Backyard Aquaponics
A place to discuss aquaponics

The first step is to decide on how much flow you want.

General rule of thumb is that you want to turn the FT over at least once per hour. In this example we have an IBC as the FT so that means we want a flow of 1m3 or more per hour.

Second if water is moving through the pipe slowly then there won't be much friction. So if we chose a pipe size that is large then we can assume there is no friction. How large is large? A pipe is large if the water moving through it is slow.

Second we set a target water speed or velocity of less than 1m/s

V=Q/A

V=velocity (m/s)
Q=Flow (m3/s)
A=pipe cross sectional area (m2)

In this example V = 1m/s and Q = 1m3/hr=0.00028m3/s

0.5=0.00028/A
0.5A=0.00028
A=0.00028/1
A=0.00028

A=(3.1416D^2)/4
0.00028=0.7854D^2
D^2=0.00028/0.7854
D=sqrt(0.00036)
D=0.019m
D=19mm

Based on this we chose a pipe that has an internal diameter of 19mm or larger which is most likely going to be 25mm.

You may be tempted to go for 20mm since it is so close to 25mm but the assumption we made at the top was that we could ignore friction because the water was moving slowly. The internal diameter of 25mm LDPE is about 22mm. Where as that of 20mm is about 17mm. Going back the other way gives us a velocity of 1.2m/s which is 20% above our assumed value of 1m/s. How much of a difference this would make can be worked out but it is a bit more complicated.

Having done the calculations and assuming the pipe is 2.5m long with one entrance, expansion, three 90 bends and one exit the friction losses add 168mm to the dynamic head using 25mm PE or 550mm using 20mm PE. Whether this is a big deal or not depends on your choice of pump.



This is my Laguna 11000. Note the short (could have been a bit shorter 32mm pipe) which goes from the pump into a threaded barb, into a 32mm to 50mm bushing, into a 50mm female threaded PVC spigot fitting, into a 50mm to 100mm glued bushing, into a 100mm 90 degree bend which goes out of the sump into the FT. Depending on how high the water level is in the sump this means that the water is flowing around 7-8000L/hr which means the pipe velocity is only ~0.25m/s. A tad excessive but when i built it I didn't understand pipe flows as well as I do now and I already had some 100mm pipe and fittings.

When I went from one 32mm pipe going from the pump and curving into the tank to the setup in the picture the flow increased so much that the water level in the FT rose 150mm and the FT started to over flow despite the two 80mm drains. That caused a HSM and a redesign of the drains.

More explanation: less simple but some good guidelines
Friction loss in pipes is highly dependent on the:

1. length of pipe - shorter the better
2. number of fittings - the less the better
3. type of fittings - gentle curves, avoid junctions
4. velocity or speed at which the water is moving through the pipe - slow is better.

1. Shorter is better because of less friction and less cost but allowance needs to be made for getting around your system. Having a pipe at waist height across a walkway may be shorter but it would also be annoying. Best design would of course not have a pipe crossing a walkway in the first place.

2. every time the water goes through a fitting there is more drag or friction. Changing pipe sizes, direction, valves (even when fully open), junctions even exits (ie the end of the pipe) and pipe joins all create friction. Some more than others.

3. Every part of a pipe line is going to create friction. Choose fittings and combinations of fittings that reduce the friction. Junctions are particuarly bad a creating friction even if the water is just flowing straight through. The Junction used in many SLOs is done with good reason but when a junction is used to make a 90 degree turn like this it creates a lot more friction than using a 90 degree bend fitting. Valves even when fully open still create friction so if you don't need a valve don't have one on the line. If using PVC pressure pipe the bend fittings are very tight 90 degree turns so it is better (less friction) to make a 90 degree turn with two 45 degree bend fittings (but costs more in plumbing fittings). In DWV pvc pipes the bend fittings have a radius of 1 pipe diameter which is pretty close to ideal to reduce friction in a bend (which is why they are made that way). Hence making a 90 degree turn with two 45 degree fittings in DWV actually creates more friction loss than one DWV 90 degree bend fitting.

Lastly many people like to add venturis to their FT inlets to increase the aeration of the FT water. There are perfectly valid reasons for doing this but they massively add to the friction losses in the inlet pipes and there are other methods that work better. Having said that a venturi may be the best aeration option for a particular system or set of circumstances. If you are going to add a venturi then you can't use this methodology to chose your pump.

4. Basically big pipes make for slower speeds so the bigger the pipe the less the friction restricts the flow. Ideally a pipe would be so big that the velocity of the water moving through it approaches 0. In the real rather than mathematical world your can do a calculation on how much does a pipe size cost to buy and install versus operate. Small pipes are cheap to buy but expensive to operate where as large pipes are expensive to buy but cheaper to operate. Working out what is best depends on you, your circumstances and a more complicated calculation than is appropriate for this post.

xtutex

Such a good post. Thanks Stuart.

For people who want to know how to do the calculations them selves see this thread:

http://www.backyardaquaponics.com/forum/viewtopic.php?f=14&t=21099&p=445935#p445935

Mr. Chignell, great post... but I can't figure out how to do the math myself...

I'm a long time browser here, but this is my first post. I have a question specific to my pump-to-FT pipe size. I couldn't figure out where to ask about it, is here ok?

Haha Found it.

Here is fine.

Can I ask what the difficulty is?

