Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

Your thoughts about splicing 3/8" pex?


I acquired a building that had frozen (and damaged) in-slab radiant heating. The previous owner has several zones and everything was done in 3/8" PEX. I've discovered and excavated the frozen (burst) lines in the slab. So now I am at the point where I need to join the blown-out areas that I cut out, then I will patch with concrete. What is the best splicing system out there?

My concern is that the radiant PEX is 3/8" , and, most systems, like SharkBite straight connectors basically have a 1/4" hole that creates an obvious restriction as opposed to the 3/8". There are some circuits where there could be 2 or 3 repaired areas on the same circuit. So any system that will help me preserve the flow would be best. What system would you use?

Thanks in advance.

Master Mark


  • icesailor
    icesailor Member Posts: 7,265
    Frozen -up:

    May I first ask how the 3/8" PEX allegedly "froze" in or under the slab? It is my experience that when the ground freezes, it pushes "out" and the water filled pipe or tube is the last to freeze, pushing the pipe wall outward against resistance from the frozen ground. If the 3/8" tube is in the poured slab, the concrete will still do the same and support the tube throughout the entire circumference of the tube. The only way the tube can freeze and split is if there is a void in the concrete. But, unless you live in a really cold area like Siberia or Alaska, the ground heat will usually protect the slab and the tube that is in it.

    Are you sure that you are not dealing with defective tube? Or PEX that was overheated?

    I have found lately that with water services put in in the late 70's. early '80's, and 100# pipe was used, there isn't enough wall thickness to support vertical ground pressures and the pipe is splitting on the sides. Also, if one causes the slightest kink in the tube when laying it out, it will split at sometime. Sooner or later.

    Call me a skeptic, but I would be looking into the mystery of the frozen PEX a little more before I got too far into something that may have no end.

  • hot_rod
    hot_rod Member Posts: 21,872

    yes every coupler you add will cause additional pressure drop. Do you know how long the loops are? The RPA guideline suggests 200 foot length with .3 gpm flow rate.

    Some brands, like Wirsbo/ Uponor use an expander and their coupling is a bit larger od.

    I like the copper type crimp couplings, they seem to have a larger od.

    A few couplings should not make a huge difference, you could calculate the flow rate and velocity and maybe bump the pump speed up. Excessive velocity will cause erosion wear and noise.

    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • McMaster
    McMaster Member Posts: 28
    edited September 2012
    How did the Pex manage to burst?

    The building was unoccupied for a couple years. You are correct that the concrete should prevent the tubing from freezing, and for the most part did. But, as I have found out, there are always unknown anomolies in every installation. As the pex is uncoiled and laid out, zip tied to rebar, etc, it's not perfectly even and does not lay perfectly level with the foam board. Guys could step on it while spreading out the concrete, etc. All locations in my situation where the PEX burst, breaks down like this: 

      (1) Where the PEX was NOT completely encased in concrete, as you mentioned, such as touching the foam insulation on the SIDE of the slab next to the foundation (where it could build up enough pressure to essentually 'blow-out' (towards the side touching the foam), or areas where the pex exited the floor (these were the 'easy' ones to fix). 

      (2) In the floor where the pex laid so low in the concrete that it was touching the foam insulation and there wasn't any concrete between the foam and the pex to prevent a blow out. This was the most common occurance. With virtually no concrete on the lower side of the pex, enough pressure could cause it to bubble and burst downward, which it did. Any leaks would spread below the concrete on the foam. These were harder to pinpoint. I wonder how a floor with the pre-formed insulating sheets where you 'snap' in your pex would have fared? I bet the whole floor would be junk because it would all be touching the foam board.

      (3) At an intersection where the PEX was tied to the re-bar AND an air pocket in the concrete existed at that point. There are a couple instances of this happening that suprised me as well. The concrete work seemed to be well done, but I suppose there are times when a couple stones around the tubing and re-bar allow a small air pocket to exist. In these cases, the holes were much smaller, but still existed.  

