Estimate flow loss when reducing pipe diameter?

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
I'm looking for your estimate (or calculations if you want to) on the flow loss when switching 19 mm pipe with 16 mm pipe on a canister filter?

I know there are some unknown variables like filter power etc. But just want your rough estimate.

Say a canister filter rated 2000 lpr with 19 mm pipes, would run how much with 16 mm pipes?
 

Zeus.

Member
Joined
1 Oct 2016
Messages
3,267
Location
Yorkshire,UK
I'm looking for your estimate (or calculations if you want to) on the flow loss when switching 19 mm pipe with 16 mm pipe on a canister filter?

Thats a big ask as theres a hit with 19mm pipes - head of water, length of pipes, bends, filter media, pump to start with :eek::eek:

But just want your rough estimate.

Less - dependant on head of water, length of pipes, bends, filter media, pump :facepalm:

I wouldn't give an estimate :arghh:
 

rubadudbdub

Member
Joined
27 Oct 2015
Messages
126
Hagen Poiseuille equation will tell you the principles. I've tried to fit your numbers to the equation, but maths isn't my strong point. An abbreviated version:

Flow = Pi x pressure gradient x radius ^4
--------------------------------------------
Constant x viscosity x length tube

Flow varies proportionally to Pressure gradient along the tube, and to the radius of the tube raised to the power of 4. More pressure or bigger tube = more flow

It varies inversely to length of tube and viscosity of the fluid. Long tubes and viscous fluids will restrict flow.

So radius has a big effect, due to it being raised to the power of 4.

The maths in your case is the bit I'm not sure about. If diameter is 16mm radius = 8mm. I'll save you some boring bits but comparing 8^4 = 4096 vs 9.5^4 = 8145

The ratio is ~ half, so if all other factors remain the same then flow should also be halved. But I'll admit that seems a big drop in flow for loosing 3mm off a 19mm diameter.

Also, Surface area of 16mm pipe 200mm2 vs 280mm2 for 19mm, by my calculation. This doesn't relate to the HP equation but will give you a rough idea that internal resistance will increase and reduce flow.

Hopefully someone with better maths will come and clarify.

Overall, I'd not be suprised if you lost at least a third of the flow, although my fag packet maths suggests half.

If you must use 16mm pipe, you can mitigate the flow drop to some extent by:
1. reducing the length of the tube to a minimum
2. maximising the pressure along the tube by raising the filter up and reducing losses due to water head pressure.
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
I wouldn't give an estimate :arghh:
I know it's quite an impossible question to answer. Was hoping someones experience could give me a hint.

Overall, I'd not be suprised if you lost at least a third of the flow, although my fag packet maths suggests half.
That's a big loss. True the total hose area is roughly 1/3 less from 16mm to 19mm.

Question is what canister to use:
2000 lpr using 19 mm hoses hooked up with 16mm hoses.
or
1700 lpr with 16 mm hoses.

I wonder how big a difference there will be between these two options.
 

papa_c

Member
Joined
22 Jan 2013
Messages
286
I would always go with the original design recommendations, as there would have been some extensive testing during development.

However if you already have the 19mm filter you could always mock up the reduction and test the output.
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
However if you already have the 19mm filter you could always mock up the reduction and test the output.
I'm currently running with this.
However, looking at the option to replace it. My problem is trying to figure out if it will make any difference. If not, it will just be a waste of money.
 

rubadudbdub

Member
Joined
27 Oct 2015
Messages
126
you already have the 19mm filter you could always mock up the reduction and test the output.

... and then tell us please. I'm sure lots of people would be interested.

If you step down from 19 to 16 it makes sense to me to keep the 16mm section as short a length as possible.
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
If you step down from 19 to 16 it makes sense to me to keep the 16mm section as short a length as possible.
Sure. As mentioned, I also have a lot of equipment connected to the filter, which further reduces flow.
Most of the equipment runs 16mm, so cannot do much 19mm. The intake is also reduced to 16mm.

I just (2 min ago) ordered a 1850 lpr filter that uses 16mm hoses. So taking my chances. Will pick it up Tuesday.
I can do a little visual testing on the flowrate on current and new filter as I'll be using the same hoses. Might even test how long it takes to fill an x liter container for some real data.
 

rubadudbdub

Member
Joined
27 Oct 2015
Messages
126
Question is what canister to use:
2000 lpr using 19 mm hoses hooked up with 16mm hoses.
or
1700 lpr with 16 mm hoses.
Do you already have some 16mm fittings/glasswork that you want to use?

If you are trying to compare different brand filters then I'd caution against assuming that the flow rates will be comparable anyway. The differences between manufacturers stated vs actual flow rates may well make any vaguely scientific comparison mute.
 

rubadudbdub

Member
Joined
27 Oct 2015
Messages
126
This was in PFK a while back. It gives you an idea of expected differences between stated and observed flowrates for various head heights.
 

Attachments

  • DSC_0000_BURST20200417091817573.JPG
    DSC_0000_BURST20200417091817573.JPG
    3.3 MB · Views: 20

Jayefc1

Member
Joined
2 Sep 2017
Messages
2,119
Location
Swadlincote
Pump 2litre from a bucket to another bucket with the two diffrent tubes time them.and you will see the reduction in flow
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
If you are trying to compare different brand filters then I'd caution against assuming that the flow rates will be comparable anyway. The differences between manufacturers stated vs actual flow rates may well make any vaguely scientific comparison mute.
Very true. Which is why deciding what to do in my case was difficult.
I do believe I will se an increase in flow going from my current cheap brand with small hoses rated 2000 lpr to eheim with standard hoses rated 1850 lpr.

Pump 2litre from a bucket to another bucket with the two diffrent tubes time them.and you will see the reduction in flow
Yes. Once I've done a test with full setup, and switched filters, I can easily test the current pump with 19mm and 16mm. I can even try with various length hoses and heights to make a dataset.*getting my nerd mode on*:p
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
The bypass allows flow around the diffuser, with the valve controlling the amount flowing in the bypass.
Thanks for the input. I do have experience with a bypass and did run one with a CO2 reactor some time ago.
The pipework that goes through my tank and my reactors are all 16mm, so it is not possible in my situation. Also, what I need is the flow through my equipment, not just return it to the tank.
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
So...
I just installed the eheim 2078 (1850lpr).
It can't run both of my reactors and uv filter at full power :banghead:
Only 3 of 6 steps, whatever flowrate that means (50%?).
Did seems like more flow in the reactors.
I removed one reactor, which was only needed because of previous low flow, and the eheim picked up a bit. Still cannot get full power, but quite a flow in the reactor:thumbup:
Now I'm just checking to see if I can get a decent ph drop without a huge gas buildup in the reactor. If possible, then I'm making a bypass to get the full flow and filtration out of it:)
 

kilnakorr

Member
Joined
16 Mar 2020
Messages
208
Location
Denmark
@Zeus. A little build up is fine, but don't want that loud splashing sound:stop:

It's a pretty large reactor, just hoping the flow is enough.
Would splitting the output between reactor and uv have an effect? It would reduce hose length on the reactor, but total it would likely be more.
 

Zeus.

Member
Joined
1 Oct 2016
Messages
3,267
Location
Yorkshire,UK
Would splitting the output between reactor and uv have an effect?

Yes the flow in the reactor and UV filter would decrease dependant on the fittings and their would be an increase in the total flow dependant on the size of the fittings used. I oversized my fittings to has the least amount of increase in resistance so less drop in output.

You do get better CO2 uptake in a reactor thats at a higher pressure, but that has a bigger impact on the flow/output and puts more strain on the pump also.
 

Similar threads

Top