Mechanical filtration choices discussion.

[koiguy1969]

If you run a 2" pipe up into a 6" pipe... that 6" pipe isnt just 3 times as large its actually about 7 times as large. you can see by this pic. not including the empty space between annd around them.
so its dropping force and speed of flow significantly.
so if your feedin the systm by 1" line and it hits a 2" upflow pipe. theres the first significant drop in speed and pressure. then it hits a 7 or 8 times bigger down flow. the debris will fall much slower, then the water slowly turns upward in an even greater displacement of flow. rising slow enough where the debris isnt carried with it.
I'm not saying your thoughts wouldnt work, but it seems to me, the more times the flow can slow and change up and down direction, the more effective the filter. do your experiment with both ways...post your findings.

 

[HTH]

Thanks for the reply it caused me to do some thinking. Yes dangerous but it is what I wanted. 97F outside so I can burn some time thinking and typing.

Settling has little or nothing to do with pressure. But for the first down pipe.I would only expect to find tiny pressure differences in the filter other then pressure due to depth.. It is a lot like filling a pail with water. The hose has some pressure but does not contribute to the pressure in the pail. Pressure in conjunction with the cross section and length of a piper determine the linear speed at which water moves through it. The linear water speed and it being uniform in any cross section are the parameters.

I am aware of the diameter vs cross section. My point regarding that is that removing the pipe to do more good then the benefit it provides, even with the objection you raised. I do not see where that top turn does anything useful. More on that below.

There is no magic in the transitions. It is not like a carburetor. One large one is just as effective as several small ones at slowing the water providing you do both without turbulence.

Suspended particles have little momentum. They are not going to go straight when the water turns as they have near the same mass as the water they are suspended in and what momentum they do have is quickly used up as they attempt to do anything but move with the water. This logic tells me the top turn is useless. Shortly after making the top turn friction with the water will have accelerated the particles to its speed even if there was a loss.

I wonder what percentage of the settling happens while the water is moving nearly parallel with the bottom. The water does not provide much up force and gravity can pull the particles down.

There is an effect that may be taking place on the bottom turn. As solids build up on the bottom they tend to snag other solids as they go by. Solids that may not have been caught had the earlier solids not been there. I see that on the top hat on my barrel filter. Solids build up on the sides as well as the bottom of the hat. Think small vortex on top of the barrel feeding the 4" down tube. I just wiped the crud off the side of the hat. I should have shot a video.

I did not offer to do the experiment, just said it would be interesting and I am fairly sure I know the results. If I did all the experiments that I find interesting I would get even less done. Part of me was hoping I could lure some one else into running it

If I can be allowed to use a word not from my generation: This post has caused me to think that a mechanical settling based filter where the water moves primarily up and down it whacked. A filter with primarily horizontal flow would be much more effective. The problem is flushing it. hmmm

Howard

 

[crsublette]

I think both of your ideas can be combined.

The idea of having the pipe increased in diameter prior to entering the container so to reduce the friction loss and flow rate of water to be closer to that of gravity and then have the water hit the "top turn" diffuser plate to further slow down the water due to the influence of any extra existing pressure created by the initial flow rate. Since, generally these types of particulate filters are utilized for larger particulates and with a water gravity flow system, the diffuser plate would have to be a reasonable distance away and, constructed in a way, still below the input at the top, so there is not a daming effect of big debris potentially causing problems such as clogging or unreasonably slowing down the totality of the initial flow rate.

Due to the top diffuser plate, the debris will be begining its descent from the top of the water volume. This would cause problems when the output is also located at the top. So, a type of wall would have to placed in front of the output and possiby quilt battling could be between this wall and the output.

This way, the drain can still be at the bottom to remove the big settled debris and a pressure gauge could be used to determine when the quilt battling is clogged.


I would differ with one point about the pressure. The water's linear speed, or flow rate, actually does provide the appropriate pressure per square inch to interrupt gravity thus keeping the particles suspended; otherwise, the particles would not remain suspended. However, the only exception to this are filters constructed to utilize the shear force of a cylcone, which the pressure created is so great from the centrifugal force that the fine particles are thrown out of suspension and this is put into practice with the multi-cyclone polishing filters, except these type of filters are too small to be for large particulate, such as is filtered by settlement chambers.

