Experiences/questions from a Cyclone Maker

[outback]

YES I dare to dabble in the dangerous waters of dust collection and cyclones!

This arvo I finished making a mini cyclone, (the same as Waynes mini cyclone on his web page). MK1 crashed and burnt on take off, i.e. all the dust etc. ended up in vac bag. I think the problem came with the inlet pipe not being at the correct angle.

MK11. more careful construction this time and a big difference in performance. As the cone was made of clear poly carbonate sheeting the cyclone action was very evident I used it to pick up everything from fine sawdust to shavings from a hand plane I have been tuning.

The upper cylinder is made from 150mm (ID) PVC pipe, when I tried to attach the 50mm PVC inlet and outlet, there was not enough room for both.
Thats really poorly explained, If you look down from the top the outlet is actually 56 mm OD, this takes up 28mm from the centre line, leaving 47 mm to the side wall, not enough for the 56mm OD inlet, to get round this I trimmed the end of the inlet at an angle.

I used a plastic drum with a screw top for the collection drum, when I turned the vac on, it sucked the sides of the drum in.

On completion of the test launch, about 70% of everything, (dust shavings etc.) ended up in the drum, the rest in the vac bag, but this was almost entirely made up of plane shavings, I guess I was asking too much but hey it was fun seeing the suckers fly!

Questions for all.
1. Should I have the inlet/outlet pipe clashing problem? If not wot did I do wrong?
2. Whilst the collection drum was made of pretty light stuff should I get this much "suction" affecting the drum?

Overall impressions.
YOWWWWWWWWWWEEEEEEEEEE
This is gonna be great for all the little toys, The drum is easily fixed if my side sucking experience is normal.
The pipe clashing may have something to do with this, ( more suction on the outlet than the inlet).
Now I feel game enough to try for the BIG SUCKER for the plane shavings, table dust etc.

[Wayne Davy]

outback,

Sounds like you have a light weight plastic drum for collection so, yes, it will collapse if your vac is strong enough (which it is by the sounds). If you read up on Bill Pentz's site about the big ones, he makes mention of some users of the big cyclones actually collapsing their 6"/4" metal ducting if they closed all the blast gates at once. Even reports of collapsing the Cyclone itself when too light a gauge metal has been used. So, get a stronger drum and that problem should be solved.

As for only getting 70% of the dust/chips in the drum, I can only assume that you have some of the dimensions could be wrong i.e Cone/Drum length/height OR the outlet pipe is projecting down to far into the cone OR the vac is so strong its sucking the dust out of the air stream. When I made mine, I used Bill's spreadsheet to calc the sizes and was very careful to match the dimensions of all the parts taking into account the blower (vac) size. I'm not getting any dust in my old Elec. Lux but it is does not have a huge amount of suction to begin with. (Have a look at increasing the cone length out to a multiple of 3.)

As for fitting the inlet and outlet pipes into the top drum, yes, mine were a bit tight as well but I simply heated up and flattened out the inlet pipe (the one in the side) just a bit so it was slightly oval which gave me the clearance I needed. I had better check that I mentioned that on the web site now that you mention it.

Cheers,

[outback]

Thanks Wayne,
The drum is light weight, so now worries there.
I also used Bills Spreadsheet to calculate everything, Whilst I was careful I am happy to double check everything.
My vac does have a fair bit of punch, so maybe that explains the dust chips thing, curiously it is almost entirely made up of the big plane shavings!

Looking at the old thread you started detailing the construction of your mini-cyclone, you fared the bottom of the cone

Cutting and Flaring out the bottom of the cone to make a ring that will allow me to connect it to a collection drum with some flex pipe. .
In the next picture the bottom dust chute has been fitted.
If I am correct the bottom of the cone should have a dia of around 50mm. On mine I omitted the dust chute, and 50mm ain't that big for letting large quantities of big shavings pass through.
My thought is the chute is not needed when it is placed in a drum, (same as you did). And it is designed for smaller particles not big stuff.
I am most interested in your thoughts on this.
Basically I am VERY happy with it, maybe a bit of fine tuning needed.

[Red neck]

outback

I would suggest that you have a velocity problem through the cyclone. Too much velocity!

