Q for expert in fluid dynamics

[CThompson]

Can I get 12,000 of water gravity feed down two 32mm non-pressure pipe fittings approximately 100mm long? That breaks down to 1.6 litres of water per pipe per second.

Craig

[anchar]

Hi Craig,

1.6 Ls -1....from what height?

merjo

[CThompson]

Is the height relevant? The issue I thought was the pipe length which as mentioned is about 100mm.

This 100mm section of pipe will be going through a 32mm bulk head fitting, fitted inside an overflow box which is 1.7m approx from floor level.

The bottom of this 100mm pipe will fall into an open topped filter similar to a sump.

[Eddie Salita]

Criaig, a question. 1.6 X 60 is 96 litres per minute. 96 X 60 is 5768 litres per hour. I assume you have multiplied this by two because you have two pipes then rounded 11536 to 12000 ltres per hour?

Is this draining the tank from near surface level where there is not much pressure, which may be what merjo is getting at. If so I would make them bigger if it was my tank. a 50mm pipe will carry almost 2.5 times the volume of water at equivalent pressure. IME weight/pressure of water is everything with flow rates.

Cheers,

Jason

[BaZ]

Hi Craig

I am assuming you are not using a 12000 L/hr pump, because as you know it will not be pumping that much at that head height.

I cannot remember what your pump was rated at.

Anyway like I said I assume you have taken pump head height into account, but thought I'd throw this in just in case

[CThompson]

The pump is an Oasi (is that how you spell it?), and is made for pumping water up waterfalls. So yes, I will loose some water through the head height, which is approximately 1.70, but not a huge amount.

The tank has an overflow box, which has been constructed shallower than I would like, at a depth approximately 180mm. The overflow box has been drilled to allow a 40mm fitting, but I have fitted it with 32mm in the hope this will be sufficient, but with a fall back of using the larger 40mm size if needed. The overflow box is drilled with three holes, two fitted with the 32mm bulkheads, and 32mm pipes, the third hole is the ‘safety’, plumbed with the 40mm fitting and the pipe attached cut at a height greater than the 32mm pipe fittings, but still below the level of the overflow box (obviously).

Jason – 12,000 lph divided by 60 (minutes) = 200. 200 divided by 60 (seconds) = 3.333333……Divide this by 2 (there are two pipes) and you get 1.66666666…….7. So this means I will need to get 1.6 litres of water through each pipe per second. Will this happen with a 32mm pipe, with a head pressure of very little?

I was at Waz’s place last night, and he had a 7000lph pump, with the water falling out of a 32mm pipe (is that right Warren?), the water level was only 15-20mm above the top of the pipe. This implies to me that I should be right with my two pipes doing 12,000 (especially keeping in mind I’ll loose some of that 12,000 litres with the head height.

I was hoping someone out there would be either a fluid dynamics expert, or have two 32mm pipes with 12,000 lph going through them, and they could say ‘YES IT WILL WORK’, so when it comes time for me to ‘suck it and see’, I will have a better idea than my current thinking of “she’ll be right mate”.

Craig

[Ash]

if it's drilled for 40mm bulkheads why not use them?

[Eddie Salita]

You can be your own fluid dynamics enigneer If you want. Get a bucket, put a 32mm fitting in the bottom, fill it, and time it. Then you can provide peace of mind for yourself. Cheaper than a fluid dynamics engineer too.

Cheers,

Jason

[CThompson]

Ash,

if it's drilled for 40mm bulkheads why not use them?

Gee! Why didn’t I think of that?

When one drills tanks for plumbing fittings, one drills the holes needed for the choice of bulk head used. A wise person will drill the tanks the bulk head fitting the next size up, so if needed it is a relatively simple matter to use a larger bulk head fitting, rather than having to get the hole redrilled.

I thought 32mm was big enough (but now I'm having doubts), but had it drilled to allow for a larger bulk head if needed. So it is not drilled for 40mm, it is drilled for 32mm with allowance for a larger size fitting.

Jason

You can be your own fluid dynamics enigneer If you want. Get a bucket, put a 32mm fitting in the bottom, fill it, and time it. Then you can provide peace of mind for yourself. Cheaper than a fluid dynamics engineer too

Gee! Why didn’t I think of that?

If you’re so confident that this will work with the precision to bet your life on it, go ahead. How does one get 1.6 litres of water through a 32mm pipe, and KNOW for certain, it was 1.6 litres… no wait, maybe it was 1.5 litres…no wait, maybe it was 1.7 litres….no wait....

That is why I was asking for help. I don’t need comments like this, I can work thaem out for myself. I was hoping there was maybe a person who has simply done it (after all, the combined number of tanks that SCP people must have must be pretty huge), or there was a person who due to their profession would know the mathematics to work it out.

Craig

[Eddie Salita]

Its simple maths craig. put 16 litres of water in a bucket If it takes less than 10 seconds to drain it will work. There is nothing difficult about this. So long as you dont have excessive weight of water over the fitting you will be getting an accurate result. It is not an advanced fluid dynamics mathematical issue like you make out.

