Will Solids Removal Still be Necessary In DWC with a Media Bed?

[Gina Cavaliero]

The ultimate goal in integrating a media bed gravel filter with a DWC system is to eliminate the need for solids removal, essentially eliminating the labor factor as well as benefiting from the additional metabolization of those solids by introducing red worms.  Is it possible to remove all solids via the media bed or will some degree of solids removal, ie. settling tanks or swirl filters still be needed?  JD Sawyer just posed this question to me a couple days ago and quoted this from James Ebeling's book Recirculating Aquaculture,

"Carbon eating heterotrophic bacteria grow significantly faster than the autotrophic nitrifiers do. Their mass can double in an hour, while it takes nitrifiers days to double. This high growth and the associated oxygen demand consequently suffocate the nitrifiers buried deeper in the biofilms, resulting in death and sloughing of the biofilm from the bioreactor surfaces... You must have effective solids removal BEFORE the high ammonia water is transferred to the biofilter. As mentioned above, the heterotrophic growth will compromise the nitrifiers ability to oxidize ammonia, mostly because the heterotrophs consume the oxygen prior to the oxygen being able to diffuse into the biofilm to where the nitrifiers are. This results in nitrifiers being starved for oxygen and then they die off resulting in complete sloughing of the biofilm and loss of nitrification capacity. A recipe for disaster in RAS is to have poor solids removal"  (Ebeling, Biofilter design, Recirculating Aquaculture 2nd Ed).

My theory is that stocking densities and feed ratios will have  a tremendous effect on this and if managed correctly, should minimize the need for solids removal.

[TCLynx]

Yes, it is all a matter of scale/density as to how well a grow bed of a particular size will manage to filter and metabolize solids without compromising nitrification.  Luckily nitrification actually takes far less space to do it's job than the space required for solids.  This is probably why so many people get away with under filtered/over stocked systems short term, it is only later that the grow bed gets overwhelmed with solids and causes disasters.  Anyway, it is likely that the whole 2:1 ratio of grow bed to fish tank largely comes from people not wanting to ever need to clean out a grow bed because of solids.

 

Anyway, We know that backyard scale aquaponics without solids removal does work quite well.  One must define one's goals with a system when figuring the design because if one wants to compete on fish production with RAS on the same footprint, then YES Solids removal will be required.  Aquaponics takes quite a lot of media bed space for solids and plants so there is no way for an aquaponics facility to compete on fish production in a foot for foot basis with RAS and since RAS doesn't want to devote huge amounts of space to filtration, they will need to remove solids before the water goes to the bio-filter.  Remember, their goal is to raise as many fish as possible in the least amount of time for the least amount of money.  Trying to be water wise or retain the nutrients are not part of their goals.

 

BYAP did some trials of a commercial type system a few years back.  They ran into problems with tree leaves getting into their rafts which caused them to shut that system down since they didn't want to enclose the area at that time.  Anyway, they were using heavily aerated bio-digesters to break down the solids.

 

There are so many ways to do things, it is interesting to see what seems to work best for different situations.

[Sahib Punjabi]

Well said TCLynx

TCLynx said:

Yes, it is all a matter of scale/density as to how well a grow bed of a particular size will manage to filter and metabolize solids without compromising nitrification.  Luckily nitrification actually takes far less space to do it's job than the space required for solids.  This is probably why so many people get away with under filtered/over stocked systems short term, it is only later that the grow bed gets overwhelmed with solids and causes disasters.  Anyway, it is likely that the whole 2:1 ratio of grow bed to fish tank largely comes from people not wanting to ever need to clean out a grow bed because of solids.

 

Anyway, We know that backyard scale aquaponics without solids removal does work quite well.  One must define one's goals with a system when figuring the design because if one wants to compete on fish production with RAS on the same footprint, then YES Solids removal will be required.  Aquaponics takes quite a lot of media bed space for solids and plants so there is no way for an aquaponics facility to compete on fish production in a foot for foot basis with RAS and since RAS doesn't want to devote huge amounts of space to filtration, they will need to remove solids before the water goes to the bio-filter.  Remember, their goal is to raise as many fish as possible in the least amount of time for the least amount of money.  Trying to be water wise or retain the nutrients are not part of their goals.

 

BYAP did some trials of a commercial type system a few years back.  They ran into problems with tree leaves getting into their rafts which caused them to shut that system down since they didn't want to enclose the area at that time.  Anyway, they were using heavily aerated bio-digesters to break down the solids.

 

There are so many ways to do things, it is interesting to see what seems to work best for different situations.

[Chris Smith]

Many months ago I made a change to my raft system to improve efficiency and made an important discovery in the process. I origionally built the system using the Friendly design with two pairs of troughs.  I was pumping each pair of troughs 5-6 gallons/minute of water. I had no readable nutrient drop from end to end in the troughs so  I decided to connect troughs 2@3 together with siphon making the 4 troughs a daisy chain. Once I daisy chained the troughs I was able to cut my pumping capasity in half saving me 50% on pumping power. I still have no nutrient drop from the begining of trough 1 to the end of trough 4.

