DIY FILTER-NO FILTER MEDIA EVER

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
Well in my fight with my dinos I needed to get a UV sterilizer. Guess what too expensive so I have an idea.

Can we pull the bacteria out of the water column with some form of particle separation technique.
The answer is yes and it uses a hydrocyclone. Essentially you make a tornado and particles collect in the bottom of the tornado. You collect the solids and toss them out. It's a continuous method and can be used in place of your filter socks too. No more cleaning socks.

So here it is.

https://www.researchgate.net/public...cyclone_as_a_maintenance_free_aquarium_filter

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

I'm going to try and build one.

This could also be used in gentler flow to collect eggs, brine shrimp.... Think about it. cleaner clearer water with no moving parts other than a pump. Like a skimmer for solids. why this is not already done I can not imagine. I guess I am about to find out.

For those that want to play along get a piece of pvc and put an inlet in the side for water to flow into. Make a cone for the other end. Blast the water into the pipe and see what you get. I'll be doing the calculations and determining the right sizes. The other part will be durability as the abrasion of the particles might be a factor.

Never know till you try. Hope you enjoy the idea sharing. Some may work some may not but thinking and sharing are the way I like to live.

Edit: could also be used on the overflow to pull out larger stuff passively powered.
 
BRS

Mandelstam

Well-Known Member
View Badges
Joined
Oct 29, 2017
Messages
685
Reaction score
1,107
Location
Malmö, Sweden
Interesting! Following! Please post more as your experiment continues, even if it doesn't pan out the way you thought. But reading about it sounds potentially promising!
 
Top Shelf Aquatics
OP
kecked

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
I uploaded a pdf in case people have trouble getting to researchgate from a none edu location. This pretty much seals the deal and describes the calculations needed. Yes a 3d printed product seems very possible BUT most 3d printed plastic won't handle the water long term. I'm thinking pcv pipe and stepping down the sizes to make the cone. The study shows we can separate particles in the 1-5um range. In fact we can classify the size if we wish so we don't pull bacteria out all the time. maybe just at night. Maybe just on a bucket of water we pulled out of the sandbed. Who knows. BUT we are on the right path. UV breaks the organics down. I think it is better to pull them out intact.
 

Attachments

  • 1-s2.0-S0263876217304215-main.pdf
    1.6 MB · Views: 423

Mandelstam

Well-Known Member
View Badges
Joined
Oct 29, 2017
Messages
685
Reaction score
1,107
Location
Malmö, Sweden
In the wiki article it says that the pressure drop from the pump should be 20-60 psi. It was a long time since I did any physics calculations but is that a reasonable pressure we could expect from a "normal" centrifugal water pump for this purpose?

Found another paper on the subject, have to pay to read it though.
"Bradley Hydrocyclones: Design and Performance Analysis"
https://link.springer.com/chapter/10.1007/978-94-015-7981-0_1
 
OP
kecked

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
Well I know the one I use for the RODI does 90-100 psi all day. Might need a prefilter (i cry thinking of it) as the size of the plumbing will matter a lot and need to be small. I have literally done zero calculation yet. Just thought of this at lunch today. Those are large cyclones. I'm thinking 4" diameter no more than 12" long. Has to be small. remember we get to do thousands of passes a day.
 

Mandelstam

Well-Known Member
View Badges
Joined
Oct 29, 2017
Messages
685
Reaction score
1,107
Location
Malmö, Sweden
With a larger pipe that pressure would be a heck of a lot smaller though... I'm just thinking out loud here but if the cyclone requires (inlet diam, etc) a monster of a pump to keep the pressure up it could be a problem...
 
OP
kecked

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
Won't know till I run the calculations to see what is required. Also reading about conical shapes. They make a difference. Also dissolved air might be a cavitation issue. Lot's of issues. Still exciting to think about. I'll try and make a calculation spreadsheet so we can run numbers. The math is not that hard. Can't do that till the weekend. The pressure is not the most important part. The rate of spin is for particles the size I want to trap. How you transfer the force and decrease the inner drag is key to efficiency. Turbulence is also an enemy so the rate of flow has to stay laminar as much as possible. That puts a limit on the spin rate. Pressure drop also counts as if we form a vacuum in areas we pull gas out and make turbulence. This is a well studied area so we just have to apply the physics to the problem. I also contacted a couple manufacturers. If I can buy it off the shelf I plan to.
 
