So I’ve gone about constructing myself a pair of under-gravel, or in my case, under-sand jets that Marc Elieson had originally come up with. Constructing Under-Gravel Jets. My design differed than the one Marc described in his article. Instead of having two different networks, each powered by its own powerhead, I opted to have two powerheads running one system of eight jets total. This picture shows how the network ended up looking.
The idea of this design was to stir up most of the waste around the perimeter and corners into the middle “lanes” of the tank. From there, the four middle jets would provide most of the necessary propulsion to slowly push debris towards my filter intakes which reside in between the two “lanes” of the tank that this design had created.
Originally I went with my two Rena Filstar filters, an XP2 and an XP3, to power the jets but the flow rate was only 300 and 350 GPH respectively and as many of you know, this isn’t enough for 8 jets. The general rule of thumb with UGJs is to have 125-200 GPH per jet and I was only giving around 75-80 GPH.
Long story short, I ordered some powerheads that would give me a higher flow rate, two CAP 3600s to be exact; each of which would give me 947 GPH and a bit over 200 GPH per jet. I factored in my rather maze like design which ultimately slows the flow rate to a nice amount just strong enough to move debris but not strong enough to annoy the fish.
Once I finally rigged up my powerheads I ran into a problem I’m sure many aquarists have ran into – poor UGJ design. The flow from the powerheads was traveling mostly out the first two jets on my network since this was a straight line from the flow source; this happened despite me having pinched the piping for these jets to avoid this problem. With me being a bit of a stickler and innovative I decided I was going to make what I had designed work, one-way or another.
I started thinking of ways to direct the water towards the middle 6 jets without clogging the pipes or slowing the flow rate. I thought of putting a sponge in there to hopefully direct the water where I wanted it to go, but thought eventually those would get plugged up with gunk and really affect my system.
Finally I dreamed up an idea of cutting into the fittings I had and inserting something that would be solid enough direct the flow and stay in place without hindering another jets flow rate. After thinking about what kind of household items would do the job I came up with this mix of tools and hardware. I decided to use a chip dip lid as the device that would scoop the water into the pipe I wanted it to go. I decided to call these inserts flow scoops.
The design ended up working perfectly. You can cut the inserts to scoop precisely how much water you want and as you can see by the angle cut, it perfectly scoops water downward without losing much momentum, which otherwise could result in slowing your systems flow rate. In the first picture below the flow source would be coming from the right to left and the scoop would be helping it to go down rather than continue on its straight path.
By using this design my two powerhead, single network UGJ system works flawlessly as I now have scoops in four areas that just so happen to be on all my “T” fittings. Any “plus” fittings which split the pipes four ways could also use this method.
The result was quite shocking and it was satisfying seeing the debris on my tank’s substrate literally lift up right in front of my eyes and stay up until it made its way to my filter intakes. At last, a job well done.
Ultimately I highly recommend this improvement to any system that isn’t working well. It’s very easy to be accurate with your flow rate out of each jet by using this method. Say you want half of your flow rate to be directed one way, just make sure that the scoop goes half way through the pipe and you’ll be scooping precisely half the water. You don’t ever have to cut the pipe more than half way, just shape the scoop differently. Enjoy!