My interesting discovery was noticed once my pumping was cut in half. With less water pumping through the fish tank the solids have more retention time to be broken down by the aeration system for the fish. The longer the retention time in the FT the smaller the solids coming out. Once I made the change my clarifier began collecting far less solids than before and I had less fines in the swirl filter as well.

I am now experimenting with pumping some water from trough 4 directly to trough 1 and sending a little less water through FT. It is more efficient to move water from trough to trough than up to the high point at the FT. With no readable nutrient drop from end to end I do not believe it is necessary to move ALL the water through the FT. I think that it may be more efficient to keep the bulk of the water looping in the troughs and adding a smaller amount if nutrient rich water from the FT to the troughs to keep nutrients up. I hope to find a good balance of moving just enough water through the FT to keep the ammonia down, keep good flow throughout the system and save some power(our power costs about $.44kwhr).

[Gina Cavaliero]

My point exactly TC, scale is paramount.  While we have valuable info that we can correlate from backyard and small systems, it still is not the same when comparing to large systems and larger densities.  Backyard scale media systems without solids removal does work well however this discussion is geared towards large scale and that is what I was intending when I started it.  Although I did not specify that in the discussion post, I thought it was inferred in that it is under the group for Green Acre where our focus is large scale growing.  That is why I thought it was important to point out that while small scale backyard or research systems can generate excellent data for their relevant size, it may not be appropriate or accurate to quantify the same data at a considerably larger scale.   

Kobus has great, specific examples regarding particle suspension and variations due to water movement in large scale as compared to small scale, not to mention the sheer volume of water, which if we consider the GAO system we are talking nearly 20,000 gallons.  This volume then translates differently for fluctuations in pH and other levels.  For instance something that may present itself quite quickly in a comparable small system with 800 gallons may take much longer to present itself, if at all, in the 20,000 gallon version.  This explains why nitrite spikes at start up that may be very apparent in a large scale system sometimes will go completely unnoticed and immeasurable in a small one.

[Sahib Punjabi]

Noted.

 

Thank you for the clarification that is / was not in your discussion "however this discussion is geared towards large scale and that is what I was intending when I started it."

 

I wonder how many "large scale growers" are using or have such mixed systems to be able to answer such?

 

God bless

 

Gina Cavaliero said:

My point exactly TC, scale is paramount.  While we have valuable info that we can correlate from backyard and small systems, it still is not the same when comparing to large systems and larger densities.  Backyard scale media systems without solids removal does work well however this discussion is geared towards large scale and that is what I was intending when I started it.  Although I did not specify that in the discussion post, I thought it was inferred in that it is under the group for Green Acre where our focus is large scale growing.  That is why I thought it was important to point out that while small scale backyard or research systems can generate excellent data for their relevant size, it may not be appropriate or accurate to quantify the same data at a considerably larger scale.   Thanks.

[Gina Cavaliero]

If we refer back to Chris' first post, he is already experimenting with the integration on a larger scale.  Although I think his media beds are a bit small in comparison to what we are doing, but the overall system still probably has comparable volume. I think Murray may know of some larger systems Sahib.  

 

Chris, your observations are interesting from when you changed your flow and its effects on your solids. Our rafts too have one single continuous flow and no measurable difference in nutrient levels are observable from start to end.  I was just discussing with Murray last night what would be a dual loop design and that it is not essential for all of the water to travel through out the system but that looping independent sections back to a centralized location, ie. the fish tank would have the same result, effectively increasing nutrient availability through out the system but also minimizing pumping and electrical consumption and not creating issues with large surges when dealing with a 200sqft (potentially 400sqft) media beds.  

However, its a bit more challenging for you in not only do you have considerably higher KWH but elevation and head heights as a factor for pumping considerations.  We in comparison have cheap electric and have to literally build our elevation to create any kind of gravity flow but as a result have much smaller, less consuming pumps for returns.  We are in the flat lands!  

[Nigel Clement]

This for me is one of the most timely and relevant discussions, given that I'm attempting to incorporate media beds with my DWC system. Thank you folks for all of your liberal and candid contributions.

[Kobus Jooste]

Nigel - I also enjoy it as I am busy trying to salvage a large system design that I started a few years ago but which ran out of momentum when the current business partners I have pulled out.  It was fortunate in a way, as I had time to study my small mixed design system and make some important observations.  The one that I think will bug large scale mixed systems the most is water velocity.  I'm not so sure what the average person in this forum has in terms of experience with large flood and drain, but it is an impressive sight.  With enough of a head, you can wash gravel with such a system (as I ended up doing during tests).  A single line of gravel beds will not stop the solids.  You should also not have too much of a drop between your components because of the velocities you create.

 

 

These beds are 7.5 meters long, about 1.5 meters wide and just under a meter deep.  When those siphons let go, the outflow almost fills a 110 mm pipe.  Nothing but the largest solids will stay behind.

