Would this work???

[RobboT]

Sorry to be a party pooper but you really shouldnt need the airline to break the siphon. Think of your toilet. With the trap installed as you posted above in post #13 (i.e., just like every trap on any drain in your house) the siphon will not draw all the water out of the piping.

HOWEVER . . . the setup you you put in post #11 is your only real chance of having a "self-priming" effect on the pump IMO. With the setup in post #13 once the skimmer empties and draws air into the influent line water will sit in the pipe to about the level of the elbow just outside the pump intake. There will be very little water sitting actually on the pump head itself. You may actually get the pump restarted by luck but I'd bet not.

If you did what you show in post #13 but instead of heading down just go up at 45deg angle then turn down you'll leave plenty of water directly on the pump head. I have a pic I drew but it exceeds all the attachment sizes.

 

[RobboT]

BTW this whole concept will only work if the venturi for the skimmer is on the effluent side of the pump. If your drawing in air on the intake side I doubt you will get enough suction to draw water up into the pump. You might get some but very little if anything.

The simplest solution to this whole thing is a PVC Check valve on the intake

 

[NateHanson]

Good point about the venturi Rob, but I disagree with the toilet-trap example. The toilet isn't sucked dry, but it is sucked down to the top of the bottom of the trap (the top of the hole you see in the toilet). When using 1" PVC, that would not leave any water for the pump to start. The toilet then gets filled as the tank continues to fill. Try flushing the toilet with an empty tank. Very little water is left in the bowl.

Another reason there's water left in the toilet is because all the toilet plumbing is about 3" wide inside the trap, and 4" wide directly after the trap. Even that 4" pipe is vented through the roof. A siphon will persist much longer in unvented 1" pipe than in the open flow that exists in your soil stack.

All that to say, without the air hose the trap will empty almost entirely.

Now, as to the question of the remaining water level with the siphon-break air hose - The water level in the skimmer and in the pump (I expect) will be roughly horizontal as it drops, so when air is in the very top of the tube (and the top of the pump housing) the siphon will be broken, leaving the plumbing, pump head, and skimmer body with a water level that's more than half-way covering the pump head. So I'd guess that it would be plenty of water to start the pump.

Those are my assumptions anyways.


(Now you HAVE to try this, to see if it really works, and vindicate, or banish, my theories!)

 

[RobboT]

True about the volume of water having an impact on the final level I had always thought this was a specific thing to the trap design of toilets though. I thought it was the reverse part of the typical double reverse toilet trap was why it leaves a decent amount of water in the bowl when its done. That's why I suggested the reverse trap. I'm no expert in commode physics though

Even if steve doesnt try this now I think Im going to have to

 

[RobboT]

Now, as to the question of the remaining water level with the siphon-break air hose - The water level in the skimmer and in the pump (I expect) will be roughly horizontal as it drops, so when air is in the very top of the tube (and the top of the pump housing) the siphon will be broken, leaving the plumbing, pump head, and skimmer body with a water level that's more than half-way covering the pump head. So I'd guess that it would be plenty of water to start the pump.

On that note it would be prudent to either make sure the horizontal section after the pump intake slopes toward the pump or is level. Any slope away from the intake will leave it dry.

This is sounding more and more like a pending science experiment in the basement this weekend . . .

 

[NateHanson]

I'm no expert in commode physics though

C'mon, now's no time for false modesty from the Porcelain Legend himself.

 

[RobboT]

Hehe I may be considered an expert in talking s---

 

[NateHanson]

When the siphon breaks, all that water that's to the left of the top curve of the trap will flow back down towards the pump, so it might be helpful to place that curve higher. That way, even if the pump is half-dry when air enters the anti-siphon hose, the water that flows back towards the pump after the siphon breaks will fill the pump!


I'M A GENIUS!!!! A-hem. Well, maybe we'd better see if it works first.

 

[sdesi2005]

I agree with Rob I think

If I start with a reverse trap, using say a 45 that will almost guarentee that there is at least the amount of water in the pipe heading up in the pump head. That would definitely start the pump, I assume.

What about the air pocket left from the broken siphon? Would that stall the pump? Or will it quickly be replaced with water?

 

[RobboT]

In either case you should try to minimize the length of piping from the trap to the tank. What probably will happen is you'll get a surge and release action for a bit until the air pocket slugs through.

sorry that was a little Yoda-speak . . . reverse the sentence order there and you get the answer to your question about teh air pocket

 

[NateHanson]

I agree with Rob I think

If I start with a reverse trap, using say a 45 that will almost guarentee that there is at least the amount of water in the pipe heading up in the pump head.

That'll only be true if the siphon doesn't keep going until that water makes it over the other side. Probably not going to happen, but I don't know until you do the experiment.

 

[sdesi2005]

I fixed the problem It was probably the easiest way to make this work and I totally over looked it.

Take a look at this:

 

[sdesi2005]

When the power goes out the most of the water will siphon out, to ensure there is enough water to restart the pump, I drilled a hole at the top of the trap and siliconed a piece of hose into it being fed from a small mini jet pump in the tank.

Therefore, even if all the water is siphoned out to the trap, as soon as the power turns back on, water will begin to fill the trap priming the pump. What do you think?

 

[NateHanson]

Cool. I like it. I think it should work.