Re: ShopNotes Full Year 2012 Issues Collection
Macabeo Eastman
The standard procedure for commercial airflow testing will not give consistent results for small shop testing, so small shop dust collection airflow testing is done without filters.
For larger systems filters are cleaned with an automated system that gives very consistent results. The cleaner and newer the filter the higher the airflow because as a filter gets used some dust ends up getting trapped in the filter strands that does not come out with automated cleaning. This process of fine dust getting trapped is called seasoning. Seasoning traps fine dust in the filter pores which adds to filtering efficiency and to overall filter resistance. A "fully seasoned" can provide up to twenty times better filtering than a clean new filter and it can so smother airflow that it can reduce a clean new airflow from 1000 CFM to barely 300 CFM when "fully seasoned". Filters that vent outside are tested when they are fully seasoned, meaning they have built up as much internally trapped dust as the filter can carry through its normal automated cleaning cycle. This fully seasoned testing determines how big our filters need to be made, at what pressure we need to clean our filters, and when to replace our filters. Most high-end automated systems track the resistance of our filters and automatically clean our filters when the resistance increases by more than 1 to 2 water column inches of resistance. Most other automated systems require us to manually do this process recording the pressure after every cleaning and adjust a sensor so it triggers a cleaning cycle when the pressure after cleaning rises more than 1 to 2 water column inches. As these fine particles work their way through our filters they cut and tear the fine fibers that make up our filter's matrix opening up our filters. The more open or "worn out" a filter gets, the lower its resistance. The high-end automated systems track the "fully seasoned" resistance levels and set an alarm when the filter resistance has fallen equal to or some amount below its fully seasoned level.
For small shops where filters are manually cleaned, it is near impossible to consistently clean our filters so our testing gives very different results depending on how well our filters get cleaned. Depending on how thoroughly we clean our filters, it takes about three cleaning cycles for a filter to get to half its "fully seasoned" performance. Although many small shops need about nine cleaning cycles to reach "fully seasoned", it can take as little over a year for a small shop filter to fully season. During the seasoning process the fine unhealthiest dust goes right through. Worse, even a fully seasoned filter slowly lets the fine dust work its way through the filter matrix at about the same rate as new dust builds. This passing of the fine unhealthiest dust is why ASHRAE who sets the standards for testing filters requires that filters which vent inside only be tested when clean and new.
A majority of small shop vendors advertise fully seasoned filter ratings and compute filter sizing based on clean new filters which is backward. They do this because this lets them advertise better performance and use far smaller filters which are much less expensive. This means their small shop filters that vent indoors do not protect our health and are so small in size they rapidly load up needing frequent cleaning and constant replacement. Most small shop dust collectors and cyclones end up needing their filters replaced fairly quickly because they were either far too open to provide good fine dust protection or so small in area they soon self-destruct.
A far better testing approach is to use the same seasoned filters on each unit to be tested, but this is just not feasible for most small shop woodworkers. Seasoning testing requires use of calibrated dust because even different types of wood dust on your filters will change the airflow. As a result, most end up with replacement filters that are fairly consistent larger commercial cartridge filters. It is best to test with the same seasoned commercial filter sized correctly for the range of airflow being tested. Such a setup ends up blowing away all the testing games and shows a far more realistic airflow across a whole range of machines and is clearly the right approach. Unfortunately, the filter needs appropriately "seasoned" to set a fixed filter resistance level, but doing so is beyond most hobbyists and small shop woodworkers. The only woodworking magazine test that went to the trouble to do this kind of testing to date is the Fine Woodworking April 2006 portable dust collector tests. Preparation for that article also tested every major brand of small shop cyclone, but when using seasoned filters the airflows were so bad that the vendor community lobbied Fine Woodworking Magazine to not even publish the cyclone testing results.
