Leaking PVC Header in Curling Club Brine System 3

[PRPA]

Quick bit of background. I volunteer at the local curling club, helping out with the ice and ice equipment. We run an Ammonia-Brine system and four years ago built new supply and return headers for the brine distribution. The original equipment, circa 1960 was all steel, but we replaced it will PVC as it is more resistant to brine running through it.

The setup is pretty simple. The supply and return headers are both an 80' length of 4" SCH80 PVC that has been drilled and tapped every 6" along the length. Each of those holes has 1" hose barb cemented into it. The system worked great the first couple of years, however this year we noticed brine starting to disappear from the make-up tank. After some investigation, we've found that four of these hose barbs have developed leaks that need to be fixed.

The problem is that correcting the problem with hose barbs that have been cemented is not so easy. Cutting/drilling out the old one requires the brine from the whole system to be drained and refilled once the job is complete. That's a major undertaking for a small group of volunteers.

This is where I'm looking for some advice/input from fellow engineers. Since the leaks are all very small and system is fairly low pressure (20-22psi), is there another solution? Like building up some epoxy around the hose barb connection? I'm thinking about a product like "JB-Weld Waterweld" or some similar putty type epoxy. Does anyone have experience with these products, especially in regards to being used in an environment where it would be exposed to cold conditions, ice build up and contact with brine?

Any input would be greatly appreciated.

 

[MikeHalloran]

Can you drag a couple of "basketballs" on a rope through the header and inflate them to isolate some of it? Clean and dry works best for PVC.



Mike Halloran
Pembroke Pines, FL, USA

 

[JavaMoose]

A previous employee at the place I work at used water-weld to "fix" some Sch80 PVC fittings on our equipment. It's lasted about a year and is now leaking again. We run sea-water through our systems, so I think it's a good indication of what that "fix" will end up doing on your system.

Might I recommend using some saddle taps to fix the leaky fittings (and other fitting as they start to leak)? This will give you a tapped connection to install some thread-to-barb fittings. Check the links below:

http://www.lascofittings.com/defaul...evStrSearch=Submit|Search||txtKeywords|saddle

http://www.conind.com/cont/pdf/Lite...-3000-00 = CONNECTOR CATALOG (WATER FTGS).PDF

http://www.plumbingsupply.com/hot-wet-taps-for-pvc-pipe.html

 

[MikeHalloran]

Amen on saddle taps of some kind. At least they don't rely solely on an adhesive bond for both retention and sealing.



Mike Halloran
Pembroke Pines, FL, USA

 

[PRPA]

Thanks for the input guys. The idea of using a saddle has been kicked around (in fact, it was my first suggestion to the group).

The hesitation lies in tha you still have to do something with the brine. Removing the 1" distribution line from the leaking hose barb would still need to be plugged or the system needs to be drained to install the saddle. That's where the suggestion of putting something over the hose barb in place came from. The fact that "water-weld" didn't hold up to sea-water is very good to know.

As for keeping the area clean and dry, with the pump off and the leak is non-existant. Since it's now the off-season, the pump is off from now until September.

 

[JavaMoose]

So, when the pump is shut down, the headers don't drain enough to permit removing the welded barbs and saddle?

I guess I would have to see a picture/diagram of the set-up, but even if they don't gravity drain, can't you stop the pump and drain through the existing barb fittings (at least drain enough so that there is minimal spillage when removing and saddle)?

 

[PRPA]

I should probably have snapped off a couple of pictures when I at the Curling Rink last night. Maybe I'll do that tonight.

The problem with the brine has little to do with the brine in the header. It's the miles of distribution tubes in the ice surface. Each one of the hose barbs on the supply header is connected to a 1" PE tube that runs the length of the ice surface, doubles back and then connects to one of the hose barbs on the return header. I haven't measured any of the lines, but based on the dimensions of the ice surface, from supply header to return each line is approximately 305'. That is a lot of brine stored in the ice surface distribution. Coupled with the fact that the headers sit some 3' below the ice surface, you have to drain off a lot of brine before you can do any work.

 

[JavaMoose]

I would disagree, given your explanation, the brine in the header is the problem. If each loop is 305' of 1" tube - that is only 7 gallons per run. If the header was above the level of the tubes, you could just simply disconnect one end of the run and drain (assuming an empty header) into a 10 gallon bucket. Of course, this doesn't solve your problem since your headers are below the tube level...

Do you have clean-outs on the end(s) of the headers or are the capped and welded?

 

[PRPA]

So, I wasn't able to get some photos, but I did throw together a simple sketch of the system.

