Measuring Dust Collector Performance - Fan Type Anemometer

[Larry Frank]

I have posted several threads that have detailed dust collector testing and along with effects of ducts and hoses. I borrowed the test equipment to do this as it is not cheap especially the digital thermal anemometer.

Now that I have the performance curve for my system, I can use a simple homemade manometer to determine the static pressure and read from the graph for my system the approximate flow.

As I read the comments and questions, it would be really nice if there were an easier way for others to evaluate their system. Many people are not going to have access to the instruments that I used or want to go through all of the work I did.

Several people have posted about using a handheld fan type anemometer. I have done that using a “HoldPeak HP-866B” anemometer but have not been happy with the variability in the results. It seems that holding the fan anemometer in the right location at the end of a hose or duct is difficult as the dust collector is trying to suck it in and any movement of the anemometer results in larges changes in air velocity.

I thought that there had to be a better way than hand holding the fan anemometer. I used a piece of ¾” plywood and made a cut out for the anemometer and secured it in place using zip ties.

20160728_132330 Anemometer Holder EM.jpg

The 4 inch dust collector hose was secured to a piece of plywood to keep it from moving and the anemometer holder was held in place with a clamp. The distance was varied by using blocks of wood as spacers.

20160728 Test Side EM.jpg20160728 Test Front EM.jpg

This system made of scraps of wood held things in place securely and it was possible to center the anemometer on the hose.

I measured the flow with the anemometer at several different distances from the end of the hose and then plotted the data with a curve fitted to the data in Excel.

Graph Fan Data.jpg

The results show that the air speed measurement is very sensitive to the distance between the end of the hose and the fan anemometer. The speed goes up rapidly as the anemometer gets closer to the hose end. So what is the right distance to measure the speed? My tests with the same hose and conditions using the digital thermal anemometer gave about 490 cfm and 5621 fpm. Based upon the graph, it appears that holding the fan anemometer about a half an inch from the end of the 4” hose would be about right.

However, when working with your own system, it probably does not make a difference if you are 3/8” to 1/2” as long as you do it the same way every time. It would work well in trying to evaluate any modifications that are made.


Once you have the flow, you can also measure the static pressure. I used a homemade water u-tube for some of my previous work. It was simple to just measure the difference in height on the two sides to determine the static pressure and gave the same results as the digital manometer.

20160728 Water Manometer.jpg

Now, with a little work and not too much expense (the anemometer I used is $23 on Amazon), you can evaluate your dust collector and any changes or modifications.

There are quite a few people on the forum with a wide range of dust collectors that are making modifications to them to enhance the performance. It would be extremely helpful if a few would take the time to evaluate the baseline for their dust collectors and then report on the results of modifications. I am certain that it would be a huge help to others seeking to do similar improvements.

[Dan Friedrichs]

I had a fan type anemometer for about 30 seconds. I tried to use it to measure airspeed on a shop vac. I now have a fan-less, markedly less accurate anemometer

[Nick Hicks]

"I had a fan type anemometer for about 30 seconds."

This was my experience as well. I guess the cheapies are not well constructed - you get what you pay for

[Larry Frank]

One thing I noticed is that if you get real close to the end of a hose or fitting, the fan goes crazy fast.

I guess what I was hoping with this thread is to show with a low cost set up, one could effectively measure dust collector performance.

There are many folks using lower power dust collectors like the HF. They are making modifications to try to improve performance. What I have tried to present is a way to evaluate modifications. Also, the actual performance could be shared with others trying to do similar things.

[Nick Hicks]

That's true but if I remember properly, the target speed for duct velocity (from Bill Pentz) was 20 metres per second (4000 feet per minute) so that would be 72 kph (or 45 mph) which isn't crazy fast so I thought the anemometer should have been able to handle it - but I was wrong :-(

I think the problem with trying to measure performance by methods other than measuring the velocity in the duct is that they can be misleading/difficult. Using a manometer to measure the pressure drop at different points along your ducting as you did shows you where the pressure losses are occurring but that is measuring static pressure. I have a technical bent as well but I couldn't think of a method to reliably measure the airspeed. Airflow at an open pipe end is extremely variable and you get different readings even at the side and center of the opening due to turbulence and skin friction on the inside of the pipe.

Ideally, you want a large opening to get maximum airflow in the duct to keep dust and shavings in suspension in the airflow while trying to minimise pressure losses. If you have a clamp style meter, you could measure the current flow to the motor as that is proportional to the work the motor is doing - more airflow = higher ampage. The motor should have it's lowest amperage draw when all ducts are closed. In the end, I just took a theoretical approach to the design using Bill's advice - minimising the use of flexible hose, wider bends, avoiding 90 degree bends if possible, having a 5 foot straight pipe leading into the dust collector and so on from Bill's website.

You probably already know all that though so I'm sorry I couldn't help.

[Larry Frank]

You are correct about the recommended speed. Some of the lower power collectors such as the Harbor Freight struggle to get close to this.