The context of these particular settlement chambers, such as this radial flow, involves seperating noticeably bigger particulates such as dirt, algae, fish excreta, etc. Very fine suspended particles are outside the context of this type of filter due to the container's size and construction. Without implementing a cyclone construction, the water would have to be incredibly still, thus the container would have to be quite large or plumbing quite involved, to remove any finely suspended particulates without the aid of other material, which I think is the point of the quilt batting and strapping in koiguy's picture to remove the finely suspended particulates.

 

[HTH]

Charles I will take another look at you post in the AM when my brain is working.

Just want to talk about pressure and force this time. We need to agree on the meaning of terms.

By definition particles in suspension will settle in still water. Friction between moving water will keep a particle in suspension so long as the water has sufficient linear velocity to keep the particle from moving down and reaching the bottom.

I would differ with one point in that the water's linear speed, or flow rate, actually does provide the appropriate pressure per square inch to interrupt gravity thus keeping the particles suspended; otherwise, the particles would not remain suspended.

There are two things here, water pressure and force imparted to particles due to linear motion.

Mix up a slury of water and dirt and place it a pressure chamber. It settles. Now pressurize the chamber and shake it again. It will settle just the same way. Clearly water pressure has no effect on settling.

Now we might say that the friction from moving water is exerting pressure on the particle. That may be what you are talking about. If that is the sense of what koiguy was saying I did not get it. Given that the context is water aka fluid dynamics I think we should be calling this sense force rather then pressure. It is a better fit. (I should go back over these posts subing force for pressure)

1. Force is the total impact of one object or gas on another
2. 
   Pressure is the ratio of force to area over which it is applied
   

I am going to run this experiment. I am going to take water from the outflow of my barrel filter and see how much additional material will settle out over time. It should give me an upper limit on what is possible using sedimentation.

 

[crsublette]

Yep, fair enough. Water in of itself does not have any inherit force in a pressurized chamber.

Yep, I got confused by the usage of the word "pressure" since we are actually talking all about the force from a flow rate. I don't think the term "pressure" properly applies in this context, unless we're talking about the pressured created by the force of the flow rate against an area. So, I substituted "force" for "pressure" as I read it, which is incorrect as ya correctly point out.

Correct, flowrate in of itself is what creates the friction that is exerting pressure on the particle area to force it to remain suspended. This is exactly the jest of what I thought koiguy was talking about.

 

[crsublette]

Generally, settlement chambers are not pressurized chambers, but I do not think a pressurized chamber would interfere with the settlement.

Actually, the construction of this is not that far off from the combination particulate + bio-ball filters out there sold in stores. There are some elaborate ones I have seen which allow ya to use a blower on the bio-media like a bead filter, that is in a seperate chamber in the same container, and then a drain with a sludge trap for the settlement chamber portion of the container. Although, I would not use this as a bio-filter. I've never been a big fan of particulate filtration combined with bio-filtration in the same container.

Here is a picture of what I was thinking. I think this is entirely doable. I can think of the easily accessible materials from a local hardware store to do this with using a plastic welder, pvc panel, pvc adapters, cut in half pvc pipe for the diffuser, a 5 gallon bucket, and a gamma seal lid.

When the quilt batting starts to get dirty and clogging, then the pressurized gauge would give an indicator of this. However, if this was a gravity flow system, rather than a pressure gauge, then a simply open stand pipe would give an indicator, that is water over flowing out of it, of when it is time to clean the batting.

 

[crsublette]

Yeah, I don't think it is helpful to get distracted by arguing about the proper usage of words. I think we know what everyone is talking about in this context as to what we are trying to accomplish with a radial flow settlement chamber.

All we are essentially talking about are the tricks of a radial flow settlement chamber to reduce the velocity of the water's flow rate by removing friction loss in a bigger pipe and forcing it to change directions in or after this pipe, which I think a diffuser plate also forces the water to change directions thus reducing the water's velocity.

I think we are trying to maintain the same flow rate while reducing the water's velocity so that the suspended particles are allowed to settle.