If you’re so confident that this will work with the precision to bet your life on it, go ahead. How does one get 1.6 litres of water through a 32mm pipe, and KNOW for certain, it was 1.6 litres… no wait, maybe it was 1.5 litres…no wait, maybe it was 1.7 litres

How hard is it to get a known volume of water and drain it thru a known hole and time it? I have simply done exactly this for myself. I will bet my life on it. But I wont bet my life on your setup. Why would I put my life on the line for your setup which I havent seen. What an arrogant thing to say....

If you want to be sure beyond belief get a large rectangular tub, fill it to 180mm depth, then drain it thru the relevant sized hole. If your good at maths(and wise) like you say, you can work it out VERY accurately. How much more peace of mind do you need. Also if you think you dont need comments like this you should try to be a little less facetious with your replys. Notice merjo hasnt posted again, probably due to your lack of abitlity to LISTEN. She is probably the closest thing you will find to a fliud dynamics engineer here(if not better), and you alienated her. Her brain eats brains like ours for between meal snacks. You say you are asking for help, You dont want help, you want to bite the hand that feeds you. Not to mention something for nothing. A FDE will ask you more questions than you could ever answer, and then give you a theoretical answer which will depend on a goodly number of varable factors. Its what they do. This is not what you want either, you want a yes or a know. an FDE or us cannot give you this, we CAN share our PRACTICAL knowledge with you but we cannot teach you, you have to want to learn. I just dont see how we can satisfy you. Hence the fact no-one here is trying.

Goodbye,

Jason

[CThompson]

Jason – bite me!

I asked for someone who could give me an answer to a question. I didn’t ask for someone to tell me to go away and fill up a bucket. If I had confidence that this would take care of all the variables that the situation would produce, it would have been done already. As I’m sure someone out there will have had a tank set-up similar and I was hoping I could benefit from their experience not from your lip.

Nothing was said to merjo that could even come close to be mistaken as rudeness – unlike what you have said to me. I answered her question, but didn’t receive further impute from her, which is her choice. I’m sure she doesn’t need someone like you to defend her.

I spend hours on these pages answering question I feel experienced enough in to have confidence in my answers. If I felt confident in my ability to KNOW for certain on this topic – I wouldn’t have posted a question to those wiser than I in this matter. At the end of the day, I’ll fill the thank, and turn the pump on. I was hoping to have a better idea from an expert or an experienced individual from the SCP before then. So wether I get an answer or not, I will find out in the end.

Her brain eats brains like ours for between meal snacks.

Yours obviously needs a refill!

If I thought I was so “wise” I wouldn’t be asking the question, I’d already know the answer.

Goodbye to you too.

Craig

PS – sorry Mods – his reply was too much for me to not respond.

[TSMill]

[Eddie Salita]

[TSMill]

Bad case of baxter detention syndrome going around.......

[Eddie Salita]

Indeed. I had to follow your edit

Cheers,

Jason

[TSMill]

haha...the sad thing is i actually did drag out my fluid dynamics text and geek it up big time in excel!

[Lee Miller]

Hi Craig,

I have a basic understanding of fluid dynamics (studies at TAFE a million years ago) but simple physics will get you most of the way to the answer you’re looking for:

6,000 litres per hour (per pipe) equals 1.67 litres per second.

1.67 litres equals 1,670,000 cubic mm.

A 32mm internal diameter pipe has a cross-sectional area of 804 sq. mm.

Flow rate (cu. mm per second) divided by area (sq. mm) equals velocity in mm per second.

1,670,000 divided by 804 = 2077 mm per second = 2.077 m/s

Acceleration due to gravity is 9.81m/s/s.

Therefore, from a standing start, the water should reach the desired velocity in 0.2117 seconds. In other words, yes, you should be able to move 12,000 litres of water per hour through two 32mm diameter pipes.

However, (and this is where my basic understanding of fluid dynamics comes in) the flow of fluids through pipes is laminar and friction comes into play. Bernoulli’s principle/equation applies. Flow rates will also be affected by the ‘head’ above the inlet and the type of inlet chosen. Bends, strainers, spinning arms and whatever other restrictions are placed in the flow will obviously slow the rate.

HTH.

Lee

[Eddie Salita]

There you go Craig. A definite Maybe. You gotta be happy with that

[CThompson]

WOW

I'll have to read that a few times to get it to sink in. Thanks Lee.

Craig

[TSMill]

Sorry Lee, i f only it were that simple!

....but you're right, you do have to solve Bernoulli's equation

[CThompson]

Therefore, from a standing start

I assume that means water on level with the inlet to the pipe, that then flows over into the pipe, without any sort of assistance of water above pushing it through?

Flow rates will also be affected by the ‘head’

I understand this to mean - the more water (height above the pipe) pushing it from above (to a point), will increase the amount of water passing through the pipe, till it reaches a maximim amount of water, afterwhich could only be increased with some sort of machanical help.