Meanwhile, for our own testing we have two choices to get meaningful comparisons. We can either go to the work to use the same seasoned test filters on each unit, or just remove the filters. Most choose to just remove the filters. If we test without filters, we then need to adjust the performance curve by adding filter resistance. The amount of filter resistance to add is a huge range. Large cartridge filters sized in accordance with filter material maker recommendations can add as little as 0.25" of resistance when fully seasoned to some of the smaller dust collector bag filters adding as much as 5". Regardless, we mostly end up needing to do our testing by disconnecting our blowers from the ducting and removing any filters because of their unpredictable impact on airflow. Testing with the filters off eliminates that big filter resistance variable during testing. Testing with the filters and ducting off can easily burn up blower motors because they can move more air and draw more amperage than the motors were made to handle. Filter resistance can also be checked later by testing with the filters.
The more air moved, the more work our blower motor must do which increases amperage draw and horse power used. Dust collection blower motors are built to handle the up to six times higher start up amperage for the few seconds while they bring a heavy impeller up to speed. So, most blower motors can run for a few minutes at up to six times their rated amperage, but running a motor over its rated maximum amperage power builds up heat that will burn up our motors. This is why the industry standard also uses an amp meter and stops the testing as soon as our blower motor draws its maximum rated amperage. Going over this rated amperage causes heat to build up and burn up our motors.
Our test pipe diameter also significantly affects test results. Air at dust collection pressures is more like water so will barely compress at all. Just like a water valve reduces the opening at a single point to control flow, the smallest opening or duct size limits air flow. A test pipe that is too small chokes off larger blowers leaving them "air starved" and unable to move maximum air. At typical dust collection blower pressures, we find that:
1" duct only moves 0022 CFM at 4000 FPM
2" duct only moves 0087 CFM at 4000 FPM
3" duct only moves 0196 CFM at 4000 FPM
4" duct only moves 0349 CFM at 4000 FPM
5" duct only moves 0545 CFM at 4000 FPM
6" duct only moves 0785 CFM at 4000 FPM
7" duct only moves 1069 CFM at 4000 FPM, and
8" duct only moves 1396 CFM at 4000 FPM
The industry standard for testing is to use a test pipe diameter that mates with the blower inlet. This presumes that the blower maker sizes their blower inlet to assure the motor is neither "air starved" nor trying to move more air than the blower can handle. Moving too much air pulls more amperage than our motors can safely support resulting in overheating that soon burns up our motors. Unfortunately, with no standards or oversight, small shop vendors choose not to adhere to industry standards and many size their blower inlets to provide the maximum possible airflow by forcing use of oversized test pipes. Also, many use impellers that are too large for their motors. Many also increase airflow by testing their blower performance with oversized ducting, no attached cyclones, filters, ducting, or hoods. Oversized inlets, test pipes and impellers move more air as does running without the normal resistance of our ducting, filters, cyclones, tool hoods, etc. Most small shop vendors and woodworking magazines also test with no amp meters so they inappropriately keep testing when their motors are already drawing so many amps that they are overheating. This is why many blower motors burn up during vendor and magazine tests.
Lots of experimenting shows we need at least ten diameters of ducting for our test pipe to settle turbulent air. We also need a few diameters between where we put our test probes and the blowers or air rushing into the blowers creates turbulence that also throws off our test results.
Sucked air comes from all directions at once, so when we do dust collection testing the air coming from the back side of our openings will rush forward and crash into the air coming in from the sides and front of our test pipe opening. To prevent this the industry standard is to use a circular disk sealed onto the end of our test duct. This keeps the air from behind slamming into the air coming in from the sides and front. Testing without this disc gives much lower results than the system provides in real use. If the disc is changed for a hyperbolic inlet shaped like the opening on a tuba, then we get even better airflow. Instead of using the disc that the industry experts recommend, many vendors use these tuba shaped inlets creating even larger than normal use airflow values. Key to proper and consistent testing is also to use a properly setup test pipe to test the whole range of dust collectors or cyclone-based systems with similar airflows. Picking the right size test pipe can make a huge difference in results and if too large can even burn up our motors and produce gravely inflated test results. A far more realistic test is to either use your maximum ducting size or a test pipe sized to handle the target airflow. To the right is pictured a cream colored test pipe which is propertly set up. It has the disc sealed on the end of the pipe with the attached needle valve with crank to adjust airflow from zero to maximum. It also has a properly configured pitot tube with dual air lines, plus a pressure sensor. This test tube is also properly connected to the inlet of the cyclone.