A vacuum cleaner will provide a high velocity flow through a small diameter duct. Conversely a standard DC will provide a much lower velocity of airflow through a larger duct. A given size particle would require a specific air velocity to carry it through a duct. The job of the cyclone is to reduce that velocity so that the airflow will not support the particle within the cyclone allowing it to fall to the cyclone collector. I would suggest that the higher velocity airflow generated by the vacuum cleaner cannot be slowed sufficiently in your cyclone to extract the larger particles therefore they continue to the vacuum cleaner bag.

A further problem here is impeller design. The blades of a high-speed impeller would be set to provide optimum performance at a given air velocity. Something like variable pitch propeller blades that operate at coarse pitch in high-speed flight and flatten to fine pitch for slow speed flight. If you slow the airflow too much the impeller efficiency will fall off further reducing suction within the system. Something akin to compressor stall in a jet engine.

There are various devices for measuring velocity but probably the simplest would be a manometer. Suction is simply a measure of static pressure and since total pressure is the sum of static and dynamic pressures, a low static pressure means a high dynamic pressure and vice versa. You would need to measure pressure at the entry and exit points of your cyclone to determine the net suction drop across the cyclone.

A manometer is a relatively simple device and I would imagine a search on the internet would reveal some designs based on a ‘U’ tube, a scale and a calibration chart for temperature variation. You would also need to build a ‘pitot’ tube and you would probably find that on the net too. If you were to research the site that Wayne suggested you may find some information that conforms cyclone entry and exit velocities to impeller entry velocity. You just need to match the two systems so that they are capable of the same mass airflow.

[Wayne Davy]

Outback,

I think that both you and Red neck are correct. Firstly, I think red neck is spot on that your Vac has too much suction for the size of the Mini-Cyclone (what the hell is the brand of vac btw - I want one). So, you may need to look at a larger design to take into account the power of your vac.

Now, before you go off and build a new one, try this. As you said that you are mainly only getting the larging shaving in the vac bag, I suspect (as you have surmised), that the cone outlet is to small at 50mm. Try enlarging it to 60 or 70mm and see what happens (ie. cut off about 10-20mm from the cone bottom). Maybe, due to the small size, the shaving as clogging in the bottom of the cone and getting pulled up the outlet pipe.

Give it a try - I dont think you will loose any current efficiency and can only gain.

[soundman]

Outback mate

plug these figures into bills spread sheet & see how you go.

inlet diameter 56mm (od 50mm plumbing pipe)
cylindre diameter 222mm
cone ratio 1.6

this will give you a outlet diameter of 110mm (od 100mm plumbing pipe)
and a trash outlet of 74mm (close enough for 75mm coro dust pipe

If check some of bills comments the outlet diameter should be significanly larger than the inlet.

Proportion, I think is the key to making these things go, I think yours don't look quite right.

I'm about a day behind you building my first of the above proportions & it looks right. Mine is steel spot welded & soldered

I half built one of larger diameter for a 50mm inlet & it looked wrong so I trashed it I pored over bills words some more & came up with the above.

I think imput velocity can be high but output velocity should be lower.

As for sucking your bucket flat.
Ive been using a quasi-cyclonic bucket for years & i can tell you even it flexes under vacume & its ribed & 2mm thick,

I'm hoping a 20L paint drum will hack the pace but we will see.

to adapt the 100mm outlet use a 100 to 50mm plumbing reducer, I recon the 100mm pipe should come out the top some distance to smooth the output flow.

If you are having difficulty geting the inlet pipe to meet the cylinder, go back to your old tech drawing books & develop the penetration out. its not hard once you do the revision.

cheers
will post pics when finished.

[outback]

The above comments certainly give food for thought. BAsically everyone is getting at the same thing, to much velocity through the cyclone, meaning some "stuff" gets sucked through, instead of being allowed to slow down enough to fall through.


Soundman may have be onto something with the inlet/outlet dia. By cheating a little and trying to use what was on hand I may have overstepped things a little, however, Waynes cyclone works really well using the same design.

Conclusion: I'm gonna play with the current model, I have added the dust chute, I will change the inlet pipe as Wayne suggested, then vary the length of the outlet pipe. Possibly make the vac somewhat less efficient to lower the suction. If this fails I will start hacking a little off the end of the cyclone bottom to allow larger particles through.