 

My modification: Run clarifiers on fish tanks.  Send clarifier solids via timed pumps to the first bank of gravel beds.  Clarifier outflow and the water that drain back out of the first bank of gravel beds goes to a second bank of gravel beds.  Between the second gravel beds and the DWC, I am going to put a fine net tank or similar to try to trap the smallest fines.  I will steer away from siphons between beds.  They are just too powerful in this design.  Simply use a perforated stand pipe design and use small level changes to "push" the water along from the gravel beds through to the DWC.  Any serious drainage flow and you will likely end up with sediments / solids in the DWC.

[Sylvia Bernstein]

Really interesting what you are doing, Kobus, and your observations about redirecting the solids and the velocity of the water.  I've come to really like the power of that sucking action to pull oxygen into my media beds, however, and I'm afraid of what would happen to their growth potential if I wasn't aerating as much.  What do you see as the main problem with having the higher velocity water flow in this design?  Is it washing the fines off the nets?  Too much for the plant roots in the nets?  What are your observations?

[Rob Nash]

in my hybrid system that has been up for a year now, i have found the regular ratios apply to the media bed and the the rafts. i go with water volume (1000 gallons) times .266 = 260 sft of gravel (i use 250sft) and for raft i used the UVI standard of 50 lbs of fish for every 86sft of raft. which in my case, at 400 lbs of fish i could add 688 sft of raft. i only added 288 sft to keep the nutes nice and high, and because i noticed the uptake of the additional plants in the raft system required more frequent top offs in the fish tank. (300 gallons/week instead of 100/week)

i have worms in the gravel and they do a great job, and with a little stirring when planting, the gravel does not clog. and then an orchard net filter between the gravel and the rafts, just to "polish" the water and keep the fines out of the DWC.

keep it simple Gina.

[TCLynx]

Velocity of water is something I've also interacted with a bit.  (Seeing as a fairly high flow and pressure is required to operate indexing valves.)  anyway, The Fast drain by siphon from a big grow bed may provide some nice aeration to the water it splashes into as well as drawing air down into a grow bed fairly quickly (but fast or slow drain the air still comes down into a grow bed so I don't think it makes much difference there to aeration.)

But I have noticed that the siphon drain will draw more fine solids back out of a grow bed than the slow drain of a timed flood and drain operation.  You can actually see this to an extent in a timed flood and drain if you pull the stand pipe out when a bed is fully flooded you will see some fines or cloudiness in the water coming from the grow bed with the fast gush of water.

 

I have also found that you can only go so far with fast flooding of a grow bed,  Like Kobus said, if he flooded from a siphon in one of those beds to the next one He would be washing the gravel, perhaps right out of the bed in places I bet.  A "flush" tank sort of means to flood beds really fast is only useful up to the point of water being able to flow down into the media.  When the gush is so fast that it is like a flash flood washing media and plants along with it, well that doesn't work so well.

 

The following example was being run with a very heavy fish load for a backyard system so might be of use to this discussion.

I've done a nearly 33 foot long by 3 foot wide by over a foot deep grow bed before.  With constant inflow, the water inlet end was pretty much always flooded and only the end with the float controlled pump was the part that really saw true flood and drain.  The solids did build up in the inlet end of that bed. 

Distribution of water into large trough like gravel beds Is probably worth the extra expense in plumbing and also the extra maintenance of cleaning that plumbing occasionally.

[Kobus Jooste]

Sylvia, as TCLynx said, the outflow of the siphon from the raised gravel beds in the filter was so strong that it would suck most of the smaller solids and fines clear out of the beds.  We washed the gravel for these beds by filling the beds with dirty gravel, filling one of the reservoirs and setting the pump into action.  With a fine filter in the canester filter we had installed directly after the pump, only one flood cycle sucked enough fines out of the dirty gravel to clog up the canister filter.  After taking it out and operating the siphons for a while, we had relatively clean gravel up top and extremely dirty water in the reservoir.  Without a serious baffle system, the first few rows of plants in a raft fed directly by a siphon such as this one will be blown out.  The net filter will also be worthless as everything will get re-suspended.  Siphons will still be potentially usefull, but then they will have to be a lot smaller.  We had only one siphon draining the bed (the unit was devided into 4, thus 1 bed was 3.5 m long and 1.5 m wide). Perhaps if you had 2 - 4 you could reduce the velocity sufficiently.  As it is now (the greenhouse was never completed, but will be moved soon, where-after I will make some changes before hopefully completing it), the siphon is too strong.

Sylvia Bernstein said:

Really interesting what you are doing, Kobus, and your observations about redirecting the solids and the velocity of the water.  I've come to really like the power of that sucking action to pull oxygen into my media beds, however, and I'm afraid of what would happen to their growth potential if I wasn't aerating as much.  What do you see as the main problem with having the higher velocity water flow in this design?  Is it washing the fines off the nets?  Too much for the plant roots in the nets?  What are your observations?