How does a backflow preventer actually work?

Jean-David Beyer

The supply to my hot water heating system includes, among other things, a backflow preventer. I know what it is supposed to do, but I do not know how it really works. I know it contains two check valves in series with a vent in between, but that cannot be the whole story.



Imagine I take a T and put a check valve on each end both facing the same way. That meets my description above, but as long as there is a supply, water would come out the vent. So there must be something like a vacuum breaker in there too. But not one I can buy at a plumbing supply store. I can invent a mechanism that would notice the differential pressure between the supply and delivery and open the vent if the supply pressure is less than the delivery pressure, but it would be too expensive, and might fail just when it is needed. So the actual design must be simple and cheap to make.



Can anyone post a cutaway view of one of these things?

Unknown

Backflow

I know what you`re looking for, and found a site that shows-it. well.Unfortunately it`s only in PDF form and I`m not sure how to convert it, as I thought PDFs will not post here,,,, sorry.

Jean-David Beyer

Is it the one of the Watts 9D-M2?

If so, I have that one, but I have trouble figuring it out.



Starting at the top, where it says "Union Outlet" is my first problem, because that is the inlet to the valve. The water hits the stainless strainer screen, flows through the primary check valve, and then overflows out the vent connection -- I do not see how they prevent that. But, assuming they do, the water then pushes the U-shaped piston down and that can open the secondary check valve, but unless there are unmarked holes in the sides of that piston, I do not see how the water gets through the now-open secondary check valve. Likewise, there must be holes in the metal piece that holds the spring of the secondary check valve.



Here is a puzzle also. Note the legend, "Vent and drain connections"? Now it seems to me that the vent and the drain connection are one and the same. But I could be mistaken here. It sure looks that way when I look at the valve. But there are those 8 tiny circles there, and they could be ports to somewhere. But where, and what for?





So I still do not see how they keep the water from coming out the vent.

Unknown

I agree with your confusion,,,

and wish I could "Doctor" the PDF to show,,,, but I can`t.



What a bummer!



Tough to get used to this New Wall,,, that`s  for sure.

Jean-David Beyer

Do I have to cough up $25, buy one, and take it apart?

I guess I could do that, but I hate to spend the money to do it.

Unknown

JD,,,

I WISH they were only $25 bucks!!I can take one apart & take pics,,,, if your interested?That-is,,,,," IF" they`ll post here ;-)

Jean-David Beyer

We cannot talk prices here, but if you search the Internet...

That would be really nice if you would do that. You do not need to take apart one that is still working, of course.

Unknown

I never do, but you mentioned it first!

I have understood the "prices"  policy here for a long, long time. However,,, you mentioned-it first.

Jean-David Beyer

you mentioned-it first

I hope I do not get in trouble for that. It was not an offer to buy or sell, just a number I found on the Internet, and I did not name the vendor.

EricAune

Backflow devise

The devise you are talking about is called a "Dual check with intermediate atmospheric vent".  More commonly known as a Watts 9D.  There are other manufacturers, but Watts is pretty good at selling these things apparently. 



The reason it is not a good idea to "take a couple of tees and a check" and build one as you have stated is yours will not be certified, nor will it work properly, you will have no way to test it when it ultimately fails...because it will. 



The design concept is related to a RPZ in that it allows for continuous line pressure and back pressure, but is limited to low hazard and small pipe installations.



Check number one (the check on the boiler side of the vent) is a standard spring check rated for continuous line pressure.  When this check fails the vent allows for drainage of any fluid.  The second check valve (on the inlet side of the vent) is a back-up only in the event that the vent is plugged or will not allow for complete flow. 



Understanding how these devises work is just as important to understanding when to use them......I would hope that anyone installing a h2o feed/fill line would have a good working knowledge of the true hazards involved. 



One should not be overly concerned with the price to purchase the devise, BUT, more concerned about the cost of failure. 



Peace

Jean-David Beyer

I agree with what you say.

