• You are viewing the forum as a Guest, please login (you can use your Facebook, Twitter, Google or Microsoft account to login) or register using this link: Log in or Sign Up

Hose diameter, bottle necking and flow..

scapegoat

Member
Joined
4 Nov 2012
Messages
134
Location
Norwich
I know this subject has been covered greatly on here but after measuring the diameters of various components I have some queries. After recently purchasing an APS 2000 external filter I discovered that although the hose id is 18mm, the spraybar right angle piece id is only 10mm with the main bar being 14mm, therefore greatly reducing flow. So I've been hunting for 18mm pipe to build one with but then today decided to measure a few other parts of the the system and discovered that there are quite a few bottle necks.

The 16/22mm UP inline atomiser has a 14mm id and the actual outlet on the filter also has a 14mm id. I'll be measuring the hydor inline when it arrives.

So surely this means that the flow is being reduced by the atomiser and the filter outlet, does this mean that the tubing doesn't need to be 18mm and the pipework doesn't need to be 18mm? Does the length the water is bottle necked for effect the flow?

Let me know if this doesn't make sense or I'm missing something

cheers
Jacob
 
Well the majority of filters dont come with 18mm pipe! most are fitted with 16 -22mm tube.
Some of the really powerful filters come with 25mm id & the small ones with 12mm id.
If you really want to keep the flow to its max then any restriction in diameter are best avoided.
 
The heater and atomiser will be ran off a T junction downsizing to 16mm leaving an 18mm straight as not to disrupt flow, however if the outlet diameter is only 14mm is this actually necessary?
 
OK cheers Foxfish, guess i'm just fussing over a few mm without actually testing to see if the system works

Couldn't find 18mm Ys so went with Ts, the 18mm goes straight through with the 16mm angling off the T, my thoughts were that the atomiser would still put the same amount of Co2 into the water and the heater would still heat the water until the required temp. The heater will have to work a little harder but hopefully not a great deal
 
Ok so now I'm really stumped :wacky: I was just piecing together my pipework when I noticed something, on the T junctions the id of the18mm bits are actually 14mm. It says 18mm on it so is obviously designed for that size tube but limiting down to 14mm is going to cause the same bottlenecking as the heater and atomiser, surely this means the split is pointless?

I can take pics if it helps explain my predicament but would really like to know if i'm wasting my time splitting.

Hope someone can help

Cheers
 
the measurements on fittings are always the external diameter so it fits your tubing. With the atomiser you can trim the ends of the connection to widen it a little but ultimately flow will always drop when using inline equipment. The only real solution as fox said is to have a bigger filter/ second filter or add powerheads to compensate for the drop.
Using T's and Y's or increasing the diameter with reactors (even empty) will also drop the flow. It really is much easier to add more flow than trying to remap and split up pipework.:thumbup:
 
my two cents....... i have x2 of the aps 2000's. one of them i have used a t bar. it splits the 22mm into x2 18mm if your looking at the piece the right hand side of the 'T' is 22mm and the left and bottom are 18mm. between this one and the standard setup i have not noticed a drop in flow (both are using the spraybar) so it must be working.
 
Cheers guys you both have very interesting points, I'm now thinking that due to the filter outlet id only being 14mm anything over 14mm won't make much difference as the amount of flow has already been limited by that.

I have 3 options, think I might run a test to see which gives better flow

a) Remove the splitters and run the atomiser and heater on the 18mm hose with hose clamps.

The trouble here is that the 16/22 hydor only has a 12mm id therefore limiting flow even more so

b) Run the system with the T junctions splitting off for the heater and atomiser

Flow shouldn't be affected by bottlenecking however the extra loop could limit it? Also the heater will have to work harder.

C) Swap the 18mm Ts for 22mm Ys, the Ys should have an id of 18mm to match the hose.

This will test whether my theory of 14mm being the limiting diameter is right or wrong, it will also make the heater slightly more efficient the b).


