• 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

Another furrowed brow in this business we call flow!

J Butler

Member
Joined
26 Jul 2011
Messages
135
Location
Leicestershire
Apologies in advance if this appears to be a foolish question, I have covertly trawled this most excellent forum for quite some time now and i’m almost ready to start applying some of what I think I have learnt to my own aquarium :wideyed: (I think I may actually miss all the theory, practical application can be a bit more intimidating!)

Now, one underlying message that seems to underpin everything is that flow is of paramount importance. I’ve got myself in a bit of a quandary on how best to achieve this however. At the minute, I’m considering buying a Fluval FX5 on the back of all the praise it has received. Originally, I was planning to add a hydor external heater and inline CO2 diffuser to this but i’m starting to have reservations with the amount of flow restriction this will place on it.

With this in mind, I’m now thinking about adding a second loop run by an external pump such as one of the eheim compact+ range that would have the heater and CO2 attached leaving the filter to do what it does best.

Now for the bit that’s way over my head, plumbing and flow dynamics isn’t exactly my forte... Would it be possible to plumb in both pump outputs into a main spraybar that will hopefully cover the whole length of my tank (DIY jobby methinks), something like the image below (apologies for the simplicity, that’s about the extent of my knowledge though :? )

scaled.php?server=217&filename=spraybarquery.jpg


How detrimental would the back flow be and would there need to some kind of check valve on each side? Just for good measure, do you think it would be worth including some kind of ball valve to isolate things for maintenance if it is at all feasible?

The crazy thing about all this is that the tanks only 180 litres :shifty: So the FX5 may well be more than enough for everything on its own but I’m going for overkill for future (hopeful) upgrades. Might need to mention that I don't want to hard plumb anything if that makes a difference, just flexible tubing if at all possible.

Thanks in advance,
Joe.
 
Hi,
At the "Y" junction of the two pipes the back pressure will be a problem. The diameter of the output manifold where multiple flows meet must be at least the sum of the diameters of the branch pipes, otherwise it's like squeezing 10 lbs of muck into a 5 lb sack. Check valves will not solve the issue of reduced flow and back pressure at the junction. That new manifold diameter must be continued throughout the rest of the piping (i.e. it can increase but not decrease).

Cheers,
 
OK, thanks ceg.

So if the output manifold is sufficient, would I be losing flow elsewhere in your opinion? Do you think it's worth persuing or is it a bigger headache than it is worth?
 
Well, it really depends on your budget and on your DIY skills. If the filter had a strong enough output then you wouldn't really need to have separate branches. You'd just accept the flow losses associated with having both an inline heater and an inline diffuser and you'd still have sufficient flow. With a target throughput of 1800LPH the FX5 has headroom to spare at a rated 2300LPH - however, to get that throughput, you must stick with the stock tubing diameter. Using a smaller diameter hose (or even stepping down the diameter with an adapter) to accommodate the heater nozzle would choke the flow.

Another option is to use the filter as your diffuser by simply porting the in-tank injected gas towards the filter intake. If components are available, the stock filter impeller could be removed and replaced with a pinwheel impeller which would chop the CO2 bubbles up even finer to get better diffusion. That would leave you with only the heater to install inline.

Using your original scheme, it would be a matter of trial and error to determine how large a manifold to use. Other flow losses only occur due to friction due to excessive length of tubing, back-pressure due to excessive vertical distances from the water level to the filter inlet, or back-pressure due to tubing cross sectional area reduction. If you avoid these pitfalls there ought not to be any other factors resulting in flow loss.

Cheers,
 
Hi

Ceg, I'm not trying to be picky as i've a huge amount of respect for the advice you give, but you quoted -
The diameter of the output manifold where multiple flows meet must be at least the sum of the diameters of the branch pipes

Shouldn't this be the cross sectional area of the output manifold where multiple flows meet must be at least the sum of the cross sectional area of the branch pipes?

To be strictly correct it could be smaller than this as the ratio of 'frictional surface' to cross sectional area changes somewhat but this difference is quite small

Hope you don't mind me adding my two penneth worth...

Shaun
 
Not at all, wade in by all means Shaun! As I mentioned before i'm a complete novice when it comes to this and any help or suggestions are more than welcome.

In this case I presumed diameter to be tantamount to the cross sectional area as we're dealing with cylindrical pipes, I think it's also safe to assume that ceg's referring to internal diameter and not the total diameter.
 
Yep,sorry about that. It's the cross sectional area that counts for the flow. You'd have to calculate the required diameter for the manifold pipe based on the two branch cross sectional areas. :crazy:

Cheers,
 
Joe. I have the same dilemma. I'm about to move my tank and and it would be more simple to combine my 2 loops into one. As you know I have a filter rated at 1700lph driving a spraybar with no inline attachments and a 3000lph external pump driving the main spray bar with a inline diffuser and heater. My tank is also 180L.

I have more than enough flow and have turned down the filter and external pump. I also run them on separate timers so when I'm injecting co2 I have high flow and when I'm not I run only the filter and give my creatures a break.

I reckon even with squeezing 2 tubes into one of the same diameter you could overcome the loss of flow by having a pump and filter which exceed your requirements and that you could adjust in terms of flow. I'd run them on separate timers.

By the way I made the mistake of thinking I needed a huge filter to provide flow. You don't. All the filter needs to do is biological filtration, you can provide your flow with the external powerhead.

I'm following this thread with interest.