The first and second quote above kind of contradict each other, but I take it you have worked out how much water will pass through the 32mm pipe, with no assistance from water above it pushing it through (from a standing start)?

The two pipes that the water will flow into are only about 150mm long, and they open (will open to) a flat sump like tray which will not restrict flow to any great degree. That is, after they exit these 150mm pipes, the water will not be impeded by anything that will cause the water’s flow through the pipe to be restricted.

Absolutely, completely and totally impressed Lee

Craig

[anchar]

However, (and this is where my basic understanding of fluid dynamics comes in) the flow of fluids through pipes is laminar and friction comes into play. Bernoulli’s principle/equation applies. Flow rates will also be affected by the ‘head’ above the inlet and the type of inlet chosen. Bends, strainers, spinning arms and whatever other restrictions are placed in the flow will obviously slow the rate.

What Lee has said is basically correct. The calculations are near enough for your needs if your cross-sectional flow (head outlet/pipes) is uniform ie. 32mm throughout. My FD is VERY rusty however I do remember that the following had to be taken into consideration:

We are assuming that the water running through these pipes is behaving as an ideal fluid ie. assume that there is no internal frictional forces between adjacent layers; assume that the fluid is not compressible (ie. density remains constant throughout the journey); assume that no turbulence is encountered (basically speaking each element in H2O has zero angular velocity about its centre) hence no "eddies" are formed in the moving water; and, assume that the velocity/ density and pressure at each point in the fluid is constant (ie. that the fluid motion is constant).

Another point to consider is the area where the water flow changes from the 40mm to 32mm (if this in fact appropriate...I'm a little uncertain about your wording here). If this is the case, then to accurately work out your problem Craig, use the Continuity equation:

A1v1 = A2v2...where A1 = cross sectional area (point of entry ie. the 40mm diameter of the head); A2 = c/s area of exit ie. 32mm pipe.

At all points, the velocity of the particle (water) is tangent to the streamline along which it moves. In a small time interval delta t (soz I can't do Greek symbols on this computer), the water at the bottom end of the pipe moves a distance of delta x1, where delta x1 =v1delta t. Since mass in (M1) must equal mass out (M2) (assuming no leaks), these have been cancelled from the equation. In short, the product of the cross-sectional area of the pipe and the fluid speed at any point along the pipe is a constant for uniform size pipe. In the case of 40mm head reduced to 32mm pipe, the speed is high where the pipe is constricted and low where the pipe is wide...obviously. The product Av (which has the dimensions of volume per unit time) is the flow rate.

Don't worry about Bernoulli's equation...the Continuity equation will work and is much simpler.

The reason I asked about height was because Torricelli's law says that the speed of water emerging from a hole is equal to the speed aquired by a body falling freely through a verticle distance (h). Hence you could work backwards and determine the velocity of the flow by taking the sq. root of 2gh (all things being constant )....however, the viscosity of water (at 20C) 1.0 x 10 (-3) (N.s/m2)...so shear strain and shear stress......and then there's Poisellville's law which states that the rate flow is greatest through the centre of the pipe (in profile)

Craig, long story short...Lee's way should suffice

merjo

[Eddie Salita]

you just ate craig and Jason's brains Merjo and proved me Wrong. Mine was the main course but

Merjo, is Poiseuille's Equation ONLY applicable to laminar flow? And can it be used in electrical circuits where pressure is voltage, and volume is current. It is relevant (i think) to working out the resistance of an unknown conductor. Keep in mind I am only an amatuer scientist. It seems simpler than ohm's law in my situation, as the material is not a good conductor, and the current is too high to test with my equipment. Not to mention that ohm's law eats my brain also.....

I searched my little heart out, but most texts are a bit to advanced for me. I was going to ask Dr Karl. But then I saw this

I guess I just want to know if the values can be substituted. But it really depends on how accurate Poiseuille's law is to non laminar flow????

Cheers,

Jason

[WAZ]

I think she just ate Lee's too - cause he was just typing it straight outta the text book - he's a clever fella - just not that clever.

Craig - stop messin about - fit the 40mm fittings - fill the flippin tank and try it.

WAZ.

[anchar]

Hi Jason,

Poiseville assumed that the flow was streamline, so he developed the relationship between tube dimension, pressure and flow rate, ie...

Rate of flow = deltaV / deltat = (P1 - P2)(piR 4)/ 8Ln

At first glance, his theory doesn't agree with common sense IMO. It is reasonable that the rate of flow should increase if the pressure difference across the tube or the tube radius increases. Likewise, the flow rate should decrease if the viscosity of the fluid or length of the tube increases. That's why he has R (radius of the tube) and P1 - P2 (pressure difference) in the numerator and L (tube length) and n (viscosity coefficient) in the denominator.

It is often applied to working out the rate of blood transfusions etc. but I can't see it's application to electrical circuits (similar formula though)...that's what Ohms Law is for

merjo

PS> I ate my brain too...40C odd here atm...way to hot for these type of posts