Fluid Engineer Explanation Please-inverted P-trap

[carlyai]

So I have 2 opened top 1000 L water tanks with the drains combining feeding into a pump, that recirculates the water through a filter back to the tank water inlets. A water flow fault caused one tank to partially empty and one tank to overflow.

Evidently I can fix the problem of a tank overflowing by installing in the tank drains an inverted P-trap at the height I want the tank water level to maintain at.

So 2 tanks and 2 inverted P- traps.

What's the theory of how this works? Why does the level in the P-trap stay at the same level as the height of the water in the tank?

 

[Surelynot]

It is definitely a P trap and not a P controller you are considering fitting?

[sometimewoodworker]

5 minutes ago, carlyai said:

So I have 2 opened top 1000 L water tanks with the drains combining feeding into a pump, that recirculates the water through a filter back to the tank water inlets. A water flow fault caused one tank to partially empty and one tank to overflow.

Evidently I can fix the problem of a tank overflowing by installing in the tank drains an inverted P-trap at the height I want the tank water level to maintain at.

So 2 tanks and 2 inverted P- traps.

What's the theory of how this works? Why does the level in the P-trap stay at the same level as the height of the water in the tank?

 

That doesn’t sound as if there’s enough information on the structure of the piping and where the outflow of the traps are going. It also sounds as if, depending on the size of the pipe in the trap, you could accidentally create an auto-siphon, and I don’t think that is what’s intended 

[VocalNeal]

So you are going to common together the tank overflows and then have the "real" overflow outlet at a slightly higher level?

 

Each tank will overflow into the other but if the level rises above the outlet then both tanks overflow simultaneously to drain?

[carlyai]

There is one of these P-trap builds for each tank and the top open end of the P- traps go to a common pipe to the pump which pumps the water through a filter back to the tanks.

I really want to know how this works.

I think it keeps the level of water of the fish tank to the same level as in the P-trap, so if the P-trap is below the level of the top of the tank the tank shouldn't overflow. 

But if one tank had a blockage in it's drain but was still having water pumped to it, it would still overflow regardless of the P-trap.

So it would probably be better off having 2 float switches so that the pump would be cut off before the tank overflowed.

[Crossy]

Can you sketch out (or use the construction drawings) what you have with some sense of height and scale.

 

A couple of floats and some relays would certainly be handy as an overflow backup.

[carlyai]

Think I've got a vid of my system now. Already a couple of mods coming for the sump.

Would just like to know why/how the inverted P trap will stop overflows.

I think I understand how the tank water level will be the same level as the P-trap. So if the P trap is lower than the top of the tank the water will not overflow from the tank.

But if something gets stuck in 1 tank drain or and water is still filling that tank it is still going to overflow.

I'll buy a couple more of those orange tank float pump cutout switches that you and sometimewoodworker put me onto.

We got a bag of fish from the local Mekong river hatchery. I wanted 60 all male tilapia of a reasonable size. No can do. So got a mixed bag of everything. Some of the fish were too small and got sucked into the drain causing one tank to keep filling and the other not to fill, so 2000 L into a 1000 L tank don't work.

Modifying the drain cover so small fish can't get sucked in to the drain pipes. Then modified P trap and pump cutout. Should be right to wind it up again.

 

[sometimewoodworker]

Unless the inflow is huge, it looks as if the only benefit of the inverted P is to keep junk from blocking the overflow. That is assuming that what you showed is all there is.

 

certainly a couple of  float switches can be set to cut off the pump if the level in either/both tanks goes too high & it’s not rocket science to design the circuit.

 

you may be able to use mosquito net and a jubilee clip as a fry protection filter 

[carlyai]

sometimewoodworker wrote 'Unless the inflow is huge, it looks as if the only benefit of the inverted P is to keep junk from blocking the overflow. That is assuming that what you showed is all there is.'

Am I right in assuming that the height I set the P trap to will be the height of the water in the tank?

 

[carlyai]

5 hours ago, VocalNeal said:

So you are going to common together the tank overflows and then have the "real" overflow outlet at a slightly higher level?