I would probably change it also, so rather than the input coming straight down vertically, the Input comes in horizontally then down with a smooth 90 degree into the bigger fitting. This way additional vertical head height will not contribute to the water's velocity. A smooth, rather than hard, 90 degree so not to increase the friction loss, which would redue the flow rate. I think water velocity would be decreased by removing all friction loss, that is initially created by the 3" pipe, and then the remainder of the water's velocity is removed by the force of the water hitting the diffuser plate. Also, it would probably be beneficial to make the diffuser plate wider than the input mouth.

I don't know...

Howard, what do ya think ??


This would be a pretty fancy settlement chamber. I would probably just be easier to go with a normal radial flow settlement chamber, which does work by allowing dirt to settle as shown by the videos although I wonder about the fella's flow rate, or a non-pressurized vortex chamber that has a prefilter on the output leading to the polishing filter or bio-filter.

 

[crsublette]

Bah, can't go to sleep... So, figure I'll think about something.


Actually, from looking at Koiguy's design, then the bucket's lid is acting as the diffusor plate.


Reading more about this radial filter and retention times on the earthangroup website. Supposedly, there still needs to be a retention time of 30 seconds (the shortest, 75gph for a 5 gallon bucket) to 4 minutes (the longest, 600gph for a 5 gallon bucket) to allow the debris to properly settle. The fella on earthangroup seems to have a coefficient in mind to determine the proper retention time for the radial flow filter, that is to say only a maximum loading rate of 5gpm per foot squared. So, say the 5 gallon bucket has a 11 inches diameter causing the surface area to be 0.66 square feet; then, multiply this by the coefficient of 5gpm per sqr. foot to get 3.3 gpm for this particulare 5 gallon bucket surface area; now, divide the bucket's volume, 5 gallons, by 3.3 to give us a retention time of 1.52 minutes retention time, which this is around 200gph being pushed through the bucket.

I would imagine anything higher than these flow rates would not allow the debris to settle and just be pushed up into the media grid. So, personal preference, I would not sit the media grid so low, likely sit it halfway or quarter from the top, due to debris possibly being easier to get lodged into it. Also, I think the additional material would reduce the buckets retention time for debris to settle.


Now, if we're talking just making the 5 gallon bucket into another device to be a particulate filter with quilt batting, not a settlement chamber, then I really like Koiguy's design. Actually, I think this design, filled with quilt battling, would make for a sweet quick polishing filter on a pond, whether it was something temporary that ya attached really quick or something permanent.

 

[Mucky_Waters]

Now, if we're talking just making the 5 gallon bucket into another device to be a particulate filter with quilt batting, not a settlement chamber, then I really like Koiguy's design. Actually, I think this design, filled with quilt battling, would make for a sweet quick polishing filter on a pond, whether it was something temporary that ya attached really quick or something permanent.

Actually I made something very much like Koiguys design last year, it didn't work very well for a couple of reasons. The water easily channeled past the quilt batting around the edges of the container, and it was clumsy to clean. I ended up building something that works much better, and have continued to use it. I posted a thread about it in the DIY forum just over a year ago. It kind of works like a canister filter for pools or shop vacs, and it does polishes the water very nicely. I have two of them and can swap the dirty one out for the clean one in about one minute, but it does take about 5-10 minutes to clean the dirty quilt batting.

 

[crsublette]

Due to how the inputs are, I can see how the channelling issue would be created. So, the water would have to be dispersed through a column, such as your bucket acts as the column diffuser.

I'd figure something like this would be used for a polishing filter rather than a larger particulate filter like a settlement chamber. Personally, I'd just do a S&G, but something like this kind of batting bucket polisher would be cheaper to build. Yeah, the cleaning of the batting is what would take some time.


Probably would make a bucket into a radial flow settlement chamber and then have this water go into a bucket made into a quilt batting polishing filter.

Also, be sure to use a smooth 90 degree, or two 45 degrees, rather than a hard 90 degree pvc fitting so to reduce friction loss.

 

[crsublette]

For anyone curious on how to penetrate the bucket, ya can use threaded male/female electric plastic conduit adatpors and install them like a bulkhead. These conduit adapters have a lip on them big enough to allow a seal, but must be quite precise of drilling the hole, do not make the hole too big. Then, use some P&L sealant when screwing the adapters together to make a tight seal against the bucket's plastic wall.

 

[HTH]

I think one can build a better sediment filter then what people are using.