If you are having problems then odds are that other people will be as well. If we could work through your solution together then that will help you and others as well.

That was a great post Stuart!

Do you think there are even some more general piping rules of thumb like:
"CHIFT PIST 1000L FT with SLO should have a minimum of 40mm pipe"

I know all the fittings and length of pipe affect it but some general sizes might be good.. at least everyone could just oversize it one size and be pretty sure it would work.

I know I just guessed when I used my 40mm pipe for above sized system… luckily it just manages the once an hour turnover rule… which surprised me.. had I known I would have gone 50mm even though the fittings are pricier.

Over sizing your drain or SLO can be a really bad idea.

http://www.backyardaquaponics.com/forum/viewtopic.php?f=14&t=21129&hilit=xtutex

General rule of thumb don't worry about oversizing you pump pipe its impossible as long as you can afford it. Be more careful about your drains.

The things is that that general rule of thumb will fail people completely if there pump is in their FT.

Alright.
I've got a 5,298 Lph pump in an IBC tote sump tank, my fish tank is also an IBC. I understand the bigger the pipe is (that's attached to the pump), the less friction loss. Using your formula V=Q/A, I get a pipe cross sectional area of 43mm, rounded to 50mm. I'm wondering if that's right.
Also, how big is too big? With my 2.5 meter vertical head, wouldn't the weight of the water in bigger and bigger pipe cause alot of back pressure on the pump, causing it to work harder?

Thanks Stuart!

Presumably this is the flow at 0 head (maximum) can you find a pump curve (graph showing the flow from the pump as the head (height) increases)) for your pump either in its manual, on it box or on the internet?



I would not use 50mm as an approximation for 43mm. Pipe manufactures often give pipe diameters withan a tolerance of +- 1mm so I'd use 43mm. Aslo while 43mm and 50mm are ~14% different their areas are 26% different. Introducing a 25% error into your calculations before you begin is not a good idea.



Think of it another way. Say you have two pairs of massive water tanks very wide and very, very deep and one of the pair is slightly lower than the other. Each pair is connected by a pipe at the bottom but one pipe is a massively fat pipe and the other is quite small. Each pair is filled until the slightly lower tank of each pair overflows.

If you poor water into the higher tank of each pair on which pair is more gravity (force) required to get the water going into the higher tank to displace water in the lower tank (via the respective pipe) and cause it to over flow?

Believe it or not the answer is the pair of tanks with the smaller pipe. Gravity is just a force and just like a pump imparts a force of water and gets it to move so does gravity. In the pair of tanks where the pipe is massive water poured into the higher tank pretty much immediately displaces water in the lower tank. The water velocity in the massive pipe is essentially 0 so consequently the friction is also 0.

In the tank with the small pipe the water velocity is greater which means the friction is greater and this creates a force that works against the water moving through the small pipe. Since every action must have an equal and opposite reaction in order to get the water to move through the small pipe more force is needed which is less case is supplied by gravity. The only way for gravity to supply more force is for the water level in the higher tank to get deeper.

If the pipe though smaller is still relatively large relative to the amount of water being added to the higher tank then the water level in the higher tank will almost remain the same (only slightly higher) but as the size of the small pipe decreases relative to the flow the water depth in the higher tank will increase.

For example say you have two IBCs connected by a garden hose. Even with a bucket you will be able to move enough water into the higher IBC so that you will see a measureable water difference between the two. If the pipe connecting them is say 50mm then a bucket will not do but a modest pump (5000L/hr) will add enough water so that you will be able to measure a height difference between the two tanks. If you increase the size of the flow (pump) into the higher IBC at some point the higher IBC will overflow because the pipe (drain ) is not big enough. If you then increase the pipe to 100mm then you will need a much larger pump in order to get a measurable water level difference between the two IBCs and a serious pump to get the higher IBC to overflow.

Hopefully from these examples you can see that rather than needing a bigger pump for a bigger pipe you need a bigger pump for a smaller pipe.

OK, so the closest thing to a 43mm pipe for me is 1-3/4 Inch (44.45mm).

EDIT: Correction, my options are 1-1/2" (38mm) or 2" (50mm). They don't carry 1-3/4" (44.45mm).

Yep, makes sense.
But, in theory, is it at all possible that one can oversize the pipe to the extent that the pump can't handle the vertical head pressure?
For example: my pump says it maxes out at 11.65 ft (3.55 meters). If I were to use a 150mm pipe (in theory only), would the pump max out before 3.55 meters?

Simple answer:

No.

If you were to connect your pump to a really large vertical "pipe" or in other words a circular tank then the water level in the tank above the level where you where pumping from would be slightly less than 3.55m (because of friction losses through the pump, small pipe connecting to the tank and the enlargement/exit of the small pipe into the large vertical pipe/tank.

The important dimension is the internal diameter. High pressure pipe has much thicker walls than low pressure pipe which means that the effective diameter of the low pressure pipe is significantly bigger than the high pressure pipe.

You might not think it makes much difference but since the area of a circle is proportional to the square of the diameter and the friction losses are proportional to the square of the velocity an error in the figure that you use for your internal diameter compounds 4 times.

Manufacturers often have their pipe outside diameters and inside diameters on their website.

How many watts is you pump?

On pump IDs I can probably give them to you because the standard sizes are pretty widely used across international boundaries. I'd need to know the type of pipe (DWV, SW or Pressure (and that class)).