Backyard Aquaponics
A place to discuss aquaponics

I think you'd be better off with a standard overflow standpipe rather than a siphon in the case of "cascading beds"...

Siphons can be notoriously unreliable if the flow is too high... or too low...

awesome advise i want it to be simple in its function and economical to run the grade of land and being able to sink the FT was trying for 1x pump .
Rupert would it work like OBO system run a spider to fill GB and standard overflow grading back to FT

Yep...

Any siphons downstream from the first will trigger without any great amount of calibration because the inflow from the dump of the previous GB will trigger it, and there are no problems with stopping, because there is no water flow at that stage in the cycle, so there is no option but to stop.

It becomes a non issue.

And the reason I suggest a siphon system was to save water.

If you cascade with a timer and siphons, you pump a little more than one GB of water to the top, switch off the pump, and then that one bed of water is all that's missing from the FT. That one bed of water, moves through the system. (or at least through that run)

By dividing the system into different runs, you can stagger them with different timers and smaller pumps to take only a minimum amount of water from the system. If you cascade standard overflow standpipes, in order to ensure the last bed gets flooded, you would need to keep the flow up to all the beds for a lot longer. Because the wave that cascades along the run doesnt have the flow stopper in the form of the siphons, the wave is a long slow drawn out one. If you tun off the pump after the first bed is full, the third and forth will be filling before the first has drained. This will mean the last bed wont ever see high tide. It will also mean a lot more water will be required to populate that "wave" that cascades through the beds.

I've done lots of experiments with siphons cascading into other siphons, and the result is always that the dump from a siphon makes a completely reliable trigger for the next, and the stop is inevitable because there is no water flowing at that point, so the bed drains and the siphon stops.

Calibrating the siphon at the top of a run is as simple as adjusting a tap. I don't ever need to do anything to my siphon and I've used the same one with pumps putting out very different flow rates (just over 50 lph to 3000 lph) because I just added a tap to control flow.

You could also cascade using siphons and a sequencer like Rup's flow splitter or my DIY sequencer. I think you can also use a device called a Flout, and there are many others.

Not necessarily... if your pump flow is too great into the first grow bed... the siphon might be overwealmed and not initiate at all....

Or, as you'd still be pumping (most siphon systems are constantly pumped)... they may not break...

Similarly... if the siphon in the first growbed doesn't initiate, or stop properly... the flow into the second may not be enough to initiate the second siphon etc...



Huh??.... it might... if you happen to set the timer to exactly the right time to initiate the first grow bed siphon... and if it drains enough to actually fill the second bed... and trigger the siphon.. etc...

More likely, you'd end up with at least one, or more beds partially filled to various degrees...

And how many timer cycles would be required to actually move the water from the first bed... through the rest... and back to the tank....



I don't know where to start with this... multiple timers and pumps.... what the??



Probably true... if you set the flow into the first bed using a ball valve...



Again possibly... with a flout... but the sequencing valve needs an interuption of flow from the pump to initiate....

It's just all so complicated... and I just don't see any benefit what so ever in using a timer with a siphon...

KISS...

Just use a timer and overflow standpipe... with the beds individually draining back to the tank... the drain will begin (slowly) as soon as the pump cycle begins... and during a typical 15 min cycle... a 500L grow bed (actually more like about 300l of volume adjusted for the media)... is filled well within the 15 minutes... and the majority of the water is delivered back to the tank via the overflow...

We're talking a 10000L tank... and about 3-4000L (max if all beds are full) of grow bed flood....

If the draw down is too great... due to the drain/flow being too slow returning to the tank... put in a sump tank with a float activated pump... to chuck the water back to the fish tank faster... ala the BYAP Family/Deluxe systems...

If all else fails... then use a sequencing valve... with the grow beds "banked/grouped"... and a timer/standpipe...

Simple.... and near faultless...

All I'm suggesting is to break the system into 2 or 3 parts and run a siphon at the top that needs calibrating like any normal, in use, tried and tested, every day siphon.

All other siphons down stream will not need calibrating and not fail to stop.

Simple and reliable.

If there is a need to sequence, a timer on each run and a smaller pump per run will be, by far, the cheapest and most reliable way to make sure that no two beds are being filled at the same time.

