Backyard Aquaponics
A place to discuss aquaponics

Here is the reconfigured gravel shroud. There are now 8 (vs. 4) columns of 7 (vs. 5) holes perforating the sides. This is 6-inch pipe. The new bell siphon cover will be 3-inch pipe and the standpipe itself is still 1 1/4-inch pipe. I didn't have much time at all tonight so I'll work on the new bell siphon covers tomorrow.

You can do what you wish, but it should look more like a thick mesh screen than a pipe with a few holes. If you have a circular saw and want to save some time. Run about six slots top to bottom between your lines of holes.

I used the suggestion to go after the gravel shroud with a circular saw. I also created a new bell siphon cover (which I fought with for days until I realized my problem was the cheap cap I initially tried using). I then ran the siphon process several times with and without the gravel shroud, and with and without the cover lifted partway up to increase flow at the bottom. There is no appreciable drop in the siphon flow rate, not even after I add in the crushed brick growth media. I'm taking this as a good sign. Next, the pump.

A friend lent me a 450 gallon/minute (1700 liter/minute) sump pump. I initially thought I'd have to build some sort of bypass to bleed off the excess pressure but I learned the height (about 4 ft/122cm) makes a huge difference in the effective flow rate of a pump. Now I fear the pump isn't strong enough. I'll have to experiment with the placement of the pump and the length of the hose from the FT to the sprinkler pipe over the GBs.

Another slight complication: The pump is DC. I was loaned a marine battery but I don't have a solar or other charger. I'll see if I have a battery charger around but in the meantime I spliced in the converter for an old set of speakers (120v AC to 12v DC) and the pump seems to work fine. If I can find a proper battery charger I'll check into how safe it is to keep the battery on a permanent trickle charge. I'd rather do that as permanent solution so there's some backup when our power goes out, which is fairly frequently.

The pump works I tested under exaggerated circumstances (hose a foot higher, time a bit longer, overestimating how much water has to be moved, etc.) and it comes out to about 42 minutes to cycle all the water in the FT (normal time should be less). Now I need to work out how much water the total system will take (GBs, FT, and whatever's in transit). I'm also going to time how long it takes each GB to kick off the siphon process. I've seen different opinions on how important the siphon process time actually is so this is more for my own information in case it's determined to be relevant. Once I get this information down I'll employ another trick I've seen listed here and set up the GBs to siphon in alternating sequence. My understanding is that this helps keep a fairly constant oxygen level in the FT.

Here's the water flowing through the sprinkler/irrigation pipe. I had to block off the holes on the far edges of the pipe and those over the bell siphons in order to maximize the amount of water landing on the grow media. I'll have to stabilize the irrigation pipe and the tube from the pump as nothing is fixed yet and things tend to slide around.

Something else I'm noticing is that the water flow varies considerably depending on the angle the tube enters the irrigation pipe. I'm thinking of adding some type of diffuser to sit between the tube and the pipe to help even the flow, possibly one of those kitchen sink inserts to prevent large food particles from going down the drain.

In the background, the Lord of Destruction is chasing some lizards. If that keeps him sufficiently occupied so he doesn't start chewing on the AP system I'll put out lizard bait all around our yard.

It's good to minimize splashing to help prevent waterborne disease problems so you'll probably want that down at the grow bed level.

Cheers

I spent much of today trying to get the siphon process working again now that the irrigation pipe is providing the water. The pump just isn't powerful enough for both GBs (either one works fine when used alone). Maybe the combination of moving the pipe down and cutting off some of the hose will result in sufficient water flow to restore this functionality.

Worth a shot

Also you may want to look into replacing that galvanized tub. It will leach zinc into your system.

The galvanized tub was a temporary addition used only for the initial testing (the water in the FT was quite dirty at that point) and is now gone, but I wasn't aware of the zinc issue. Thanks -- I'll definitely keep that in mind.

Reducing the vertical distance between the FT and GBs didn't resolve the siphon process. I even added a reducer to the top of the standpipe and the flow is still too slow, even if I tilt the irrigation pipe so that water only comes out on that single GB.

