Poorly plumbed rainwater tank

Full admission, I am making this post in hopes that SaveH2O will see it. I have been reading through this forum have come across many of his helpful posts on the subject of rainwater tanks and plumbing.







My new build journey has finally come to an end and unfortunately, I have since identified some major issues with the rainwater tank installation (which was done 1 day before handover!).

The pump is drawing water from the bottom of the tank which is the worst quality water. The bottom hole should have a drain valve installed and instead the hole above it be used for the pump.

The inlet pipe has no mosquito mesh. The overflow outlet is too high and may not have enough head for heavy rainfalls.

I am in the inner suburbs of Melbourne and if I calculated correctly from BOM (AEP 5% @ 5mins = 10.9) my 1:20 ARI will be 130mm/hr.


There is approximately 890mm of head from the bottom of the quad gutter to the inlet pipe's invert. ALL downpipes from the house (except for the front) are charged to the tank in order to achieve the required fall to the kerb's legal point of discharge. Total roof area is approximately 300sqm with a 22 degree pitch roof. Pipes used appear to be 100mm DWV.



With the issues I have already identified, is there enough head to fit leaf diverters to the downpipes? Ultimately, I want to bring this system up to scratch and ensure it is able to with stand a heavy downpour.

Unfortunately, what you have identified are substandard practices that are nevertheless still compliant.

You have a good grasp on industry terms and what constitures good practice, probably more so than the majority of plumbers and installers. This will help a lot.

This is correct. The 130 mm/h is based on an average rain intensity of 10.9 mm over a 5 minute duration. This equates to an average of 2.1666 mm/min. It is very important to understand that this is the minimum average intensity over a 5 minute period to qualify as a 1:20 ARI. The next higher intensity is a 1:50 ARI and above that a 1:100 ARI which has a minimum qualifying intensity of about 50% higher than the minimum 1:20 ARI intensity. This is why I advise people to calculate a minimum 20% safety margin into the hydraulic calculations to allow for proper gutter slope and positioning...which is uncommon.

The IO on the riser certainly looks like 100mm DWV.

Water doesn't flow paper thin from the top of the vertical riser, you need to substract 50mm from the head.

Because the pipe drains the other 7 downpipes from both sides of the house, we needn't worry about the roof slope and wind driven rain in this instance.

We will just surmise equal roof areas draining to each downpipe.

Minimum intensity 1:20 ARI:
9 DPs draining 300 sq m roof plan area = average roof area of 33.33,
7 DPs draining to the tank = 233.3 sq m.
233.3 x 2.1666 = 505 lpm. V = 0.99 metres/sec.

I would design for 600 lpm.

A thing to remember about downpipes is that each one acts as a water tower to repressurise the carrier pipe which has friction losses along its length. The 2 DPs closest to the tank are close together and also close to the tank and so the ability of the pipe to drain to the tank is not an issue. One issue however is that the 100mm pipe has been used along the entire length and this results in a gradual build up of silt and other muck at the head of the system due to the insufficient flushing velocity compounded by Melbourne's rainfall being about 90% drizzle and light showers.

Correct but still compliant. Easily fixed.

The inflow pipe is also dumping concentrated water through a meshed inlet that is above the valve that supplies the pump. Compliant - Yes. Stupid - most certanly! The tank's top meshed inlet used should have been the one at the other end.

You will also notice that the IO on the vertical riser is very high and also facing the pump! IOs used to drain subsurface pipes are not efficient.

Remember the 0.99 metres per second velocity I calculated during a minimum 1:20 ARI? The inflow from that pipe during any rain intensity approaching a 1:20 ARI would generate a lot of splash.The overhang is too great.

It is best practice to fit a flap valve. The bend used is the wrong one, a flap valve fitted to that elbow would remain open.



If you ask the plumber who installed the tank what the tank's overflow capacity is, he won't know nor will the salesman or the manufacturer!

The pipe is 104mm but the flashing on the mesh perimeter will close that to about 100mm. AS/NZS 3500.1 has a chart that conveniently shows the overflow capacity of different size horizontal pipes with different water heights above the invert. A height of 100mm above the invert is obtainable (4.67 L/sec) but a height of 125mm (5.72 L/sec) is not.

The 'box' the overflow pipe attaches to and the sealed top inlet does interest me as it would require a vortex to travel horizontally to the furthest top meshed inlet. I have never tested this scenario and I can find no references. Do you know the make and size of the tank as I would like to look into this but my gut feeling is that if there was any syphonic effect, it would be short lived as the velocity through the outlet would be insufficient to prevent shear down the vertical pipe and so any flow rate gain would most likely be inconsequential. If I am correct, then the overflow pipe's capacity is way, way, way insufficient.

