diameter cyclones"

He does note that a 6" cyclone to his design will work well, but with a shop vac, which of course is a totally different proposition to an impeller based extractor in terms of suction (lots more) and air flow (a lot less).

With a 3hp motor it looks like I'd be pushing my luck with a 16" cyclone too, so I may well need to go for the harder route of making more space.

I would be interested in pressure loss data though, as I'm less worried about separation efficiency (venting outside, so mostly interested in just separating the visible dust - some smaller particles making it through don't worry me). So, I'd prefer low pressure loss and average separation to high pressure loss and really good separation.

Quote from: tomservo on August 26, 2015, 09:41:24 PM
Well, it's been a while with no replies but I'll give it a go.. I think 16" is too small for that much flow - if you run the math, with 16" OAD, that's 15.5" with a 1/4" wall, minus 6" for the outlet, probably more like 6.5" to account for whatever it's made of, and you only have about 9" left, so that's 4.5"of width for the incoming air. I'd bet the pressure loss will be huge, but with 3HP behind it, you may do all right.

Thanks. I'd guessed it might not work well.

The problem is that if I go to the effort of ripping out some existing infrastructure to get more space I would then have a chance of fitting in a Pentz cyclone design. I'm assuming that might be worth the difference in effort.

Quote from: dabullseye on August 27, 2015, 07:37:43 AM
i would not want a push system cuz stuff is still hitting the impeller. the reason i went to the tophat was i once sucked up a baby food bottle full of sanding sealer mix. it shattered when it hit the impeller. since then a few scotchbrite pads and a few pieces of sandpaper have been sucked up but they stayed in my tophat or fell into trash can below tophat. now imagine if a chunk of wood gets sucked up and gets wedged in impeller housing.

The motor+impeller unit I'm using is from a dust extractor that pushed air+dust into standard bags/filters so has a "material handling" impeller (i.e. one that can take a beating). It does actually have a guard/grill at the front of the housing to catch anything substantial. I.e. I'm fairly confident it'd be OK used as a push system as that's kinda what I'm doing now (just with no separator at the end).