I've tried adding reducers to the bottom of the drain with mixed results. The siphon process will finally kick off but the outflow is too slow to successfully drain the GB in a timely manner.

Unless anyone else has any ideas then getting a bigger pump is the best option I can come up with at this point. I'll work on that over the next few days.

I'm not 100% sure of what you're trying to do but you could ditch the siphon and just run Constant Flood. Basically running the pump 24/7. This way you don't have to worry about having enough flow to trigger the siphon. I've been running CF for several years and others have as well. Check out the BYAP Trials thread for a comparison of three different types of systems over a year long period. Your goal should be to turn over the volume of the fish tank once per hour. If you're close that's probably good enough but if you're not then you need a bigger pump.

Jaymen, nice work on the gravel guard.

In your photos, are the tops of your bells above the level of your gravel? I have found that the top of the bell needs to be the maximum height of your water level in your GB - usually 1/2-1" below the gravel surface, and the standpipe needs to be a further 1/2 to 1" below the top of the bell in order for the siphon to kick in. The GB slowly fills until the water starts to over top the standpipe (under the bell) and continues to rise to the top of the bell which eliminates the air and the siphon kicks in. The GB then empties at a fast rate until the water level drops to the notches in the bottom of the bell and it starts slurping air until it breaks the siphon. If the siphon doesn't kick in, the flow is too slow (it just continues to overtop the standpipe), and if it doesn't break, the flow is too fast, leaving your GB empty.

I am guessing at your system setup... I gather the barrels are ~50 gallons or 200 litres? If so, a 1/2 barrel would contain maximum 100 litres of media & only ~40 litres of water, so should only take several minutes to fill up at a reasonable flow. You want to turn over the volume of your fishtank each hour, so if your fishtank is 200 litres, and you are splitting your flow between two 1/2 barrel GBs, you would be aiming to fill them nearly 3 times an hour (3x40x2=240 litres/hour which is only around 60 gallons/hour). Be careful with sump pumps, I have read that they can use a lot of electricity and are not designed to run 24/7.

You need to deliver the water at the gravel surface to keep the gravel dry to prevent algal growth which can clag up the gravel around the outlet. You also want taps or valves on each GB so you can adjust the flow. That is, pump to a single larger pipe with individual controllable outlets to your GBs. Jayendra uses elbows on the ends of his outlets to his GBs which he can twist up or down to regulate the flow (cheaper than taps). I think there are photos on his Yarragon Cafe thread.

scotty435, I've been leery of CF because of the prevalence of mold here in south Louisiana. Any time there is standing water in a plant pot the plant itself ends up rotting away from the roots up, even with those plants that grow naturally in such an environment. This may very well be different in an AP system, especially with the running water. I think the concern about mold is just so ingrained that I didn't even objectively consider it.

I have access to another pump. I'm going to try that out first since I already have everything set up for e&f. If not then I may give CF a shot.

Thanks, joc. The bell covers and standpipes are both too high. I will correct this. I did confirm that the flow is insufficient by pumping water directly into the GB and observing the slow flow without the bell cover on. I have to conclude that the pump is simply insufficient and may in fact be failing.

Good guess on the capacities. I measured the pump outflow and figured the FT water was getting completely cycled once every 42 minutes, so it works for the FT but I still need faster movement for the siphon. Luckily, I have access to a replacement pump. I'll be able to install it within the next few days.

I built the irrigation pipe with traditional dirt farming concepts stuck in my head, so it isn't optimal for AP. I can minimize splashing, keep the gravel surface layer dryer, and recover a lot of grow area by cutting the arms shorter and either drilling larger holes or just removing the caps. I've already lowered it to the surface of the GBs. If it looks like flow control is needed then I'll try Jayendra's elbow solution.

This picture is from many revisions ago (no grow media, pump, or bell siphon) but it shows the basic layout of the FT, irrigation pipe, and GBs, just in case this helps explain anything about my configuration.