By this time Soundman will have completed his, and I can learn more from him.

[Red neck]

Outback

Just some further thoughts:

The top section of the cyclone does the sucking and blowing. The wall design handles the change of velocity. Too low a drop is inefficient. Too high a drop is inefficient. The cone helps gravity remove particles. The cone shape will have some effect on velocity control. At some point in the cone the airflow must change direction upwards. The heavier matter should fall below that point so it continues to drop and cannot be supported by the exhaust airflow. Something like the balance point of a fulcrum!

If you play with the length of the vertical pipe (and I am basing my theory on cyclone cross sections I’ve seen published) you should be able to move the ‘balance’ point up or down the cone depending on whether you shorten or lengthen the vertical exhaust pipe.

I would expect that velocity into the cyclone should be in the vicinity of 3500 to 4000 fpm. Your vacuum is possibly much higher than that.

[outback]

Just finished another play session.
When I made the cyclone I inserted the outlet pipe to the depth according to Bill's spreadsheet, but because it was round 50 mm pipe and not a rectangle section, you can't get the measurement right from the top of the cyclone and the bottom of the inlet pipe, which is what you are saying Red.
So, I played with the pipe length, as the cone is polycarbonate, it is easy to see what is happening. By adjusting the length, you can see the larger shavings swirl at different speeds and different heights up and down the cone. So it is pretty easy to find the "sweet spot".
As the upper section is 150mm pvc pipe, (160 OD), I should infact have an 80mm outlet, far bigger than my crummy 56 OD now. This has gotta have a major impact on air velocity as it leaves the cyclone. I will try and rat out something at least close to 80mm and try this.

Also when I made the vac less efficient, ie "suck less" by allowing air to enter the vac suction hose, the whole setup became more efficient, that is more junk ended up in the drum, of course I had less suction at the inlet ducting.
Even so the efficency rating has increased from 70% to 'bout 90%. Not too shabby.

[Red neck]

Outback,

If you can see what is happening inside the cone try using small lengths of wool secured with duct tape but with a tail long enough to indicate the airflow. If you stuck them in a line, say one inch apart or closer, down the wall of the cyclone, you would get a very good picture of the airflow. At some point you should observe the change of direction from a downward spiral to an upward flow. Using a telescoping tube in your vertical pipe you could slide it up or down until you find the correct airflow reversal point.

I flew an aircraft once and the engineer had stuck wool all over the starboard wing to 'watch' the airflow through various flight regimes and configurations. Very interesting experiment!

Re your pipe diameters; if the airflow is entering through a large pipe and exiting through a small pipe the velocity at exit will be higher than entry (Bernoulli's theorem) subject to any loss between the ports.

[outback]

Thanks for the idea on wool tails, I will try this as soon as I can.

As far as the pipe dia's is concerned you have said betterer than me, my idea. (????)

After having a good play with the now infamous spreadsheet, you have the opportunity to adjust most sizes, but if you leave well enogh alone and just pick your duct size, cyclone dia and you let it come up with inlet, outlet sizes it gives you a rectangle inlet duct 38 X 75 or 2850 sq mm, and an outlet dia of 75mm or 4415 sq mm, almost a third more capacity, this will help slow velocity, as opposed to using the same size inlet and outlet.

I gotta admit though, If I don't get any more improvement I'll be happy, but it sure is fun trying.

[soundman]

I am happy to report that I have just had my firts cyclone running & it works well as far as I can tell so far. seems to seperate all but a small ammount of fine dust a has minimal flow & static preasure loss. dimensions as above.

sugestions for your experiment.

Go with an inlet of 32mm electrical conduit (it will mate with most standard vac hose.
and an outlet of 90mm storm water pipe.
it should all fit in the interior of 150mm pipe (with a little shove & a bit of heat) & give you the desired effects.
5/8 diameter seems to be the recomended output tube depth(journal of cotton sience)

I'd be interested in how you go.
I will post pic of mine soon. (still on 35mm)

cheers

[outback]

Just finished doing a bit more playing and fine tuning of the Lil' sucker.

I changed the drum to a 25l (app) steel drum, which has a clip to hold the lid as well as a foam seal, all help to maintain a good seal, i discovered a small air leak between the cone and drum lid where it enters when I first attached it, this had a major influence on efficiency, at first I had wondered what went wrong!

