J. Phil Thien's Projects

Call me thick headed (I don't mind  :) ), but I'm having a little difficulty visualizing how the filter would be situated in the separator chamber. It's not clear from your description, but it sounds as though the filter is the same height as the top hat (12 inches), so I picture it as sandwiched (and sealed) between the baffle and the top plate of the separator. If so, then all outlet air would pass through the filter side walls.

From what I've picked up on this site, I see a couple of potential problems with the design. First, I assume that the filter has exposed pleats, which will likely create a lot of turbulence, so will impair performance.

Second- and more critical- if air is passing through the entire height of the filter, you're going to get a lot of bypass: air (and dust) that will hit the filter before they get a chance to separate at the baffle.

Third- the transition from 4" round pipe (~12.6 sq in area) to a 1/2" slit (if 12" high = 6 sq in) sounds like a real bottleneck, creating a lot of resistance to air flow. Please verify my math, but 600 m3/hour =  10 m3/min = 353 cfm, which is just adequate for 4" duct. I'm not sure, even with the great vacuum your blower can draw, how much that resistance will affect performance.

Finally, to me the thought of disassembling a separator to clean or replace a filter, then making sure that its airtight on reassembly seems like a real chore.

Sorry, I don't mean to be a Debbie Downer- just want to throw out some potential pitfalls as I see them. If you have the time and energy, It would be interesting to see how your design actually works. Good luck!

Okay, Colin, there are more experienced and wiser heads in this forum, but until they jump in, I'll make a few comments based on what knowledge I've picked up along the way.

1. I think a fiberglass baffle would be fine- in fact probably better than a lot of builds. Ideally, the thinner the baffle the better, esp. at the baffle opening. Many builders bevel the bottom side of the baffle slot edges (the top surface should be entirely flat). The drawback of a thin baffle is that it might need support, typically by hanging rods or dowels from the top lid.

2. The air entering the separator ideally flows smoothly around the outer edge, so the best way is to have the inlet as close to the edge as possible, with the entering air tangential to the circumference. (BTW, this is a major argument for a so-called "top hat" collector, which sits on top of the waste container- it allows entrance from the side, rather than the original design, with an internal inlet 90 degree bend inside the separator). If you angle the inlet opening toward the outside, make sure that air flow isn't likely to "ricochet" off the side- that is, that it isn't at a big angle.

3, 4. If you must go with a 3" inlet, I'd position it near the top, rather than the bottom. As I understand it, the more time the air flow can spend circulating around the outer edge of the collector (before being sucked up the outlet in the center), the better the dust separation. With the inlet at the top, the air flow has a chance to spiral down the edge, theoretically allowing a longer path for separation.

However, the air flow should hit the baffle in the solid section: the 120 degree section without the slot. If the inlet sits low- right on top of the baffle, the solid section begins right about at the inlet. If the inlet is higher, with air flow spiraling down the edge, the ideal placement of the solid part of the baffle is more complicated.

5. 3/4" sounds good- narrower slot traps smaller particles, but is more likely to clog with large pieces of debris (wood shavings from a planer, for example). If you're dealing mostly with small particles (dust), go with a smaller slot width.

6. IIRC, I've seen some threads here suggesting that corrugation may actually improve separation. Size-wise, however, a 16 inch diameter collector sounds a little small for a (nominal) 1200 cfm collector. If you're willing to invest a little more effort, I think you'd get better performance by making a top hat separator to fit on something like a 32 gallon Rubbermaid Brute container. Plenty sturdy, and allows a 21" wide separator. With that, you can go with a 4" inlet; the 3" inlet will really constrict air flow.

Check out retired2's thread for a lot of information, particularly on design and build of a top hat collector.
Final caveat: the above suggestions may be completely bogus! I hope that smarter contributors will confirm or refute. Good luck! John

Sorry to hear about the results of the impeller's maiden voyage- I hope there wasn't any substantial damage or injury involved.

It does remind me of centrifuge calculations I did in my former life working in a hospital lab. I estimate from your pictures that your impeller diameter is about 20 inches (508 mm). The equation for relative centrifugal force is RCF (x g) = 1.12 x radius (mm) x rpm.

At 2850 rpm, your impeller was pulling more than 2300 g's. Just something to consider in deciding how sturdily you want to make your design- or whether you want to continue in that vein (or vane? :) ). Just saying, I think there's a reason most impellers are welded or cast metal. Also, precise balancing is critical.

QuoteFinally, if I remove the restrictor ring in the 50-850 and only connect to a single machine via 6" flex pipe, would I risk damaging the motor with the configuration I've described?

I think it's safe to remove the restrictor ring. I assume you're referring to the sheet metal disc that's spot welded just to the inside of the impeller intake, which reduces the intake diameter. I removed the ring in a 50-850 that I had, by drilling out the welds, and had no problems.