I have vivid memories of cleaning the rocks out of a 55 gallon barrel filter on a hot day. It is enough to keep me from attempting a 250 gallon S&G.

The difuser/baffel you have in the diagrams is going to cause turbulence. A cone like I used in an earlier drawing would smooth the transition from a small to a larger cross section.

 

[crsublette]

You got some good ideas man. I think the diffuser plate, as ya describe it as a "top turn" plate, would be an awesome addition to make a better settlement chamber.


The problem with the raised short walls of the baffle is that it is going to get filled up with debris but the entering water turbulence might just flush out any debris that might get trapped in it. I'd be up to trying it.

Yeah, I was wondering about if a convex plate would properly reduce the turbulence or if it would simply just displace it.

Rather than the convex plate, substitute it with a sharply angled concave plate with a cone glued below it. I'm actually liking your idea better.



For folk who may or may not know...

There are many variations of settlement chambers in an effort to find that "better" settlement chamber. The typical minimum size, that is completely dependent on the flow rate, of a basic settlement chamber is around 160gallons and all the way up to 1,200 gallons, which is crazy for just a large particulate filter. This is due to the fact of what it takes to reduce the water's velocity and the velocity is what is attempted to being addressed in a radial flow settlement chamber. So, there have been variations made such as VMS settlement chambers (like the Answer, which isn't made anymore), vortex settlement chambers with brushes or a vortex with a pre-filter. There have been all sorts of settlement chambers variations.

Personally, I am not a big fan at all of settlement chambers, except for I do like how they are simple and cheap to build. I think settlement chambers is old school technology that is being quickly replaced by sieves, which sieves unfortunately are a bit more expensive to DIY due to all of the extra material and plumbing. Here's the best DIY sock filter sieve that I have seen. Supposedly, the going rumor is that sieves are supposed to be "self cleaning", but I think this is just marketing hype since all of the sieve builds I have seen require assistance in removing the debris without losing quite a bit of water. The RDFs are the only truely self cleaning particulate filters, but they still dump quite a bit of water from the system's cleaning of the sieve screen.


A quick aside... Rocks should never be cleaned out of a properly built and pre-filtered S&G filter. I know folk that go on for 3~5 years before ever dismantling them, that is if they do it, since it is not necessary at all. These S&G filters are polishing filters so they would have to be properly pre-filtered to remove large particulates out of the water prior to entering the S&G. Although, I have seen them used as a large particultate filter, which I do not recommend due to clogging. S&Gs should be properly blown out every 3~5 days, which only takes a flip of a switch, opening the waste outlet, and around 5 minutes. S&Gs are not settlement chambers, they are polishing filters and a side effect is it also does some bio-filtering. I know there are concerns of obligate anaerobic zones forming, but, if the S&G is properly blown, then this is of no concern. S&Gs can be of any size, that is even a 5 gallon bucket, but the depth of the rock layers and size of the container will determine the filter's flow rate and the blower's strength.

I would go so far to say that S&Gs are better than the cheap bead filters due to the fact that bead filters require a stronger water pump to properly operate them. S&G's don't need a strong pump. I even know a guy that has replaced his expensive bead filters with S&G filters and have obtained much better results for his pond. Of course though, S&G's do have a bigger space foot print. I have also seen a fella build a 200 gallon S&G, but he was using one really big blower that was tied off to supply the multiple diffusers at the bottom.

S&G's are essentially bogs, can put some floating plants on top of the S&G, except far better than a bog since they are not a ticking time bomb like a bog can sometimes become over time.


Talking mechanical filtration is quite fun and tossing around that idea football quite fun as well. I think it is neat how quilt batting is used so much on this forum since it does work and it is cheap.

 

[crsublette]

Bah, I forgot to add a note to my pictures. The orange stuff in the polishing filter is the quilt batting strapped around the column diffuser pipe.

 

[Litzi1964]

I felt like I was totally over the top in pond filtration, but a 10 to 1 ratio of gph to volume? Wow.

I have a submersible box filter attached to a 350 gph pump. I also have a combined filter-pump unit that moves 325 gph. I also have a 1200 gph pump with a prefilter, and it's attached to a canister with a pool filter. The output is a fountain and an underwater current. I initially tried just the fountain and I had a geyser.

My pond volume is about 350 gallons. I have 11 goldfish between 2 and 4 in long.