It requires 1.21 m head for a 100 metres long 104mm ID pvc pipe to flow at 600 lpm. This is not a wet system that has slow flowing water at the head and 6 recharge stations along its length, two of them close to the tank.

You have about 840 mm head but let's say 800 mm.

It really depends on how much head you lose fitting the leaf diverter but looking at the photos, I think that 300mm max loss is achievable. This would leave 500mm head which I feel is plenty given the closeness of two DPs to the tank. I hope that the plumber used 45 degree junctions at the two DPs nearest the tank but that of course is best practice.

Thanks heaps for your reply!

I've read your reply a few times and if I'm understanding it correctly, I should be ok making the following changes:

• Install drain valve where current pump intake is plumbed
• Install pump intake in the outlet above the current one
• Remove current riser bend and intall a mosquito proof flap valve
• Fitting leaf diverters to the 7 downpipes around the sides of the house shouldn't affect their ability to drain to tank

The tank is a Yarra Plastics 2000L slimline: http://yarraplastics.com.au/slimline-water-tanks/
Unfortunately as you have correctly surmised, the documentation from the tank manufacturer is lacking.


This is what I was afraid especially because I'm on reactive clay soil. Can this be remedied by creating another overflow outlet lower down and joining it with a T section to the existing overflow? I've also seen some of your previous posts recommeding using a bellmouth inside the tank but I'm not sure how this can be achieved...

In regards to the build up of silt from slow flow - I've also seen a diagram on this forum of a DIY sediment trap. I'm assuming this is not possible for my current scenario as the concrete apron has already been installed?

• It is difficult to desludge that style of tank but yes, a ball valve needs to be fitted. A cheaper alternative is to seal it with a poly threaded plug.
• Certainly much better than having the pump connected to a drain valve with the inflow dropping from directly above but that higher outlet is really for a tap to fill a watering can etc. The tank has the threaded fittings molded into the tank, using the higher outlet is obviously your only immediate alternative.
• Yes and the vertical section of pipe also needs to shortened. You can buy 5, 15, and 30 degree 100mm DWV bends in either M+F or F+F, I would use a 30degree bend.
• The system's required head. at a given flow rate is difficult to say accurately because there are less friction losses upstream due to lower velocities plus there are several recharge points (downpipes) that harvest different roof areas along the pipe's length and I don't know whether the pipes have been plumbed with 45 or 88 degree junctions. What we do know for certain is the pipe between the downpipe nearest the tank will carry 100% of the inflow and we also know what the friction loss will be along its length. This also includes the friction losses through the junction (either 45 or 88 degrees) and the bends as equivalent pipe lengths plus the riser pipe. This downpipe is fortuitously only about 3.5 m from the tank and the two closest downpipes are only about 2.4 m apart. A quick calculation and experience tells me that 500mm heads at all DPs will be fine.

The lack of critical information is common place and a blight on the industry.

Unfortunately yes and a common problem.

Roof drainage regulations end at the downpipe and stormwater pipe regulations do not apply to rainwater harvesting diversion pipes.between the downpipe and the tank, only from the overflow pipe to the LPOD. This is something that few understand.

Having large pipes all along a wet system is crazy as they have no flushing velocity at the head of the system. The only time these pipes flush sediment is after the pipes have build up so much sediment that the flow restriction generates turbulence. If organic matter also accumulates, bacterial die off often combines with the sludge to create a resinous black gooey tar like substance that won't flush. This can take many years to build up though.


I believe that the overflow is more efficient than normal by virtue of the nearby sealed inlet restrictting a vortex forming.

I suggest as a first step you phone Yarra Plastics and ask what the tank's overflow capacity is with a meshed 100mm overflow and the nearest top inlet sealed. If they don't know and I would be more than shocked if they did, they should do a long overdue flow test and you should ask for this. It is important to have this information before planning the next step.

Also ask where the best place to fit an additional overflow pipe is. These tanks often have different requirements and you don't want to risk voiding the warranty.

Fitting an air gap to the vertical overflow pipe and removing the overflow outlet's mesh will increase the overflow capacity about 20%.

A bellmouth fitted to a vertical drain pipe will dramatically improve the flow capacity but there has to be a suitable storage height above the pipe's crest. I get regular enquiries about tanks overtopping and some of those have had high level overflow elbows fitted inside the tank. It is becoming an increasingly common issue.