A note of importance. You do not need to measure static pressure along the pipe. For my shop tests, I measured at one point just where the pipeline into my shop. Whatever you do downstream from that point will show up.

I found that you can effectively measure using a handheld fan anemometer. But, you need some way to hold it in the proper position. Not easy, but can be done.

Once one gets an idea of the performance curve of their dust collector, all you need to do is measure changes in static pressure. A homemade u-tube manometer works great. I built mine with 1/4" clear plastic tube, blue dyed water, piece of plywood and zip ties. It costs very little and does not take long to make. You can attach this near your dust collector and not have to move it.

My goal with thread is to encourage people to measure performance and share the information so that others can benefit. It wood be informative to know the change in static pressure with a dust deputy or Thien top hat.

Can anyone post some information from their system.

[David Kumm]

You need an anemometer rated at 8000-10000 fpm if testing at the port. I hold the fan tight against the fitting and then move it around the opening and take the lower end of the range. I suspect the reading may still be a little high but all I'm looking for are relative numbers in comparison the the motor amp draw. Since I run my collector off a vfd I can also approximate the cfm change when adjusting speed. What the measurement really tells you is the effect of restriction on your specific impeller. What my method doesn't tell me is how much cfm I lose when the machine is attached. I do know that with my system, long runs, lots of gates, a standard backward curved low pressure impeller did not produce the cfm I desired and increasing rpms did not increase it much as the fan is designed to top off so as to not overamp the motor. Once you correlate amp draw with cfm, you can get a pretty close estimation just by watching the amps. Dave

[Ole Anderson]

Just testing a 3.5" opening on a blast gate (fed from a 5" drop) with a 2 HP SDG I was getting 11,250 fps or 128 mph (752 cfm). I expect the end of a 2.25" shop vac would be even faster. Used a fan type Velocitor anemometer.

[Larry Frank]

I am so glad to see people posting their results. I agree that you can not use the fan anemometer when connected to a tool. BUT.....if you hook up a manometer you can see the effect.

[Nick Hicks]

That's good although the air velocity will drop when the air enters the 5" section of the pipe but the airspeed should still be fast enough (5508 fpm). There is a good web page to assist with duct velocity calculations at the following webpage if interested:

http://www.engineeringtoolbox.com/du...ons-d_883.html

The shop vac may have a faster airspeed but the 2.25" cross-sectional area limits the amount of air that can flow through so the cfm would be reduced.

[David Kumm]

Ole, are you sure your meter is testing correctly? I don't mean to question but I've tested all kinds of impellers and my current 7.5 hp system with a high pressure fan won't do as well as your number. My 5 hp Oneida impeller was was closer to 7-8000 fpm. Dave

[Larry Frank]

Please note the graph in my own original post. When you get too close to the end of the hose, the numbers go up dramatically. When you get close you greatly reduce the area, as the anemometer covers the end of the duct/hose. I found with a 4" hose end I needed to be about 1/2" from the hose end.

[Ole Anderson]

Ole, are you sure your meter is testing correctly? I don't mean to question but I've tested all kinds of impellers and my current 7.5 hp system with a high pressure fan won't do as well as your number. My 5 hp Oneida impeller was was closer to 7-8000 fpm. Dave

Keep in mind it was measuring the 3.5" blast gate as an orifice fed by a five foot 5" line tied into the 7" main a few feet from the DC. I don't have any reason to believe the unit measures incorrectly, as my other measurements seem reasonable.

[Larry Frank]

The easiest way to double check measurements is with a manometer. You can make one cheaply with 1/4" clear plastic tube and colored water. Stick one end into the duct and measure the difference in height. Then just check on the performance curve from Oneida and read the cfm. The curves that Oneida provides are really close to reality based upon what several members on the creek have measured.

I measured around 700-750 CFM at about 9.5" static pressure on a 4" opening. I was using a 5 hp unit with a 15" impeller. My measurements were done in a standard method using a digital thermal anemometer. I have done a lot of comparisons with a fan type anemometer and unless you find a way to hold it stable a short distance off the end of the duct, the values are all over the place. If you hold it too close the number go up and are not correct.

[Larry Frank]

Some the discussion in this thread were about readings that seemed too high when using the Fan Type Anemometer and this bothered me. I wondered what could be going on so I took a close look at my fan anemometer and noticed something. The outer ring of the anemometer is fairly large being. I took some measurements of it and found the following.

Fan Dimensions.jpg

The outside diameter is 3.378” and the inside dimension is 2.578”. This means that the outside diameter covers 8.962 sq in. But the important issue is that the ring between the 3.378” diameter and the 2.578” diameter covers an area of 3.742 sq in.

If one is testing a 3.5” duct, this means that the ring covers about 39% of the area and this will concentrate the flow in a smaller area and result in higher than actual velocity readings. This effect is especially critical when the fan anemometer is held close to the end of the hose.

On a 4” hose, it would cover about 30% of the opening and again could result in too high readings.