I would not really consider making an anti-backflow device by using two checkvalves on a T. But that is all the explanation I could get from the Internet on how they work, and the explanation is clearly inadequate -- even ignoring that it would never be certified as anything other than as a piece of modern sculpture. ;-)



The main fault of the 2 checkvalves on a T is that water would pour out the middle the entire time the supply was turned on. And looking at the diagram of a Watts 9D-M2 did not enlighten me. I could see pretty much how it was supposed to work, but I cannot see how they turn off the vent in such a way that it is closed in normal operatioin, but open if the supply pressure gets lower than the delivery pressure. The only possibility is that they might have a way to move the cylinder that encloses the U shaped piston up until it hits the bottom of the assembly that holds the primary check valve, and guarantees that it moves back when the vent port needs to be open. But I not only see no such mechanism, but from the drawing, it appears that the cylinder cannot move.



If you look at their fancier valves, there is a third item that basically opens and closes the vent port, but I cannot find it in the 9D-M2. My valve is a different brand, but it looks just like it. I doubt it if my contractor would like me to take it apart to see how it actually works.

Jean-David Beyer

Another explanation for these valves.

I found a better explanation here; I am sorry the link is so ugly.



http://www.google.com/url?sa=t&source=web&ct=res&cd=1&url=http%3A%2F%2Fwww.michigan.gov%2Fdocuments%2FMDA_Plumbing-n-CrossConnectionControl_27629_7.pdf&ei=RmOJSsasJt-BtgeG3NjnDA&rct=j&q=ASSE+1012&usg=AFQjCNGjqjqB5-tyhLFkaC98NJ9QCrJQjQ&sig2=Dlj_UoOJ3Ex78_DbXMfSKA



The explanation starts on page 7.



It seems as though what appeared to be holes or ports in the Watts diagram are actually a weak coil spring. It is tough to describe how it really works, but if you look very closely at their figure, and read the text, you can probably figure it out.

hot_rod

backflow

Let's see if I can post some pictures. I'd be glad to send you a "returned" backflow. Many of the ones we sell in Europe go with the adapter to install a vent pipe with an air gap, as all backflows spit from time to time. It depends on the pressure inside those checks. I'll try an exploded view also.



hr

hot_rod

actual valve "exploded"

these get returned from time to time. 99% of the time they have crud under the check surface. Screens help, but the smallest speck of crud will cause them to drip.



I like to take things apart to see what makes them tick. With this valve disassembled you see how the checks and different spring tensions work together.



hr

Jean-David Beyer

It looks to me as though your valve is better.

It looks to me as though your valve is better than the one I figured out this morning. In yours, the check valve assembly on the input moves when there is net pressure in the valve, closing off the vent port. In the one I have, they use a rubber diaphragm to do that and I would guess it is more likely to fail; though not often. Execution of the design may be a more important factor than the design itself, but who knows. It also looks as though your valve has better flow (less resistance) than mine, though other than initial purging, that probably does not matter. They had no trouble purging my system when they put it in.



It would be fun to have one of them, but I do not wish to put you to the trouble and expense of sending it to me.

hot_rod

a quality

valve, as you can tell by the design and you will see in the machine work and brass castings.



The offer stands if you want this returned one for observation. Or a paper weight



hr

Jean-David Beyer

I would be delighted.

OK, I accept your offer.

I will e-mail you my address, etc., off-line.

Jean-David Beyer

Thank you.

Thank you so much for that returned backflow preventer valve. Nothing like taking something apart with the bare hands to see how it works.  It is so well made that I wonder if Caleffi make rebuild kits for it. If they do, it seems it would be easy to rebuild it if the failure were due to crud inside the mechanism. I suppose it is a question of the relative cost of labor vs. the cost of a new unit. But it seems a shame to toss a defective one into a landfill, or even the melting pot.



I know that Lawler make rebuild kits for their valves, at least the 9700 series one I use in my darkroom. I rebuilt that one about a year ago. It filled up with deposits because an O-ring started leaking and I did not fix it right away. It just dripped in my sink, so it did not seem urgent. I won't make that mistake again.