It's my understanding that the ring was added in later versions of the 50-850, as a means to prevent possible motor burnout resulting from numbskull user behavior. By that I mean running the DC with no duct work attached, and possibly without a filter. With no restriction to air flow, the motor will move a lot of air, and draw a lot of amps- so risk burning out. The restrictor prevents unimpeded air flow, but also impairs performance.

As long as you have a filter and a duct attached to your DC, you should have enough restriction that motor burnout isn't an issue, and the restrictor ring isn't needed, or even desirable.

Your first question is a little confusing to me, in part because of some imprecise and possibly mixed up terminology. I think that the term "drop" usually refers to the distance between the bottom of your outlet (center) pipe and the surface of the baffle. The rule of thumb for that is 1/2 the diameter of the pipe. So, with a 4" pipe, there should be a 2" gap between the outlet pipe and the baffle. This position is a compromise. With a higher pipe outlet you'll get better air flow, but that's because more air will go straight from the inlet to the outlet, without swirling around the perimeter of the separator, so more dust will escape (bypass). A lower pipe would potentially give better separation, but a narrow gap between the baffle and the outlet will restrict air flow, and result in degradation of performance.

I think you're confusing "drop" with "slot width", which is the width of the opening that arcs 240 degrees around the outer edge of the baffle. The "standard" slot width is about 1-1/2", which is also a compromise. A narrower width (3/4" or 7/8", or so) should give better separation (collect more dust), but it's more likely that larger particles, such as shavings or long strings of wood (planer and jointer) will clog the slot. A wider slot won't clog, but won't separate as well. If you're only generating fine dust, try a narrower slot width.

As I understand it, the orientation of the inlet opening (elbow) to the baffle slot is complicated. My read on it is that you want the air just leaving the elbow (just entering the separator) to "land" on a solid section of the baffle. If the bottom of your inlet is at the same level as the baffle, then the inlet opening should be at the beginning of the solid (120 degree) section of the baffle.

If the inlet opening sits over the slot- or the distance between the inlet and slot is less than 120 degrees, more of the rapidly moving air just entering from the inlet will go down into the waste collection section below the baffle, and will stir up (scrub) and resuspend the dust below, and impair good separation.

With some designs, the separation chamber is taller than the inlet, so there is a vertical distance between the bottom of the inlet and the baffle surface. The air flow entering the separator will spiral downward, and may or may not "land" on the baffle where the solid section of the baffle is located. I dimly recall that Phil Thien posted about using chalk dust to trace air flow. Check out the posts, though, before trying chalk, though- it's bad for cartridge filters.

I'm a relative newbie myself, so some or a lot of the above might be bogus. Maybe wiser heads will correct or corroborate. Just my .02

Okay, I assume that the intake to your DC is on the bottom (opposite the motor). Looking into the impeller from the intake, the impeller spins counterclockwise, so the common wisdom is that you want the incoming air to "match" the impeller spin. My personal image of it- as a non-engineer- is somewhat like trying to jump onto a fast moving merry-go-round" it's a lot easier and less traumatic to hop on if you run in the same direction as the merry-go-round is spinning. The left side separator would match your DC in this case.

It just occurred to me that rather than describing rotation as clockwise or counterCW, which depends on your viewpoint, it might be easier to describe and visualize by labeling flow as either right- or left- handed thread. Your DC matches left thread. Just a random thought.

Okay, I'll give my .02, as a relative newbie myself, until wiser and more experienced posters jump in to elaborate a/o correct.

1. With a "standard" separator, the air exiting spirals in the same direction as it enters. Common wisdom has been that if the spiral is the same direction as the rotation of the DC impeller, air flow will be improved to some degree, esp. if the distance between separator and impeller is short.

However, there are data which suggest that a straight (non-spiral) air flow produces the best impeller efficiency. To that end, now many builders are putting "air straighteners" in the exit port of their separators, to eliminate spiral air flow. For a lot of info, check out retired2's extensive thread.

2. I'm not aware of any hard and fast rules re: cyclone size and DC power, other than the larger the separator, the more it robs DC performance, so more power is needed. That said, 32 gal Brute seems okay for a 1 hp DC; I wouldn't go any larger, assuming you're making your separator diameter about the size of the Brute.

3. Other than size-DC performance penalty, no optimum size. With a top hat design, you can put a smaller diameter collector on a larger waste container. Container size then depends on convenience (weight, difficulty emptying, etc), available space, cost, etc. A taller waste container is better than a stubby one: as the dust pile rises and gets closer the the bottom of the baffle, more of it will be scrubbed and sucked back up into the separator.

Okay, Dennis, I'll add my $0.02, or the equivalent in Finnish markka. Your most recent post clarifies your plan.

First, as far as duct size. The general rules for maximum performance are to keep duct runs short, wide, and straight as much as possible. That said, assuming you have several dust producing machines you want connected, the typical way is to go with a large diameter trunk line, with smaller branches to individual machines. With a 4 kW blower (16+ amps, I assume), I wouldn't go with less than 6" main duct size, or 4" for tool connections.