T
 
Morning J Butler (very formal) try googling pumps in series v parallel. Basically what you propose is putting 2 pumps in parallel which means that in theory the resultant flow will be the sum of each individual flow (although in practise it is usually somewhat less). As long as the resultant pipe diameter is capable of dealing with the resultant flow then you shouldn't get any problems beyond smaller pipe = less flow. Back flow is only an issue when one of the pumps is stopped or so much smaller than the other that it is overwhelmed and some of the flow from the other pump diverts through the additional branch. Check valves will prevent this but also reduce flow and are not usually necessary on our size systems.

Just for info if you plumb a second pump in series (directly into the outlet of the first pump) then you add the heads of the pumps together ie. the equivalent vertical height which they can raise water to. Effectively you reduce pressure on the system which enables the overall flow to increase but this is limited by the capacity of the smallest pump.

Neither option is as good as separating the pumps into their own loops though ! Have fun
 
I agree that it would be easier to keep both filters separate and have a separate spray bar for each. Between them they could stretch across the whole tank like your diagram. Then simply have the intakes for each external in the opposite corner to each spray bar and you'll get excellent mixing of the water and good circulation.
 
Charlieh said:
Morning J Butler (very formal) try googling pumps in series v parallel. Basically what you propose is putting 2 pumps in parallel which means that in theory the resultant flow will be the sum of each individual flow (although in practise it is usually somewhat less). As long as the resultant pipe diameter is capable of dealing with the resultant flow then you shouldn't get any problems beyond smaller pipe = less flow. Back flow is only an issue when one of the pumps is stopped or so much smaller than the other that it is overwhelmed and some of the flow from the other pump diverts through the additional branch. Check valves will prevent this but also reduce flow and are not usually necessary on our size systems.

Just for info if you plumb a second pump in series (directly into the outlet of the first pump) then you add the heads of the pumps together ie. the equivalent vertical height which they can raise water to. Effectively you reduce pressure on the system which enables the overall flow to increase but this is limited by the capacity of the smallest pump.

Neither option is as good as separating the pumps into their own loops though ! Have fun

Thanks for all the replies guys. Sorry about the name Charlie, I wasn't trying to be formal, I opted for something easy rather than quirky for speed, for shame! ;)

I must admit i'm surprised at back flow not being an issue if I shut off one of the pumps when not required, would water not head back down that way when pressure has built a little at the spraybar?

Would CO2 distribution not suffer if the majority is isolated around one spraybar, or am I underestimating the supreme power that is diffusion? I suppose either way would be a slight compromise. It would be nice not having to fit in an extra spraybar though.

I've been having a look around for various parts, does anyone know if variable sized Y connectors even exist or would it have to be a fixed size connector with the intlet of whichever hose that doesn't fit enlarged or reduced with an appropriate fitting?

Just an aside, is there a safe way to secure fittings to flexible hosing without the risk of them coming apart, or is good ol' friction genrally sufficient with our tanks?

Thanks again guys,
Joe
 
nry said:
Could you add an inline pump instead? You may as well buy a more powerful filter though...

http://www.force4.co.uk/848/Rule--iL280 ... ign=pid848

Hi nry , my plan was to buy an external pump and run it inline with a heater and CO2 diffuser attached without having to worry about reduced flow on account of filter media ect.

I'm buying a Fluval FX5 so filtering should be more than covered as it is but didn't want to run an external heater and diffuser on it also as I would have had to reduce the pipe size for it, meaning it would take a further hit.

As for what inline pump to buy, going with eheim was more for familiarity and reputations sake than anything else.
 
Hi again Joe (that's better!)

If you run a filter on one branch and the extra pump includes an inline heater then I doubt you'll want to turn either off in which case back flow won't be an issue. However if you only run co2 off your extra pump and stop it when co2 is not in use then whilst you would get back flow from the filter through the extra pump and its intake back into the tank when co2 is off, does it really matter ? Also you don't have to use spraybars for each return from separate pumping loops you can use a combination of outlets arranged to get the best circulation.
 
Well, I went ahead and took a bit of a leap and attempted to combine the two loops without increasing the outflow diameter considering the flowrate was going to be somewhat overkill anyway, and preliminary tests seem to be promising! :D

I did struggle a fair bit to find something suitable with the large 25mm FX5 pipe diameters and in the end settled for the following y-connector (along with a few threaded adapters and hosetails).

t-pipe45.jpg


I'll first say that measurements are very approximate, I used a bucket with a litre scale on the side and a simple measuring jug to take the water down in the bucket to the nearest litre, not a meniscus in sight! That said, the approximations should be good enough.

Eheim compact + 2000 loop with inline atomiser and heater:
-approx. 2.8L/10 seconds -> 16.8L/min -> 1008 lph

Fluval FX5 loop:
-approx. 7.3L/10 seconds -> 43.8L/min -> 2628 lph

Both loops combined:
-approx. 9L/10 seconds -> 54L/min -> 3240 lph

So, if I take the projected flow rate at full capacity as the sum of both the individual loops:
3636 lph - 3240 lph = 396 lph lost when combining the two loops

So, roughly 12% reduction when doing this, an acceptable loss in my opinion considering the y-connector may help dissolution of CO2 and also reduce the amount of hardware in the tank. Whether this is the case will have to wait a little longer while the rest of the tank comes together :oops:

(edit: I just went over my numbers again and noticed my fx5 is putting out more than expected. Foolishly I stopped the flow by turning off the power rather than removing the pipe straight away. Although that said, the water did stop pretty much completely when doing this so didn't think it would be a huge problem at the time. I should have done repeats :rolleyes: )
 
Back
Top