    This is a building where there is a lot of curtain wall glass that goes from ceiling to floor. Much greater chance of cold air coming off the glass as it hits the floor. It is located in the great lakes area and we get a couple days of -20*F every year. My guess is that either the LP bill was not paid and the tank ran empty (the tank was empty when we took posession), and/or the boiler failed to fire and nobody was around to report the problem. Amazingly, not one single radiant circuit on the second floor (which is in gypcrete) burst. Not even the tubing exposed to the air (on the 2nd fl).
  • McMaster
    McMaster Member Posts: 28
    edited September 2012
    Back to joining the PEX

    Back to the question about the 'preferred' coupling system to maintain best flow, My current thinking is:

    1) Create and use custom couplers from thin wall, large ID stainless steel and crimp them in place. Basically I would machine 'ribs' on sections of seamless stainless steel tubing that is 3/8" OD, but much larger ID than the standard brass couplers. Due to the slightly larger tubing size, I would accomodate by using polybutyl tubing style copper crimp connectors (the bright copper colored ones for PB tubing). These have a slightly thinner wall and slightly larger ID. They should create a nice hold onto the tubing but not over-crimp the tubing that I feel a standard ring would do. Where my repair gap is a few inches, I am thinking I could use a single piece of stainless to bridge the gap and not use PEX in that area. This would 'halve' the number of connections v.s. splicing in tubing and therefore reduce my connection points. How long I can go, I don't know. I haven't figured out what the expansion of, say, a 12" piece of S/S at 180*F would do to a section of concrete and what expansion/contraction might do to my PEX splice joints. I am concerned that too much expansion/contraction over time might ruin the joints. So maybe shorter couplers of 1 1/2" or less would be better. I could leave a slight gap of, say, 1" on either side of the joint by wraping the PEX in that area with a layer or two of cloth or another material to allow for some compression in the tubing rather than at the joint should the S/S coupler expand. Anyway, using S/S would yield a strong and corrosion free connection resistant to abrasion. .

    2) Another direction would be to use expansion type fittings that "hopefully" would be larger in diameter. I haven't zereo'd on on what the ID would be on these and I don't currently have an expansion tool. I'm not sure if one works with all, or just certain types of PEX. I'm a little hesitant to burry these type of connections in concrete that I haven't used in the past. I feel confident about copper compression rings.

    3) A possible 3rd direction that I have considered is a hybrid connection using a larger S/S coupler with a 3/8" ID with machined ribs and expand the PEX as required to fit over the larger coupler, then crimp with a copper ring. This would virtually eliminate any flow issues. I have a lathe that will allow whatever machinging is necessary, but I don't have the expander tool, and, frankly, I'd rather not machine out new jaws for the compression tool if I don't have to use custom copper compression rings. So I'm kicking around this idea at the moment. But, if the flow will be acceptable with the s/s tubing I currently have (.305" ID) compared to a .365 PEX tubing ID that makes life easier. The tubing is similar in size to 3/8" brake line. I was going to run some water flow tests with a machined S/S coupler and see how it compared to regular 3/8" pex, but I have not done so yet. I can already tell there is a bit of resistance just by blowing through it, but it is certainly a lot less than the standard SharkBite couplers that have a .235" ID. I was also going to do some heated tests with a customer coupler and see what the pull-off would be at 180*f. I haven't done this yet either. 

    I know this is all off-the-wall thinking and I'm sure every contractor reading this is right now is probably rolling your eyes back, but my main concern is maintaining a balanced system by not increasing flow restrictions while applying a coupling system that will not leak. I'm actually not that concerned about the leak part, I think that is pretty easy to nail down, mostly the flow. I'd like to stay away from talk about insurance liability because right now, flow is the primary consideration.

    In the lower level of this building there are appx. 15 zones on two separate manifolds (yeah, I know it's a LOT). There are multiple zones for each room. Each circuit is pretty short, I'd guess between around 75' -90' each give or take. About 1/2 of the circuits required repairs, some circuits had up to 3 sections that need to be spliced/repaired. 

    I'd love to hear your feedback / ideas. Thanks
  • Gordan
    Gordan Member Posts: 891
    PEX A can be expanded

    The others can't.

    Are you sure you want to homebrew something for a customer? Liability seems like it could be a concern.
  • McMaster
    McMaster Member Posts: 28
    edited September 2012
    Re: Homebrew

    Thanks, I'll take a look. Off the top of my head, I seem to recall it being PEX B but it's been a while since I looked.