 

Each tank will overflow into the other but if the level rises above the outlet then both tanks overflow simultaneously to drain?

My setup vid. Later will just be changed to connect the P traps in the sump pipes if needed. I'm not sure I need them if I have float switch pump cutouts in the tanks.

[Crossy]

Be careful linking your "traps", you may inadvertently create a syphon which would drain the tank if it overflows, I assume your fish wouldn't like that.

 

[Artisi]

How about a sketch showing the system layout. 

[VocalNeal]

1 hour ago, Crossy said:

Be careful linking your "traps", you may inadvertently create a syphon which would drain the tank if it overflows, I assume your fish wouldn't like that.

 

 

Not if the first verticle elbow is a Tee?

[carlyai]

10 hours ago, Artisi said:

How about a sketch showing the system layout. 

I posted a video of the layout. Not good enough? 

The only difference to the layout in the video will be that both the drain pipes exit the bottom of the tanks then ascend vertically to the inverted P-traps (set at a level about 1000 mark) then descend to join the drain pipe.

The other pipe setup is the same.

Another modification will be in the sum to change the sump covers as they are hard plastic and warped during heating and don't deal properly.

 

So setup looks like vid but P-traps added.

[carlyai]

Soooo the way I see it is that the level of water in the bottom of the P-trap will be at the same height as the water level in the tank. (Must be someone's law.)

Without any pipe blockages, if the water flowing into one tank is more than the water flowing into the other tank the tank won't overflow as it will only reach the level of the P-trap and the drain will drain more, so tanks won't overflow. 

That's my main premise.

Stopping pipe/pump blockages due to small fish being sucked into the drains will be completed with drain cover modification and sourcing correct fishlings.

[Pogust]

If you let the air go trough the mesh that prevent fish and other things being sucked out you will avoid the problem. Combine the air nozzle with drain mesh and let air keep it clean. If you want the same level in both tanks, why not just put a T at each drain pipe and combine them? That will always keep the same level in tanks until you block one drain. That you can't with air cleaning it....

 

[carlyai]

37 minutes ago, Pogust said:

If you let the air go trough the mesh that prevent fish and other things being sucked out you will avoid the problem. Combine the air nozzle with drain mesh and let air keep it clean. If you want the same level in both tanks, why not just put a T at each drain pipe and combine them? That will always keep the same level in tanks until you block one drain. That you can't with air cleaning it....

 

I have a T at each drain point. 

I have a fix for the sump and fish getting sucked into the drain.

If the efficiency of the pump drops off and more water is pumped into the first tank than the second tank then my thinking is that without the P-traps the first tank will overflow, but with the P-traps the first tank will drain faster and not overflow.

With the P-traps each tank can only hold up to 1000 L.

There again I didn't listen to the science teacher at school, unfortunately.

[Pogust]

1 hour ago, carlyai said:

If the efficiency of the pump drops off and more water is pumped into the first tank than the second tank then my thinking is that without the P-traps the first tank will overflow, but with the P-traps the first tank will drain faster and not overflow.

May I suggest you have a valve/restriction on the water feed line from pump to tank (at each tank) so you have an over pressure there. If pump allows it. That way you will have the same flow to both tanks even if the pump slow down a bit. It looks to me you do it unnecessary complicated for yourself. On the other hand you probably have much fun doing it... ????

[carlyai]

2 minutes ago, Pogust said:

May I suggest you have a valve/restriction on the water feed line from pump to tank (at each tank) so you have an over pressure there. If pump allows it. That way you will have the same flow to both tanks even if the pump slow down a bit. It looks to me you do it unnecessary complicated for yourself. On the other hand you probably have much fun doing it... ????

Thanks. I have valves that can be adjust on each tank water inlets. Trouble is (when this gets going) it will be working 24/7 and sometimes we won't be there. Eventually I'll connect it all to Uprosessors and IoT but could be a while.

[carlyai]

Here's the architectural drawing. At the moment just pumping from the fish tanks thru the filter then back to the tanks.  

