J. Phil Thien's Projects

Thanks to Phil Thien's design and all of the other contributors to this forum, I purchased the DeWalt DW735X and constructed a separator build for it. Below are a few of the photos from this project during construction and the results from first two tests, both of which were better than any of my expectations.

Briefly, the build is from 1/8" hardboard, 1/2" birch plywood, an automobile air filter element, and a 10-gallon metal bucket. The design was predicated on the whole assembly fitting on the circa 1963 Craftsman radial arm saw mobile stand which I refurbished, beneath the table top thickness planer. The hardboard was used for not just for the primary baffle, but also was used for the vertical walls of the separator. This required the two 48" long strips to first be slightly dampened, formed, and dried prior to their lamination. Wooden dowels were used to maintain precise registration throughout construction and during final assembly. The first tests were performed with the full 10-foot length of hose and the assembly situated on top of my table saw to better observe any mishaps. The 8/4 poplar board of test one produced zero chips in the filter compartment with only a very little bit of fine dust visible after wiping. After a thorough cleaning, the second test used a birch board with identical results.

Howdy Phil,

Just a quick note of thanks for all of your excellent effort in designing the Thien Separator and for sharing it with the world. I've just completed a build for the DeWalt DW735, which after some tests this morning, works quite well - more on that later. Last night I was playing around with photographing it with a little bit of light painting and found this shot kind of fun and worth sharing with you and visitors to the forum. Not to spoil things for you, but the white arc is actually from my waving a small fluorescent behind the separator.  ;)

Experiment #1, Test 1

Not sure exactly what caused my first trial to pretty much utterly fail. I suspect that the hand plane shavings should never have been included while vacuuming the mess up off of my shop floor which has been building up since when I first started work on this endeavor, having the vacuum in various dissembled states. Still, maximum sized chips needed to be tested. These were normal size shavings from a 2" cutting iron that created a major clog and was part of this experimental first running of the Thien Separator that I designed to fit into my 32-year old shop vac - Sears Model No. 113.178081, should you care about such details. More details are shown in the attached drawings and photos.

Placing the completed Thien Separator assembly into the 16-gallon plastic barrel and aligning the vacuum head with its integral intake port and the 2" PVC fitting required some attention, but fit perfectly. By studying the dust pattern around this connection, there is evidence that the seal between the PVC fitting and the molded plastic lid was good. Likewise the weather stripping seals on both the top and bottom of the 9" x 2" filter did their job. Also, the filter did its job as it was quite packed with dust in the creases of the filter's paper folds. Soundly rapping the filter against the inside of the barrel loosened most of the debris but not all of it; I'll need to find somebody that has an air compressor and try blowing the rest of it out before Experiment #1, Test2.

https://www.woodcraft.com/products/trash-can-cyclone-lid

The great luck in particle separation afforded by the Thien Separator is well proven, largely through anecdotal evidence of not having to clean the filter for the original dust collection system or shop vac, sometimes for years. What remains to be evidenced of its success is the commensurate redesigning of the air filter. For example, in my 30+-year old Craftsman 16-gallon shop vac, more than 1.3 gallons is occupied by the filter.

So the question that follows: How much of the air filter's design criteria needs to be predicated on solely not creating any back pressure on the blower? Is there a standard equation used in sizing air filters which incorporates, among the various air flow components, a dirt factor (for lack of a better term)?

Are you familiar with computational fluid dynamics?
and if so,
Do you have software for performing these types of computations?

Portion of Hector's sketch (thanks Hector!) with drop slot shown highlighted in red. (Sorry about having used the attachment feature - I'm unclear on how to get images to display inline on this forum)

I've read that the the width of the drop slot (wds) is best suited at 1.125" and that an ideal diameter of the Thien Separator (labeled with an uppercase D in the sketch above) is equal to or larger than 20". I haven't read (yet) any discussion that mentions the wds being a ratio with D, while implicit in other readings, wds is always fixed at 1.125" irrespective of D.

Similarly, I have yet to read any discussion as to why the length of the drop slot (lds) is 240 degrees of arc. Phil Thien's original presentation; i.e., a separator lid, not the more popular so-called top hat, seemed to have been predicated on a shop vac, and a 2.5" round waste port inlet injecting from the top of the separator.

What miserable failure(s) has been discovered?; for example, using a continuous drop slot (full 360 degrees) and using a drop slot width of (D/2 * 0.1125)?

Phil Thien mentioned that he had gone through nearly fifty designs/variations during the course of approximately two years. Such persistence is inspiring with the results being most informative and very much appreciated! Equally informative have been other's experiments which have contributed to the validation of: rectangular section for the waste port inlet, elliptical bellmouths, and air straighteners. There's most likely more that I haven't learned yet as the posts here are numerous!

I've been wondering about the science of the baffle's thickness. I began to visualize it like a cutting iron on a hand plane only instead of shaving the wood, it's shaving the the particulates swirling in the wind. I'm wondering if introducing an angled edge might help improve the baffle's performance. The fibrous edge could be bolstered to withstand the particulate bombardment with some CA glue. Chamfering the edge would also allow the baffle's thickness to be increased for structural enhancement of the Thien Separator without degradation the of its performance, or at least that's the idea.

The angle of the chamfer, the material of the baffle, and the anticipated types of particulates abrading its edge make for further consideration. As a starting point, a 45 degree chamfer seems reasonably durable and any matter striking the chamfer surface would still be deflected into the waste bin.

The lower edge of the chamfer might also be eased to mitigate back edge turbulence.