As a side effect, having more than one pump is ALWAYS a good idea and especially so on a big system where a pump failing might mean having to rent a truck to get rid of all the dead fish.

-The cost and running costs of 3 small pumps running for 15 minutes each compares favourably with the cost and running cost of one large pump.

-Timers are $20 for 3 from byap.

-Timers are an inexpensive and easy way to sequence.

-A siphon in each bed in a cascading system will trigger decisively from the one above and cannot fail to stop as there is no water flowing in at that point in the cycle.

-The reason I mentioned that "slightly more than one GB" of water would be needed is so that there is no issue with timing when the first siphon triggers. Running the pump for a bit longer than the first GB needs to fill and trigger the siphon, means you can dump a stack more water into the GB than the siphon would be able to cope with at the stopping stage. This means the siphon will start without fail even without calibration. Stopping is a non issue because the pump is off at that stage.

There's nothing complicated about it, but I'm going to stop posting here because I don't want to confuse anyone. Too many cooks and all that.

I'll start a new thread in the general section to discuss this approach, and I'll post a link to it here.

Rup know's what he is talking about and his advice is sound.

A pump failure wont kill your fish... unless you're relying on water return from the beds alone for aeration...

And I wouldn't... especially in a system of cascading beds... dependant on water return from siphons triggered by a timer to the first bed...

Oxygenate your tank with an airpump.... and your fish will survive any pump failure...

Bullwinkle, I hear what you're saying... and I've never run a "cascading" series of grow beds... but the experience of Kuda, and I think someone else that did so.... wasn't very successful...

And I'm not sure of the "cost savings" of multiple small pumps... to deliver water to x raised head in a cascading system... over and above a single pump... small pumps will often struggle to deliver a lot of flow above more than a metre head...

Nor would I generally suggest a timer... on a siphon based system... although I see where Bull is going in terms of a timed flood of the top bed(s)... to trigger a siphon, and subsequent downline beds/siphons...

But has anyone actually done it??... I'd like to see pictures..

I posted this here Bull for the benefit of the original poster...

The draw down of the fish tank seems to be the major consideration, and opposition you have to a timer/overflow standpipe system...

I don't believe the draw down would be significant...

The Bamarang system has 6 x BYAP 500L grow beds based on a timer/overflow standpipe configuration.. albeit with a sump tank and float activated Grundfos 10,000L pump to throw the water back to the 3000L fish tank...

And that never draws down more than 25-33% max....

With the immediate slow drain through the drain holes in the standpipe... and about 10 mins of the 15 min cycle in "overflow".... with the slope and gravity feed back to the fish tank in this proposal... and a 10,000L tank.... with most of the grow beds only half the size of the BYAP beds....

I don't see draw down to be a problem...

Totak GB = 16000 ish

thats quite a lot of grow bed.

Oops, I wasn't going to post here but my keyboard just went off when I was cleaning it

Nah.. you're way out Bull...

There's 4 x 500L (bottom of drawing) = 2000L
And 14 x 300L (top two banks) = 4200L

A total nominal capacity = 6200L

Fish Tank = 10,000L+


Compare that to the Bamarang system...

6 x 500L grow beds = 3000L
3000L fish tank...

Of course if you allow for the media... approx 60%... then the actual draw down on the tank is only about 3000L... if all beds were full... (and weren't draining... which of course the would be with an overflow standpipe)

Remember... the top two rows of grow beds are only 1x1x0.3 ...

The bottom rows of beds are 2x1x0.3...

that was a quote

Ok.. I just reread it.... I'd assumed a 300mm depth...

But see he's intending a full IBC.. ala OBO....

I wonder if he realises that OBO does not do a full drain of his IBC grow beds...

I knew there was something niggling me...

OK... on the basis of the above... then I have another concern...

With an entire IBC grow bed... you will need to lift the water from the previous bed up to potentially a metre to fill the next bed...

Don't know what the fall of the ground is... but will a siphoned drain be able to lift a metre head... and fill the next grow bed entirely to initiate the siphon.. and onward...

Or will the head... be too great... and/or if the gradient isn't sufficient.... wont the water equalise....