In commercial systems with multiple ducts working all at once, our mains and blowers grow large enough to handle all the airflow combined. In small shop systems we use tiny blowers that barely have the capacity to collect from a single large machine at a time, so we shut off using blast gates all but a single run. As a result our main ends up needing to be only as large as the airflow needs of our largest run, a whole different ducting design. Using mains that are too large or using down drops that are much smaller than the mains can build dangerous dust piles in our mains that pose a serious fire risk as any spark can quickly be blown into a serious fire. Air engineers established through years of testing and experience that we need 350 CFM air volume moving at 4000 FPM air velocity for good chip collection. Since most small shop vendors build dust collectors and cyclones move double this much airflow, we use a 6" diameter test pipe. Likewise, air engineers established through years of testing and experience that we need 1000 CFM with a ducting airspeed at about 4000 FPM for fine dust collection and transport from our larger small shop tools and dustier operations. Since Area=CFM/FPM we can use these two numbers and a little math to calculate we need almost exactly a 7" diameter duct for moving this airflow and airspeed. Since most fine dust collectors and cyclones are built to move some extra air, to give a fair test we need to test with an 8" diameter test pipe. This also is the size test pipe we should use for our ducting in our shops if we target for this same 1000 CFM.
Many 1.5 hp and larger small shop dust collectors and cyclones have a 6" inlet which may split into two or three 4" separate runs. These systems do not move enough air to be tested with more than 6" diameter test pipe. For those systems that claim higher airflows, bumping up to a 7" test pipe is optimum for moving 1069 CFM at 4000 FPM but also requires having a motor able to handle up to 5 hp. Bumping up to a 8" diameter test pipe is optimum for moving up to 1396 CFM at 4000 FPM but also requires having a motor able to handle up to 6.45 hp. We end up seeing a 2 hp powered cyclone with 14" impeller testing with over 1500 CFM, but also pulling more than 4.5 hp.
To test for good fine dust collection, we should target for 800 CFM which requires use of a 6" test pipe. We also need to put on the end a ring sized at least three pipe diameters large. This places an 18" diameter plywood donut shaped plate on the air inlet end. The plate reduces the "vena contracta" effect that simulates a restrictive inlet. I chose to use 6" S&D PVC for my test pipe because it has lower friction, constant size, and is easily sealed to the blower inlet. The testing standard calls for locating our 3/16" pitot tube hole at least 1.5 pipe diameters from the inlet (9" minimum) and still have at least 8.5 pipe diameters (51" minimum) to let the entering air stabilize before hitting the pitot. I made my 3/16" pitot tube hole at 12" from the inlet end. Rather than have to pull the tubing off the top of the pitot to set each SP with my needle valve I use a second gauge attached to a simple probe to measure SP (vacuum) at the same time as I measure velocity pressure with my first gauge. The testing standard also calls for the SP probe hole to be located at least 9" from the inlet and have at least 18" of test pipe before it to ensure clean air flow. I made the second hole for my probe perpendicular to the pitot hole at 12" from the inlet. This spacing and order preserve the required clean air to the pitot and probe.
One other fairly serious concern with test pipes is the ASHRAE and Dwyer Instrument test protocols clearly state that the air needs tested at different levels inside the pipe then averaged. Testing just the center gives a maximum airflow that drops significantly as we test closer to the pipe walls. Personally, I choose to just test the center with each test done using exactly the same test pipe to provide a good consistent result between units. I also know that if I test at 0", 1", 2" and the center at 3" then average will give me much less airflow.
Another advantage of the way I set up my test pipe is it saves me from having to make and mount a separate pipe for the Max SP and the Minimum Amperage tests. Since no air is flowing in either of these tests, I could make do with a short stub of ducting 12" or longer with a well-sealed connection from my meter to that duct. Alternatively, I can just use the simple probe in my long test pipe and seal the pitot. To seal the pitot connect a single tube to both the top and side outlets.