    It's my place, not a customers. I don't look at it as "Homebrew", rather improving upon an already existing connection concepts. I am hoping you guys will offer thoughts on my ideas, or suggest something better ...based upon sound reasoning, or experience.
  • SWEI
    SWEI Member Posts: 7,356
    PEX standards
    1. ASTM F 1807: Specification for Metal Insert Fittings Employing a

      Copper Crimp Ring. These fittings are intended to be used in 100

      psi cold and hot water systems up to 180*F. Requirements for

      materials, dimensions, and markings used on the fittings are also

      included in this specification.
    2. ASTM F 1960: Specification for Cold Expansion Fittings with PEX

      Reinforced Rings. The same standards apply to these fittings as F

    3. ASTM F 2159: Standards Specification for Plastic Insert

      Fittings Employing a Copper Crimp Ring. These standards establish

      requirements for sulfone plastic insert fittings and copper crimp

      rings for PEX tubing. All other standards are the same as F 1807.
    4. ASTM F 2080: Standards Specification for Cold-Expansion

      Fittings with Metal Compression-Sleeves. These fittings and metal

      compression sleeves are intended to be used in residential and

      commercial, hot and cold, potable water distribution systems. All

      other standards are the same as F 1807.
  • Zman
    Zman Member Posts: 7,549
    edited December 2013

    "If you can't explain it simply, you don't understand it well enough"
    Albert Einstein
  • Steamfitter66
    Steamfitter66 Member Posts: 117
    Unless your

    making 20 repairs per circuit its not going to make any practical difference. I would use the best available for the type of tube you have.

    I dont know why anyone would put 3/8 in a slab unless it was a cap. Even the I would go with a 1/8 to 1/4" more concrete.
  • Weezbo
    Weezbo Member Posts: 6,232
    i have quite a few 3/8Ths wirsbo fittings .

    adaptors , couplings mostly.

    i question the price of what you are considering on the over all scheme of things both in locating the breaks and the materials ...

    seldom if ever are the breaks one pinner leak that can be cut square and rejoined.

    i have repaired frozen slabs . sometimes finding as many as 7 or 8 breaks where i thought there only to be one.

    care to enlarge on your concept a bit more?

    because i hate to send you off without considering costs up front , i feel it to be "Reactive ,Homeowner, cobble that does not allow for possibility of there being a better way to accomplish a task at hallf the cost .

  • Weezbo
    Weezbo Member Posts: 6,232
    ok, well ...

    i have work to do . so i am outta here for now .

    i am not inclined to give you this however , let me suggest that there is at least one aspect beyond the immediate , i too must consider, when i go to do my work today .


    decades ago , Rehau had 3/8ths on the Market and installed here in America . before them Stadler ...

    this tubing was in metric though.

    Hauser pipe may have made it in metric b pex too although i do not know that .

    not all pex is made by two companies . there are of ten years back over 258 different pex tubing and close to that amount of various connection systems .

    maybe re inventing the wheel is not what you should be "After' at this juncture , you asked for some answers well, ask . there is a great deal of experience here on the wall and off the wall too.

    there may even be some things in the "Off the wall" sections that do not stray from or are exactly the answer you need to hear . You may disagree thats ok it is up to you to do your own Thinking ..because ultimately that is what people do . and they tend to only hear what they want to hear.

    i know that is true for me , sorting out what i hear sometimes lets me really hear what is being said.

    this post i offer as like a quiet whisper in your ear a nudge in reading between the lines on the replies that you get , like you said , i do not want to discuss the insurance liabilities at the moment... well, i too said the same thing to you in my first post ...and more...

    today i went to look for pre 80's Pex fittings in mm

    here is a site page


    this may expand your horizons in your endeavours.


    there is a compression fitting in their language a Kobling ...lol any compression type with sleeve fittings were and still are in vouge ... the brass or stainless insert little bit different than nominal 3 /8th's the different manufacturers have different expander tools and it may even take two different expander tools , and their corresponding 3/8ths heads, to make some minor adjustments to reality ...i am going with might have done that in the past ok?
  • ChrisL
    ChrisL Member Posts: 121
    Coupler restriction

    I would just use a regular 3/8 coupler.  The resistance of the coupler is equivalent to 3 feet of 3/8" tubing, which is not that much in my opinion.

This discussion has been closed.