# Siphon Breaker Vent Pipe Location

1. May 15, 2014

### rollingstein

Siphon Breaker Vent Pipe Location (Sketch attached)

Say I need to maintain a roughly constant liquid level in a tank (T1 in attached image), one way is to have a U-loop on the drain pipe.

Problem is, this can get a siphon started in the geometry shown & entire T1 empties into T2. Not good.

One way to break a siphon is to use a vent pipe (V1 / V2 / V3) as a siphon breaker (I think).

Question: Does it matter if V1, V2 or V3 is the siphon breaker pipe location? I think all ought to work but wanted to check. Intuitively V2 seems best, but sometimes mechanical constraints might make V2 hard to implement.

Last edited: May 15, 2014
2. May 15, 2014

### Wes Tausend

...

Any except V1 will work in my opinion. The "atmospheric end" of the vacuum break must be higher than the maximum top of the loop riser level though. The downside near the V1 could conceivably drain T1 down to the level of the V1 tee, if V1 were used. T2 should also be vented above the loop riser level and probably the top level of T1.

Wes
...

3. May 16, 2014

### and7barton

I'd have thought that only V2 would be workable - Both V1 and V3 would drain down the upper tank. The top opening of vent V2 could be located at any height, provided it's attached to that pipe anywhere higher than the liquid level in T1.

4. May 16, 2014

### rollingstein

Can you elaborate on why you think V1 & V3 wouldn't work? I'm still confused.

Which path would V1 / V3 drain T1 via? Can you explain a bit more?

5. May 16, 2014

### rollingstein

Interesting. Can you explain why? Sounds non intuitive to me.

6. May 16, 2014

### and7barton

V3 would drain off T1 simply by gravity, not syphonage. The water would simply leak out of V3.
V1, if the water in T1 became high enough to run over the high loop, would not operate as a syphon breaker, but would probably leak water out of it, in addition to some water running down into T2.
V2 would act as a syphon breaker, but I would ensure that V2's nozzle was a few inches higher than the pipe loop to prevent any dribbles from it.
But why not simply route the pipe horizontally from the required level in T1, and down to T2, and NOT from the bottom of T1 ?

7. May 16, 2014

### rollingstein

I don't think it would. Maybe I need to clarify my diagram. The vent V3 is a "L" shaped pipe with the long vertical leg going above the liq. level in T1 (exactly as shown in the diagram). So by gravity water can only rise till T1's liq. level but it can never drain?

Right? Or you still think V3 & V1 can drain T1?

One ( quite stupid) reason: Existing nozzle is at that location! :)

A better reason: There's utility to maintaining a vapor seal between T1 & T2. The u-leg filled with water provides that.

Last edited: May 16, 2014
8. May 16, 2014

### and7barton

Ah. I see now what your diagram means. Those orange lines you've put in, I assumed they were just pointers to the position of the vents...... You mean, they are actually depicting the vent pipes themselves ?
If that's the case, then here we go again - I'd suggest that V2 and V3 would be fine to position syphon breakers. V1, I'd suggest, is not a good place, it's low down and I don't think it would prevent syphoning.

9. May 16, 2014

### rollingstein

Yes. Exactly. Maybe I shouldn't have used orange. My bad.

Maybe you are right. But I still don't see a fundamental explanation why V1 wouldn't work.

10. May 16, 2014

### and7barton

I think V1, being "over the hump" of the loop, and too low down, wouldn't prevent any syphoning, at least until the water level in T1 had lowered to be level with the tapping, at which point air would be sucked down V1 and break the syphoning. The latter situation might well apply to V3 too. Experiments need to be made.

11. May 16, 2014

### Travis_King

If you have the room, you could make the first vertical leg out from T1 greater than 33 feet in height. No siphon will form then. Otherwise, V2 is your best bet, then V3. I don't like V3 though because it will be filled with water most of the time. It should still work though.

12. May 16, 2014

### fleebell

I suggest connecting your vent line to that elbow between V2 and V3. But instead of an elbow use a T fitting there and connect the vent line to the top of the T. That should prevent siphoning from happening while allowing the over flow function to work without any problems.

Also as an extra measure you could put a slight restriction in the line out of Tank T1 smaller than the piping or simply use one size down piping from the tank to the T fitting the vent is connected to. If it has a valve, just don't open it more than 3/4 or so. That will prevent an overfilled tank from possibly trying to dump the excess faster than the overflow plumbing can handle it which could possibly cause a small amount of siphoning as the tank got to the correct level even though it is vented. (strange things due happen now and then) That restriction will pretty much guarantee it can't siphon.

Both horizontal lines in the drawing should also have a small drop in them toward tank T2 as to prevent liquid from just setting in the lines when an overflow does occur.

Last edited: May 16, 2014
13. May 19, 2014

### Wes Tausend

Sorry it took so long to get back. Life got in the way.

You can use a thought experiment to see each arrangement in a different light. The arrangement can be reduced to a simpler set-up to more easily see it work. You can re-draw your dipiction to look like each case of the below suggestions.

Excluding other vents and incorporating V1 as your vacuum break vent is equivalent to just T1 with a simple siphon hose curved over the edge of the rim. To be entirely equivalent, one end of the siphon must reach the bottom of the T1 tank and the other is over the rim, bent down and ending as open at the V1 tee junction.

The T1 rim can be lower than shown; slightly below the heighth of the top of the U-loop level if you wish, but it doesn't have to be. The hose-loop can be at the dispicted height of you U-loop or as high as you want (but less than Travis_King's 33 foot rule*) as long as it drops and only terminates at the level of the V1 tee. In each case it will siphon T1 liquid down to the end of the hose level, but no more. V1 won't work, I was right about V1 not working for you.

The same "thought" method can be used to determine what V3 will do. The volume of the V3 riser represents a small additional volume of T1 as though it were a U-shaped continuation of T1 itself. It can be re-drawn with the first U-loop upright leg going straight up and your present U-loop tapping the combined liquid volume at the low point of present V3 tee and going up a bit and over. The tee where V3 now connects now becomes the low siphon tap on the combined volume, T1 + upright. So it now appears to me that T1 will also be siphoned lower than you want by the long downside to T2 thereby siphoning T1 down to the height of the V3 tee junction. I was wrong that V3 would work for you. Poster and7barton was corrrect, at least in his first post here.

Only V2 will work. This arrangement is the same as a siphon hose, or a hole in the side of tank T1 simply terminating at the level you desire and somehow spilling into T2.

A good thought experiment is at least a third of every successful theory. A final third would be testing a model to proof, if the real project is labor intense. In this case you can skip the middle third, the math, altogether.

Wes
...

*After 33 feet, water will vaporise before it can experience a low enough vacuum to lift it further.