I think your major challenge is getting your impeller set up appropriately. As much as I admire Mr. Wandel's skill and creativity, I would be extremely leery of scaling up his wooden homemade impeller to fit a 4 kW motor- it just looks like a disaster about to happen. Design of an efficient impeller (at least to an amateur like me) appears very complex. Maybe you can find a used impeller of appropriate size with a burned out motor, or check out the Cincinnati Fan web site- there's a ton of information there. In the end, it may be more cost effective to get an intact DC blower and find another use for your motor.

I see both BernardNaish's and your points of view regarding dispersal of fine dust to the outside. It really depends on your locale. If your shop is in a more rural, less densely populated setting, fine dust will either settle to the ground or be so diluted that it causes no significant health risk, esp. compared to other airborne hazards. If you live near others, though, it would be more neighborly not to discharge dust in their direction. It begs the question, which I'm not sure anyone can answer, as to how close is too close.

Disclaimer: I am by no means an expert in anything! Good luck with your project.

Based on what I've gathered about DC and related topics:

1) When you state "before" the separator, do you mean as a standalone build, as opposed to one that's built inside the DC (just below the intake port and above the waste bag? If so, I suppose it has some additional effect on performance, but I can't say how much. OTOH, with it before, you avoid most of the chips (and more dangerous pieces) from reaching the impeller. Also, emptying a trash can is usually easier than those bags.

2) IMO, there's not much readily available (certainly for the price) that can beat a Wynn filter for collection efficiency, including the 1 micron bags on the 2 hp PSI. I haven't looked much, but am not aware of any bag that matches a pleated cartridge filter. IIRC, the Wynn has over 200 sq ft of filtering area, which improved performance from reduced back pressure.

3) Unless you don't have the space, I'd think hanging onto the Reliant would be preferable to the 2 hp PSI. If you can't afford two Wynn filters to set it up, maybe you could get one for now, and do one of two things:

a. If the Wynn would fit your DC1B, use it there for the time being; it would improve performance, and maybe even offset putting in a Thien. When your budget allows, get a second Wynn, and put both on the Reliant.

b. Get one Wynn filter and install it onto one of the housings. Put a baffle into the "y", to block air flow to the other housing. Even with using only half of the collector, with the surface area of the Wynn (compared with bags), you're still probably ahead with air flow. (Just a thought, and based on no personal experience).

Hopefully, those with more knowledge will either verify or contradict the above. just my .02  :)

I doubt that replacing a bag with a second filter would have a significant effect of performance. First, the Wynn filters have pretty large filter surface areas, so I don't think that resistance to airflow by one filter is a major factor in limiting performance, so a second filter would decrease effective resistance only slightly (diminishing returns). Duct run length, size, bends, etc, probably are greater factors.

The only scenario I could see a second filter making a big difference would be if the first filter was so caked up with fine particles that it really restricted flow. It that's the case, you have much bigger problems- bad top hat design/build. Also, if that's the case, the second filter, replacing the bag, would clog up even worse than the upper filter.

I assume, though, that your top hat must be working so well that a plastic bag is almost superfluous- otherwise, you wouldn't be thinking of removing it!

I think the money you'd spend on a second filter would be better used in saving toward a bigger motor/impeller system.

Just my .02, and hope that somebody with more expertise can correct, as needed.

What a relief that such a good looking build is working well, but what kind of wood are you working with that leaves light blue sawdust?  ???   

Mike-

Thanks for the clarifications- I'm always learning more from this site.

BTW, I don't know what type of chalk Phil has used, but I'm planning on using snap line chalk- it seems to be ground pretty fine, and dirt cheap.

Regards, John

What a shame that such a beautiful build isn't working. :(

A couple of thoughts that come to mind, having pored over recent threads on this site:

Phil commented in one thread that for heights over 1H, the orientation of the drop slot might need to be changed vis-a-vis the intake port- he suggested the chalk method to trace the path.

Also, can you adjust the height of the bell mouth over the baffle- maybe lowering it a bit would reduce pass through.

Just thoughts from a newbie (in the process of building my own).

Interesting technique! My experience with fiberglass is pretty limited (patching holes), but it seems like you just laid the cloth and resin right over the styrofoam- that you don't need any mold release agent (e.g. wax). Is that so?

BTW, it never occurred to me to use eggs as weights for glue-ups!  :)

QuoteIntroducing a handful or two of bright chalk dust into a new separator should give one a good idea of where to position the baffle.

Maybe you could elaborate on the positioning? Assuming that the chalk makes a spiral pattern along the wall, is it correct to say that the solid portion of the baffle (i.e., no slot) should begin where the chalk spiral intersects with the baffle? That is, should there be a 120 degree landing zone ("runway") on the baffle for the air path to take before it hits the slot.

In my mind, the alternative is that the spiraling air path would shoot straight through the baffle slot into the waste container, which would create a lot of turbulence, and stir up  collected dust. Of course, my mental image of what's going on has often been in error ???, so I appreciate any clarification.