After this I gave it a good workout in the real world. I did some planing then a heap of sanding using my ROS, the final result was a very small amount of the larger, light weight plane shavings in the vac, (I tend to think these are acting somewhat like a feather, inthat they take very little air to keep them aloft), I couldn't detect any measurable amount of fine sawdust in the vac. This is the main concern as the vac has a paper filter, this will block up real quick with fine dust.

So now I can spend more time making sawdust, but no-one can tell. :D :D :D :D

[soundman]

Since my last post ive had my 222mm cyclone up & running, (be it strung up on a ladder).

I gave it a real hiding, first I stuck the hose into the previous dust seperator that was almost full & and sucked up the contents, then on to the contents of a 1/4 full dust extractor.

just about filled the 20l waste drum. the dust was a good mix including a heap of fine sanding dust, router chips, aluminium swarf, & chips from the spindle moulder.

all I got out of the vacume cleaner was a fine film of dust on the canister & on the filter bag, a pinch of little bits & 1 dady long legs spider (still alive).

i think one problem with shop vac duty is that the airflow is not consistent & may stop & start suddenly due to clogs & such.

I think this may effect the over all efeciency.

Important point
while the rig was running I saw arcs over 25mm long jumping from the metal cyclone body to the ladder frame it was hung on.

From my reading this is to be expected from metal dust hardwear.
I fitted an earth jumper lead, end of problem.
I would strongly recomend that any metal cyclones, ducting & bins be earthed. ( as seems to be normal practise).

cheers.
no pictures yet.

[outback]

I quite agree with the varying vacuum created through a nborma vac as opposed to a regular DC, causing a dgree of irregularity i performance.

I have noticeed with mine that when sucking big piles of rubbish that is when the system seems to suffer.

This will never happen under "normal operating conditions."
The 'Lil sucker is just that for little toys, for the big toys a BIG Sucker is required, but now, with my new found experience, I am pretty confident I can manufacture one of these as well.

BIG thanks to Wayne for his help and inspiration for this project, also to Red and soundman for their input during the manufacture and test run stage. Now I'm free to be an all day sucker!

[Wayne Davy]

Outback,

Here's your certificate mate (although I would like to see a pic of the finished product ;) )

[outback]

Aww shucks, thanks Wayne

Pics will come. SWMBO is having a birthday in a couple of weeks, and she is getting digital camera, but no one tell her OK. Once it is here I will happily post apic of my basic model cyclone.:) :p

[Neo]

Hey Outback,
a quick question for you. How did you acquire a polycarbonate cone for your mini cyclone? If you made it could you please tell us how? I can think of few things better whilst enjoying a quiet ale, than watching the dust go round my cyclone.

Thanks

Pat

[outback]

Pretty simple really.
I made it from polycarbonate sheeting I had on hand. I originally bought some to make 'lil transparent windows in a hopper of a machine I use for work.
it was purchased from the friendly hardware shop man, pretty dear per/M but worth it, as you only need the best part of bugger all for a 'lil cone.

A note of caution, when you form the cone, getting the seams to meet neatly, and stay like it ia all a bit tricky, I riveted mine, then applied a seam of tape along the join to make it air tight, and to neaten it. I gotta admit this was the hardest part, cutting the sheeting is a snap, rolling it is no probs, just keeping it all together whilst you join it.

Watching the dust, shavings etc. swirl around and down into the collection bin is kinda cool.

[Bob Willson]

Did you contemplate butt gluing the edges instead of riveting them?
I ask because I have had thoughts on the subject myself and had intended to use a polycarbonate glue called Acrifix 192 that gives a glue joint that is practically invisible.

My original application was going to be for a fish tank filter but the basic problem would remain the same and I will get around to building a cyclone eventually.

PS what thickness of sheeting did you use to get the small radius required?

[outback]

Sheeting was 'bout 0.75 mm 1mm max, so butt joint not possible, gluing the overlapped seam would be terrific idea, still have the problem of holding it all still, but the end result would be great.

