Dust extraction - sizing

Hi everyone.
I have an existing dust extraction system installed in my shed that isn’t performing as well as I’d like so I’m about to upgrade it. I’ll do all the usual things like replacing elbows with sweeps and other things to minimise suction losses.

I’m also considering upgrading the ducting from the existing 100mm to 150mm but I first want to make sure this isn’t a false step. My extractor is a 1.5hp 2 stage cyclone unit with a 150mm inlet and produces a nominal 1100 cfm of airflow (please excuse the imperial units).

The maximum length of the main ducting is about 6 metres with 3 inlet points, each supporting a few machines. The biggest loads would from the table saw at 400cfm and the thicknesser at 600cfm (estimated). It’s a small hobby workshop so I would never run more than one machine at a time.

My question stated simply is, will an upgrade of the main duct from 100mm to 150mm increase the suction or decrease it?

Thanks in advance for any input.

Cheers

Andrew

Yes it will help a lot if you increase the inlet to 150mm. I would recommend that you increase the discharge ducting by 1 and a third. I assume that you are using a BCC blower?

Many people go for oversize ducting. It needs to be large enough to handle the flow without too much pressure drop but not any larger. The larger you go the lower will be the duct velocity. It might seem counter intuitive but blockages are more likely with overlarge ducts because of the low velocity which allows the dust particle to drop out of the flow.
To answer your direct question, firstly with a question. Just what do you mean by "suction"? The suction at the work place depends on the suction at blower minus the pressure loss in the ducting. So as long as the duct is large ENOUGH then the pressure loss will be low in comparison to the suck at the blower. However, air velocity is what you need to consider as far as the ability to pull in the dust is concerned.

YES indeed, velocity is the most critical part of a dust collecting system. Velocity is a function of flow volume and duct size. I taught in a school shop that had a very large, very expensive, poorly designed, system. It had mostly 3 or 4" drops, with a 6" on the 24" planer. Each had a gate to block off flow to unused machines. As I recall there was about 8 drops. Those entered into a large duct over head. It started at about 24" at the collector and dropped in size after passing a few drops. Ending about about 10".

The 24" main duct was way too large, even if all of the drops were open. They could not provide enough air flow to keep velocity high enough in the main duct it was always plugging up. The was especially true with the dust collecting filters just a little bit dirty.

The link here is in regard to an HVAC blower system. But the principles are the same. Lots of factors come into play in your design. Bigger is not always better in this case.

There are tables that show required velocities to capture various particles.

Good Luck.

Read through the post by hemmjo,above !

You could also find several papers on the internet about Fireman's hydraulics. That is an excellent source of technical information as they have to know about water flow, hose size, pressure and number of connections. It does not matter that they focus on water. Water is a fluid as is air, air is just a different density.

There are more PDF's here than you can shake a stick at.

https://www.google.com/search?q=fire...client=gws-wiz

Thanks to all for your advice. I’ll get started on the project with a greater degree of confidence.

Tonyfoale and hemmjo are absolutely correct. Prioritize duct velocity over pressure drop.

I too shall unapologetically mix units. 100 mm duct has a cross sectional area of 0.00785 m^2. 150 mm duct has a cross sectional area of 0.0177 m^2. At 1100 CFM (31.1 m^3/min or 0.518 m^3/sec), that gives us average duct velocities of 65.9 m/s for the 100mm duct and 29.3 m/s for the 150mm duct. Using hemmjo's very excellent table (https://www.engineeringtoolbox.com/p...rts-d_134.html) it looks like you're teetering on the edge of running into problems with that 150mm duct because your air velocity is too low.

Stick with the 100mm duct.

Nova robotics - thanks for the excellent calculations. I’d done some of the same myself but I have more confidence in your numbers.

Based on your calculations with 150mm ducting and a duct velocity of 29.3 m/s, using Hemjo’s table it looks to me like that velocity will accommodate all or most activities in my wood workshop - in many cases with reasonable head room. Or am I missing something in that simple interpretation eg losses due to various factors?

Regards

Andrew

Quote:

---

Originally Posted by andrew mcgregor

Nova robotics - thanks for the excellent calculations. I’d done some of the same myself but I have more confidence in your numbers.

Based on your calculations with 150mm ducting and a duct velocity of 29.3 m/s, using Hemjo’s table it looks to me like that velocity will accommodate all or most activities in my wood workshop - in many cases with reasonable head room. Or am I missing something in that simple interpretation eg losses due to various factors?

Regards

Andrew

---

That's average velocity within the duct. Boundary layer effects mean the air velocity near the walls of the duct are much slower, and around elbows and other features you'll find all kinds of dead air. If you're getting near those velocities you're going to accumulate little deposits of crud all throughout your ductwork. Imagine how bad the hose on your vacuum cleaner would get if you turned the speed of the motor all the way down to the point where it would just barely lift dust. That's the situation you're trying to avoid.

Quote:

---

Originally Posted by nova_robotics

That's average velocity within the duct. Boundary layer effects mean the air velocity near the walls of the duct are much slower, and around elbows and other features you'll find all kinds of dead air. If you're getting near those velocities you're going to accumulate little deposits of crud all throughout your ductwork. Imagine how bad the hose on your vacuum cleaner would get if you turned the speed of the motor all the way down to the point where it would just barely lift dust. That's the situation you're trying to avoid.

---

Take notice and use 100 mm duct.

But I might be wrong. My gut feel says strongly that you want high air velocity. But gut feel can lead us astray, so I just worked it out from a head loss perspective. Going from 4" round duct to 6" round duct will save you about half a PSI. The 4" will have about 0.7 PSI of loss, and the 6" will have about 0.1 PSI. For a shop vacuum that may only make -3 PSI (guessing here), that may be significant. You might very well get 10-20% more suction at your endpoint by increasing the duct diameter.

It's a tradeoff. All I know is I don't want to mess around with removing sections of duct to try to unclog them, so I'm installing 4" in my shop.

Quote:

---

Originally Posted by nova_robotics

For a shop vacuum that may only make -3 PSI (guessing here), ................................

It's a tradeoff. All I know is I don't want to mess around with removing sections of duct to try to unclog them, so I'm installing 4" in my shop.

---

I have measured various shopvacs and none get to 3 psi with useful flow. Here are the results of tests of different shopvacs, without further explanation. The axes show pressure in inches WG, for reference 28" is 1psi.

Attachment 41505 Attachment 41506

Quote:

---

Originally Posted by tonyfoale

I have measured various shopvacs and none get to 3 psi with useful flow. Here are the results of tests of different shopvacs, without further explanation. The axes show pressure in inches WG, for reference 28" is 1psi.

Attachment 41505 Attachment 41506

---

I am amazed you have this information.

Quote:

---

Originally Posted by nova_robotics

I am amazed you have this information.

---

I needed the info when designing my flow bench.

Attachment 41507

There are a few other things to keep in mind when assembling your system.

If your main duct is on the floor, just skip to #3.

1) If your main duct runs over head, the tap from each machine needs to enter the main duct on the side. It they enter from the bottom, eddy currents created by air and shavings passing over the unused openings, will soon fill the ducts entering from the bottom.

2) If your power unit is below the overhead duct, the duct and elbow rising up to meet main can be the larger size to help minimize pressure loss. The shavings will fall into the collector anyway.

3) If your duct is plastic, you are also building a static electric generator. There are a a few ways to eliminate that. Like this, OR this, or any similar method.

Quote:

---

Originally Posted by tonyfoale

I needed the info when designing my flow bench.

Attachment 41507

---

That's interesting. I've never had the opportunity to mess with one, but I've wondered about mounting a loudspeaker to do Helmholtz resonance tuning as well as flow. Is that a thing that people do?

If you are collecting or conveying any combustible material and you don't want to get shocked, you may want to run a solid copper ground wire inside the entire duct run. Ground the end to a good earth or system ground to eliminate static build up. I can attest to this because I have been knocked on my ass a few times!

Quote:

---

Originally Posted by nova_robotics

That's interesting. I've never had the opportunity to mess with one, but I've wondered about mounting a loudspeaker to do Helmholtz resonance tuning as well as flow. Is that a thing that people do?

---

You are trying to modulate the flow and see where the intake or exhaust manifold is at resonance ?

With a sufficiently large diaphragm woofer you can move a fair bit of air. The problem is that is is not unidirectional. It is push pull. Tony I think is looking for unidirectional flow.

I do a fair bit of vacuum bag veneering and clamping. I can tell you that a decent small shop vac can get you quite a serious vacuum for those purposes. that 3 PSI is over every square inch of surface. pretty useful for clamping. I have helped one machine shop come up with a vacuum plate arrangement that allowed them to machine things with no visible clamping method. A deeply drawn vacuum can give you near 15 PSI.

When I started doing this in the early 90's I broke many pieces of wood and molds learning their structural limits. Also discovered a neat thing call a variable vacuum switch. The reverse of a compressor switch.

Quote:

---

Originally Posted by mwmkravchenko

You are trying to modulate the flow and see where the intake or exhaust manifold is at resonance ?

---

Pressure waves can be very important to the performance of highly tuned engines. I pointed out that the suggestion to evaluate this on a flow bench would have problems due to the need to emulate very high exhaust temperatures, because the wave velocity various with temperature.

Quote:

---

Originally Posted by mwmkravchenko

.... Tony I think is looking for unidirectional flow.

---

Actually I test flow in both directions. The wave action at certain parts of the cycle at different RPM causes reverse flow. It is good if you can reduce the reverse flow without reducing the forward flow. Hence the two way testing.

Quote:

---

Originally Posted by tonyfoale

Pressure waves can be very important to the performance of highly tuned engines. I pointed out that the suggestion to evaluate this on a flow bench would have problems due to the need to emulate very high exhaust temperatures, because the wave velocity various with temperature.



Actually I test flow in both directions. The wave action at certain parts of the cycle at different RPM causes reverse flow. It is good if you can reduce the reverse flow without reducing the forward flow. Hence the two way testing.

---

OK Makes sense.

Speed of sound varies with the temperature of the medium it is measured in. As does the amplitude of the sounds power. Easier to have a nigher amplitude for a given power input in a warmer medium, air mixture in this case.

Warmer air easier to transmit sound through vibrations in such a system. It is still the compression and rarefaction of air.

Quote:

---

Originally Posted by tonyfoale

Pressure waves can be very important to the performance of highly tuned engines. I pointed out that the suggestion to evaluate this on a flow bench would have problems due to the need to emulate very high exhaust temperatures, because the wave velocity various with temperature.

---

I was mostly thinking for intake tuning. That's usually where people go after resonance. Variable intake runners and whatnot. Then you don't have to mess with any of that pesky hot exhaust.

But for the exhaust, if you made a closed(ish) loop system and ran 1/3 air, 2/3 helium you could pretty closely approximate hot exhaust at room temperature. That would get you your resonance tuning. A bottle of helium is cheap. I don't know how many tests you'd get out of it though.

Quote:

---

Originally Posted by mwmkravchenko

You are trying to modulate the flow and see where the intake or exhaust manifold is at resonance ?

---

Yup. Intake and airbox tuning is super important on engines. You can even achieve a light supercharging effect (I think about 5 or 6 PSI) by getting the high pressure reflections to hit the intake valves just right. A lot of effort has gone into tuning the airbox and intake of the car or truck you drive right now. It's pretty neat stuff. The problem is you only get resonances at one or two RPMs, and the rest of the RPM band doesn't benefit. I had an old '85 Porsche 944 with a weird flapper thingy that would open and close off chambers on the intake specifically for changing the resonant frequency of the intake as the RPMs changed. So by changing the internal volume of the intake they could selectively create a whole bunch of different resonant frequencies. My Porsche 928 has a V8 where the intake runners are static, but grouped in pairs. To smooth out the powerband each pair is a different length and tuned for a different RPM. So (for example, not sure if the cylinder numbers are correct but you get the idea) as you increase in RPMs, cylinders 1 and 5 do most of the work, then cylinders 2 and 6, then cylinders 3 and 7, then cylinder 4 and 8 as you wind it out. You have no idea how much I hate this, but I understand why they did it. Just to smooth out the powerband so you don't get one big punch at a certain RPM, and flat spots everywhere else. It's all very interesting, but I'm only an armchair expert. Never actually done it myself. A big part of the Wikipedia page on intakes is devoted to variable length intake manifolds and Helmholtz resonance.

https://en.wikipedia.org/wiki/Inlet_manifold

Quote:

---

Originally Posted by nova_robotics

Yup. Intake and airbox tuning is super important on engines. You can even achieve a light supercharging effect (I think about 5 or 6 PSI) by getting the high pressure reflections to hit the intake valves just right. A lot of effort has gone into tuning the airbox and intake of the car or truck you drive right now. It's pretty neat stuff. The problem is you only get resonances at one or two RPMs, and the rest of the RPM band doesn't benefit. I had an old '85 Porsche 944 with a weird flapper thingy that would open and close off chambers on the intake specifically for changing the resonant frequency of the intake as the RPMs changed. So by changing the internal volume of the intake they could selectively create a whole bunch of different resonant frequencies. My Porsche 928 has a V8 where the intake runners are static, but grouped in pairs. To smooth out the powerband each pair is a different length and tuned for a different RPM. So (for example, not sure if the cylinder numbers are correct but you get the idea) as you increase in RPMs, cylinders 1 and 5 do most of the work, then cylinders 2 and 6, then cylinders 3 and 7, then cylinder 4 and 8 as you wind it out. You have no idea how much I hate this, but I understand why they did it. Just to smooth out the powerband so you don't get one big punch at a certain RPM, and flat spots everywhere else. It's all very interesting, but I'm only an armchair expert. Never actually done it myself. A big part of the Wikipedia page on intakes is devoted to variable length intake manifolds and Helmholtz resonance.

https://en.wikipedia.org/wiki/Inlet_manifold

---

I have some new things to learn then. Funny thing I started cluing in on most of what you were saying here from your last post.

I have done similar work with distributed resonances in subwoofer enclosures. Same idea. The world is not as big and mysterious as many would have us believe. You just need to be able to connect the dots. Physics is physics no matter what the application.

Quote:

---

Originally Posted by mwmkravchenko

I have some new things to learn then. Funny thing I started cluing in on most of what you were saying here from your last post.

I have done similar work with distributed resonances in subwoofer enclosures. Same idea. The world is not as big and mysterious as many would have us believe. You just need to be able to connect the dots. Physics is physics no matter what the application.

---

So you probably know more than me then. I've never done any of it, and I only know from what I've read. It's all the same stuff, just different application.

It's funny because as a teenager I used to rip all that stuff off my engines, thinking I was reducing restrictions. Oh dear...

Quote:

---

Originally Posted by nova_robotics

So you probably know more than me then. I've never done any of it, and I only know from what I've read. It's all the same stuff, just different application.

It's funny because as a teenager I used to rip all that stuff off my engines, thinking I was reducing restrictions. Oh dear...

---

No I don't know more.

I don't know more than Tony either. I just have done similar work. Which means I can understand the ideas and the goals. Just not exactly what you have to do to get it accomplished. I hang around on this forum because there are quite a few intelligent people. And I constantly learn things. Some new, some old. But I learn new and intriguing applications. It get me thinking and I appreciate this a lot.

Quote:

---

Originally Posted by mwmkravchenko

And I constantly learn things. Some new, some old. But I learn new and intriguing applications. It get me thinking and I appreciate this a lot.

---

Same same. How gray the world would be if we couldn't explore new ideas. I think if I ever get into such a rut where I was just slogging the same thing over and over without learning anything new I'd probably drive my car off a cliff.

I have a good few hundreds or more hours on our old super flo flow bench.I have missed it dearly the past 12 years. I was going to make one hear at the house for my own stuff but...Im trying to down size and not doing much more engine& head work these days.it's too much like work to tear car or toy appart to do it. I dont really need flow bench any more, but it sure is nice to have and test with. theres lots of hidden hp in cylinder heads. especialy older ones. I would love to pul the ehad of my new cars and do them up, but..again too much like work.and I nolonger race. but what gets e is...Ive never messed with those heads....so I dont know exzactly what they may or may not need without having one in my hands. if I ever need to remove one Ill know what to do when I see it.probably wont be a lot.or it may be a lot, you just never know...untill you do.