[carlyai]

There's also a pump between the drain and filter.

[Crossy]

I think you need a balance pipe between the tanks other than the pumped link.

 

How do the vertical overflow thingies in your video relate to the pumped stuff? If they're not too close to the pump inlet you may be able to use them as a balance with a simple link (and a valve so you can drain a tank).

 

 

[carlyai]

48 minutes ago, Crossy said:

I think you need a balance pipe between the tanks other than the pumped link.

 

How do the vertical overflow thingies in your video relate to the pumped stuff? If they're not too close to the pump inlet you may be able to use them as a balance with a simple link (and a valve so you can drain a tank).

 

 

The vertical P-traps will replace the drains as they are now. So out of the drain up to the vertical P-trap then back to the drain. I'll break the drains at the drain connectors, install the  P-trap vertical inverted U pipe then back to the drain. They form a closed loop: tank drain out, combined drain, pump in, pump out,  filter in, filter out, combined tank in, tank in.

Everything worked well when just water (before the fish). I presumed the tanks levels would stay the same as equal atmospheric pressure on both.

A problem arose when I introduced too small fish into the system. (They were the only ones available at the government farm at the time.) I introduced them and they got stuck in the drain cover and partially blocked up the drains.

I've got a modification for the drains so this shouldn't happen again.

I think if no one can say these inverted P traps won't work, then I'll give it a go.

 

[Crossy]

I still have zero idea what these "inverted P traps" are or how they're supposed to allow the tanks to balance when they are on the pump suction.

 

But I'm just an electrical/electronic engineer.

 

[sometimewoodworker]

8 minutes ago, Crossy said:

I still have zero idea what these "inverted P traps"

Fancy overflow pipes, that if fed with too much flow can auto-siphon and so empty the tank. ???? 

[carlyai]

4 minutes ago, Crossy said:

I still have zero idea what these "inverted P traps" are or how they're supposed to allow the tanks to balance when they are on the pump suction.

 

But I'm just an electrical/electronic engineer.

 

Sorry.

That was really my original question.

I'll try again.

If I have a tank marked from 0 to 10 and have a drain inverted P-trap at the 10 level, then the tank and P-trap will fill to the 10 level.

If I have the P-trap at the 5 level, then the tanks and P-trap will fill to the 5 level. Even if more water is flowing into the tank as the drain will drain faster and the tank water can only get up to the height of the P-trap water.

So this was my question.

No one seems to have answered it.

This is assuming no blockages anywhere.

 

[carlyai]

2 minutes ago, sometimewoodworker said:

Fancy overflow pipes, that if fed with too much flow can auto-siphon and so empty the tank. ???? 

Ok, got that. But what is too much flow?

So do you agree with my last post or not?

[carlyai]

May as well make an executive decision and try it. There are 2 outcomes. It works or it doesn't. If it doesn't work it can be a talking point. ????

Thanks for all help.

[sometimewoodworker]

11 minutes ago, carlyai said:

15 minutes ago, sometimewoodworker said:

Fancy overflow pipes, that if fed with too much flow can auto-siphon and so empty the tank. ???? 

Ok, got that. But what is too much flow?

So do you agree with my last post or not?

You have the physics correct. 
 

Too much flow is the amount that will cause the air in the down pipe to be sucked into the outflow. This will change an overflow system to become an auto-siphon so emptying the tank. Changing the clean out screwed fitting at the very top to a mesh filter that will allow air in but keep crunches out will make the auto-siphon possibility much less and would probably eliminate much more than a 10% loss of volume 

Edited October 25, 2020 by sometimewoodworker

[sometimewoodworker]

Just now, sometimewoodworker said:

You have the physics correct. 
 

Too much flow is the amount that will cause the air in the down pipe to be sucked into the outflow. This will change an overflow system to become an auto-siphon so emptying the tank. Changing the clean out screwed fitting at the very top to a mesh filter that will allow air in but keep crunches out will make the auto-siphon possibility much less and would probably eliminate much mor than a 10% loss of volume 

is an example