The ends of the headers were finished with welded caps. We installed 1" ball valves at the end of both runs for draining the system. There are 4" isolation valves between the pump/chiller and the ice surface, as well as some smaller valves used to bleed air from the system (I've shown them at the tail of the header, but I think there are actually three sets: head, tail and mid-span).

It seems, at least from my experince, that any option involving a distribution tube being re-moved and re-installed means draining the system. We wanted to avoid this, if possible, since it's a big task for a group of volunteers. Maybe it was just wishful thinking that there would be something out there that would accomodate that thinking.

It does seem that the correct thing to do is bite the bullet, drain ths system and fix the four leaky fittings properly. Thanks a bunch for your input.

 

[donf]

Check this link for a product called "Stopit".

http://www.indumar.com/

Don't know if it's recommended for use with brine.

Godd luck.

donf

 

[btrueblood]

"the correct thing to do is bite the bullet, drain ths system and fix the four leaky fittings properly. "

I'd say you'd be better off fixing all of the fittings properly, leaking or not. Or else plan on fixing some more leaks next year, and the year after that...

 

[PRPA]

It's easy to say "...you'd be better off fixing all...", but the reality is its not going to happen. And you are correct that the liklihood is that if a couple are leaking now, more are going to follow.

That being said, as a club we don't have the money (~$35k, I've made some enquiries) to replace every one of those fittings with a saddle and hose barb. It would be nice if we did, but the fact is between the building and equipment there are lots of maintenance tasks and not a lot of money (or man-hours). We've worked hard to get our heads back above water and it would be a tough sell to convince the membership to spend that kind of money on something that is not critical at this point. The bottom line, regardless of the decision, we'll have to live with fixing the leaking fittings and leaving the rest. A bunch of preventitive maintenance isn't an option at this point in time.

 

[Cuemaster64]

PRPA,

Groovlok saddles ((U-Bolt Style 7045, with FPT connection, for new barb)

http://www.anvilintl.com/ProductSearch/ProductInfo.aspx?ProductId=4576).

Procure new Barbs (possibly reducing) for fitting your line size.

Sequence of repair.

Open all possible drains to reduce the amount of liquid in all lines. Cut off affected barb hoses and clamp off temporarily. Cut off or remove old barbs. Use a jig to drill out new holes to accomodate branch holes. Strap on your new saddles and reinstall your hose.

As always this is only a suggestion.

Regards,

Eric

 

[btrueblood]

$35k?! Wow, how many branches are there?

How about cutting in socket weld tee fittings, with reducers? Still spendy, and more labor, probably.

 

[PRPA]

The headers span across six 15' sheets of ice, making each of them some 90'. On both the supply and return side there is a hose barb approximately every 6", or 180 hose barbs on each header (360 in total).

Based on the suppliers I talked to a 4" x 1-1/2" saddle complete with a reducer and new 1" hose barb runs somewhere about $90 per connection. Factor in some thread selant and the liklihood of needing some spare hose clamps and distribution tubing and it's a ~$35k bill.

 

[JavaMoose]

Is it only leaking (now, at least) on the supply header? Does it still leak with the pump off? The reason I ask, if you turn the pump off (no pressure) and it stops leaking, then you can dry it. Try the water weld, worst case it performs like it did for us and only works for a year (FYI we were outdoors, with it exposed to sunlight and average 85*F temp year round). If it only lasts a year, maybe that will give time for a fundraiser to replace the headers? Maybe not, but at least it buys you a year.

 

[PRPA]

The four locations only leak when the pump is running. With everything shutdown for the summer, there is no leak. The fact that for the next couple of months the pipe can be cleaned and dried is one of reason we wanted to eplore alternatives to fixing the problem.

All four leak locations are indeed on the supply header. And all four leak almost identically, although one is a bigger leak than the other three. Personally, I think that the fact that water weld didn't hold up to seawater is a red flag for me. Seawater is a weaker saline solution than we are dealing with, so it would seem to be a good comparable to me. At this stage, my preference is to drop in a saddle to replace the leaking fitting. It's a pain, but it should eliminate the problem.

After that, we put together a plan to replace a portion of the good fittings with saddles and repeat that replacement over a couple of years.

 

[JavaMoose]

Great, glad to see you went with the "proper" solution. Hope everything works out well.

 

[JavaMoose]

Hey PRPA,
I came across these the other day when looking for some odd PVC and thought of you and your brine distribution system. Something you might want to bookmark/keep in your back product for the future. Looks like the largest they list is 2" main, but it might be worth checking with the mfg. if you ever do replace the entire system.
Java

 

[vpl]

Java, Were you going to attach something? Your post reads like you found an intersting link, but there's nothing that gives a clue as to the company.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.