I could be totally wrong as i don't know a great deal about flow dynamics, this forum has offered so much information on it for which I'm very thankful, I guess I just want to make sure i'm getting the most out of my filter

Thanks

Jacob
 
Hi,

A small restriction given by a short bit of something won't make all that much difference. It will make a difference, but not much. If you take a long piece of pipe which is a small diameter then you will see a noticeable effect.

Water flow in pipes is analogous to current flow in wires, where voltage is pressure (or voltage drop across a resistor is pressure drop across a length of pipe), current is flow rate and the pipe thickness is resistance. If there is a small value of resistance in series with a circuit there will be a small effect on the current flow (and a voltage drop accross the resistor) If you have a 2 of these resistors then there will be twice the effect and so on. This doesn't help if you don't have an electrical background, but a lot more people know about Ohms law than flow through pipes.

There is added complexity in pipes is due to the surface roughness and vortexes which have a big effect on flow. (Do you remember the term gas flowing as applied to polishing the inlet manifolds of car engines to get more power?)

There are a lot of on-line flow calculators and this is one.. Pressure Drop Online-Calculator if you put in sensible figures it will calculate the pressure drop (be carefull with the flow rate - 2000l/h = 2m^3/hour). It doesn't give you a flow rate i.e. working backwards but it does give you some idea of the effect of the restrictions you will be creating.

It's also interesting to look at the effect of roughness in pipes AKA fluval.

Cheers, Mike.
 
Cheers Mike very informative, this has flipped my ideas again however think i'm starting to get a little bit of a grasp on the matter. I guess the easiest way to find out is to have a little test today, I'll just test a) and b) and see which gives better flow rate and go with that

thanks
Jacob
 
I'd be very interested what you find with that. I have split my fx5 output with Y pieces to include heater and co2 atomiser and including the jubilee clips it came to about £25 worth of hardware, which was the cheepest I could get it.

Looking forward to your results :) Mike.
 
Well unfortunately I just tried to test it done something extremely stupid :sour: I didn't install the UV as I didn't think it was necessary for the test, this however meant that there was the live UV bulb connectors in the filter it doesn't take a genius to work out what happened next :meh:. So now I guess I'll wait for it to fully dry out and install the UV and try again or I might have broken it.
 
Just discovered the switch for the UV was on, this was a stupid mistake I made but you would assume the switch to be off when sent new and also something in the instructions saying to make sure you install the UV before turning on the filter. So anyway I've turned that off and attached the glass cover to stop any more water touching it. On with the experiment!
 
Oh dear :(
Hope it dries out OK. If you can leave the wet connectors on a radiator for a day or so it should be OK. Best to make sure they really are dry in case something starts to rot....
 
OK so test results are in!

Note that I've removed all media for this experiment.

a) Heater and atomiser inline with 18mm tube with hose clamps.

10l in 37 seconds which if my calculations are right gives 973 litres per hour.

note - small leak in atomiser connection to 188mm tube, nothing an O ring, elec tape or silicone won't sort out.

b) Heater and atomiser running off the bottom of a T junction and back to the bottom of a T junction.

10l in 36 seconds which should give 1000l per hour.

note - hardly any water was passing through the loop making Ts pretty useless.

This means there was a 6.7% drop in flow between b) and a). However as hardly any water passed through the loop I think I'll make do with the 6.7% drop and got with a) as the atomiser and heater will be a lot more efficient. I would have liked to have had some Ys to test with as well but as the test was a massive chore to set up think I'll leave it at that

*As Spikey-Mike later points out there was actually a 2.7% drop not 6.7%
 
Hmmm, I make it 2.7%....

You'd really want to arrange the flow into the base of the 'T' to make sure you got an equal split.
Thanks for posting the results anyway. Looks like I probably wasted my £25 quid :)

Cheers, Mike.
 
Now there's one I really should have tested! Just dismantled everything and cleaned the small pond I created on the kitchen floor ;) I guess that as I need the heater and atomiser in the system it wouldn't have made a great deal of difference to my end decision but would have been very interesting to see how much difference the inline equipment makes
 
You're right there Mike! Looks like I got the 7 and the 3 the wrong way round :crazy: maths never was my strong point.

That's made me even more sure of my decision now, a) it is!

thanks
 
Back
Top