BRS

Mandelstam

Well-Known Member
View Badges
Joined
Oct 29, 2017
Messages
685
Reaction score
1,107
Location
Malmö, Sweden
Just had a thought about fabrication... If 3d printed abs plastic doesn't hold up in the long run it can at least be used to print a negative and then create a mold. I'm sure there is some suitable and castable materials out there to be used. Heck, maybe even concrete lol. I assume it won't be that large so the wall thickness is not really an issue and doesn't have to be uniform. It's just the shape of the empty space that's important really...

As usual I'm too far ahead of everything so I'll just see what you come up with lol.
 
OP
kecked

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
Here is a calculator to do it all along with a diagram to explain it. This looks very possible to build. Let me know how you do. I'll publish my results and selected sizes. I'm going to try and 3d print this in nylon and see what happens. If I get that far I''ll put it on thingaverse. Part of the tricky part is the waste. It looks to flow out the bottom so how much water is this going to waste.
It might turn out to need a post filter to trap the waste and let the water out. At the 1-200um size that is going to plug fast if you are concentrating the waste soooo. maybe a hybrid where you concentrate the waste and then treat in a UV unit and release. Not quiet what I had in mind but a possible solution. This would then just be a pretreatment concentrate. Hoping I am wrong and can just make the solids collect in the bottom. Maybe I'm finding out why this isn't common.
 

Attachments

  • Cyclone-Design-and-Analysis.zip
    29.7 KB · Views: 193

reefwiser

LMAS
View Badges
Joined
Nov 24, 2013
Messages
7,451
Reaction score
9,359
Location
Louisville,Kentucky
biggest problem with this is that the size of the dino's is too small for this type of filter to work. We use this type of filter in the food industry to remove food bits from our vacuum systems. It is mainly for large particles for food. Not the micron size of dino's Best to use a Diatomaceous earth filter which will filter into micron size needed
 
Top Shelf Aquatics

ithk21620

Active Member
View Badges
Joined
Mar 15, 2012
Messages
463
Reaction score
246
Location
Kellner
Hydrocyclones are used in potato processing all the time to remove the starch from the process water after the potatoes are cut. You can buy them in a wide variety of sizes and materials. The problems I see with using them would be:

1) The pressure needed for them to work properly is around 65 psi so you would need a pump to feed the hydrocyclone which is possible but most people prefer to run fewer pumps than more.
2) You would constantly be wasting water. For them to run properly you always have a steady stream of water coming out of them. In the potato industry we take the water from them and then run that water through a second set of hydrocyclones to recover as much "good" water as possible. So if you wanted to do that you would need yet another pump but you would still be wasting water in the end.
3) You could get around wasting the water by running the underflow discharge through a filter sock or something similar but then why wouldnt you just run filter socks in the first place.

I am by no means telling you it wont work, I am just trying to explain why I don't think it would be an efficient filter for an aquarium.

Here is a company that makes hydrocyclones if you wanted to buy one that is already engineered to play with.
https://hydrocyclone.com/Urethane-h-clone.htm

Edit:

Forgot to add that potato starch ranges from 15-75 microns and dinos range from 15-40 microns so it could be possible to remove dinos with this method.
 
Last edited:

ithk21620

Active Member
View Badges
Joined
Mar 15, 2012
Messages
463
Reaction score
246
Location
Kellner
Here is a picture of a hydrocyclone skid that is used for removing starch

 
OP
kecked

kecked

Active Member
View Badges
Joined
Dec 9, 2017
Messages
338
Reaction score
198
Wow that's great info. I'm going to see what they charge for one. Another possibility is to run it and let it all flow back in. According to this article the g force is like 5000 G so it might kill the bacteria. http://infohouse.p2ric.org/ref/13/12362.pdf

The best I've been able to calculate is an 8" wide 24" long unit with around 96% efficiency. 1/8" inlet at 60 psi with a 1/8" underflow outlet and 2" overflow. Trying to design around one of the pressure booster pumps used for RO/DI water. I wonder if I put a box on the bottom if the solids would fall out the bottom and collect and the water flow back up through the center clean flow. Now there is no wasted water.

Hey if you don't ask or try you never know BUT it is looking less viable by the day. I nthe end it might make a nice water polisher water change device. Clean the water while you take your cut out for water change time. I think this is looking about the same cost as the UV method.
 
Avast

Do you have any special reefing upgrades or projects planned coming up?

  • Yes (please tell us what in the thread)

    Votes: 92 52.0%
  • NO

    Votes: 64 36.2%
  • Maybe (tell us)

    Votes: 19 10.7%
  • Other (please explain)

    Votes: 2 1.1%

Online statistics

Members online
2,591
Guests online
6,012
Total visitors
8,603
Coral Frenzy
Top