[SteveI]

I used a polycarbonite too (from Bunnings) - Laserlite - this is their website

http://www.laserlite.com.au/product...ge=Introduction

I used pop rivets but sealed the overlap with silicone (in this case All-Clear)

Just held it together with clamps to drill two holes top and bottom - for pop rivets, squeezed in the silicon then drilled the rest of the holes and inserted the pop riverts - sealed up well

[mmmmbeeeeer]

Has anyone tried using perspex for the cone? If so, I was wondering how one would deal with the static electricity. I was thinking about running flat copper tape or similar around the inside of the cone and grounding that.

I just like the idea of being able to see what's going on :-)

Pete

[craigb]

Don't know the answer to your question, I just wanted to say I love your nickname :D

Why didn't I think of that :( :D

Cheers
Craig

[Glen Bridger]

Hi,

To overcome the problem with the 50mm PVC pipe, use 50mm electrical conduit. Remember elect conduit is tubing and measured on the outside diameter, except 100mm. The other thing to remember is to use the actual outside diameter of the 150mm PVC pipe used for the cyclinder into Bill's spreadsheet. I have some at home which measures 161mm OD.
I've yet to make one, but it is on my long list of stuff to do.

Glen.

[mmmmbeeeeer]

think of Homer Simpson when you say it and you've got the idea...

:D

[mmmmbeeeeer]

Back on topic though...

I've spent a couple of hours reading through all the different threads about cyclones and I have one question which someone may be able to answer.

I, like most people its seems, am vertically challenged in my workshop and I was wondering whether anyone has tried mounting the blower to the side of the cyclone and using 6" or greater duct from the cyclone to the blower?

Pete

[outback]

You really need Wayne for this one, but I'll have a go.

If you leave the dust extraction system (DES) as a pull through, that is machine to ducting to cyclone to blower, ans isolate the blower from the cyclone i am sure you will get a reductiuon in efficiency. The only extra vertical you need is basically the width of the blower.

If you are contemplating a DES on a push through system That is machine to ducting to blower to cyclone you will find you receive an increase in air flow over the pull through setup. :). I am pretty sure Bill's site recommends this option. The downside of it this is all the dust, chips, nails, chisels, planes, (OK I'm gettin' carried away). go through the fan, possibly causing damage. BUT, on a normal bag DC this happens anyway.

Hope this helps, if not wait for Wayne he'll set you right.

[Sturdee]

I, like most people its seems, am vertically challenged in my workshop and I was wondering whether anyone has tried mounting the blower to the side of the cyclone and using 6" or greater duct from the cyclone to the blower?

Pete


Pete,

Have you seen the posts by Geoffs on his cyclone set up, I think it may be what you are looking for.

Details are http://www.woodworkforums.ubeaut.com.au/showthread.php?t=6014&highlight=results and on his website http://users.ncable.com.au/gsyme/woodwork/index.htm

Peter.

[mmmmbeeeeer]

Thanks Peter, that is indeed what I was thinking of although I was thinking more along the lines of using standard ducting.

Now I'll have to ponder the merits/otherwise of push-through versus pull-through...

One day I'll actually get my act together and build something. :o

[Wayne Davy]

Pete,

When I was researching Cyclones, I too thought about making a Push Through type. Being unsure of how it would go, I asked 'The Guru' Bill Pentz in the USA. Bill sent me the following response:


In theory there is no difference in efficiency between a push through and a pull through. In practice, most hobbyists try to make do with a 2 hp or smaller blower that is severely air starved. The result in a pull through is bad performance because the cyclone adds even more resistance on the inlet side. In a push through, it works far better. So, the answer to your question is the push through will actually work better for you with that 2 hp setup, except of course the things hitting the impeller. Make sure you use a metal can with a push through! The potential of a spark from a piece of metal hitting the impeller and getting put into your collection barrel is far higher with a push through.
Bill also sent me a pic of a cyclone that was made Mobile with the blower at the base like Geoff's (see attached)

Personally, I decided to make a pull through mainly as I was worried about a possible spark in the bin as Bill mentions.

Have you seen my site btw - might give you some help/ideas. Check it out at http://members.optusnet.com.au/waynedavy/

Best of luck,


(woo woo - 1000 post's by little old me :D )

[craigb]

think of Homer Simpson when you say it and you've got the idea...

:D

Yeah I got the reference. :rolleyes: