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.

What type of pex tubing to use for hot water heating with cast iron radiators?

Jamso
Jamso Member Posts: 9
edited August 2018 in Radiant Heating
I know this topic has been touched on before here....but everyone's situation is a bit different as far at the project they are planning and many of the threads are at least a few years old so some things may have changed in the industry or peoples' opinions may have changed or maybe now they have new or different experiences with pex for heating...so I'm asking it again for my current situation.

My house is in Duluth, MN, (gets pretty cold here) it was built in 1925, and it has 16 cast iron radiators, using hot water. The boiler is about 11 years old and is working well. I've been told by boiler maintenance workers that the system used to be steam based on the size and pitch of the threaded iron pipes. The pipes are 3" in the boiler room and then taper down sequentially via reducers and tees to 3/4" as they run out toward the different Tees and radiators. The boiler has 1 1/4' coppers pipes that come out of it, which then feed into the 3" iron pipes in the boiler room. The large iron pipes that feed the radiators hang down a good distance from the ceiling, so they are often in the way, they look kind of obtrusive and they make it difficult to put any kind of ceiling in, I bang my head on them from time to time, etc...so I'm hoping to replace at least some of the big pipes with pex or a pex type product. I'll likely replace one or two of the big pipes with a manifold and then run pex from the manifold to each radiator that runs off that particular run. Which is about 8 radiators. Many of the radiators are upstairs and the pipes run though walls etc so I wasn't planning on replacing those pipes but just disconnecting the pipes in the basement up to the elbows or connectors that are between the basement floor joists and just run my pex from the manifold to those fittings.
(For the ones on the floor just above the basement I would replace the hardware with new valves, elbows and run a new piece of brass pipe down through the floor and connect the pex to the new brass pipes up in the joist spaces, as well as put new decorative brass escutcheons around the brass pipe. (My current pipes coming up through the floors are painted and dinged and look bad, so I want to replace them...and i have a friend with pex on some of his radiators and the pex runs up through the floor to the radiator which I think looks bad...hence i'm doing the brass pipe thing).

Anyway, I know for sure that I should use an oxygen barrier product, and originally I was just going to use oxygen barrier pex, but then have been reading threads that recommend pex-al-pex. So I'm looking for clarification as to whether or not it is really necessary or at least highly recommended to use pex-al-pex over o2 barrier pexin my case, or if there are other types i.e PERT that would work well for me.

Based on what I've read the problem with normal o2 barrier pex is that it expands and moves which makes noise as it rubs against floors, clamps, etc. In my case the tubing would just end up being up in my rafters as I'm not going to be putting it in walls or in subfloors etc so my guess is that it would have less to hit or rub against...but maybe I'm wrong on that and the expansion and noise will be worse than I might expect from general o2 barrier pex. So, basically I'm looking for a cost effective Pex option...but also want one that isn't going to cause me problems i.e. noise. I'd rather spend a little more money on something and be happy with it as well as not have to redo it.... but also don't want to spend the extra money on pex-al-pex if I'm unlikely to see a benefit from it for my particular project.



So...any advice you have on pex choice type (pex-al-pex vs general o2 barrier pex, pex a vs pex b, as well as information on PERT being an option) and possibly even brands Uponor, Sioux Chief, Nibco, Watts etc would be appreciated.

Comments

  • hot_rod
    hot_rod Member Posts: 22,024
    You pretty much have the right info. Pex A and the PERT is a more flexible tube to work with. If the O2 exterior EVOH barrier is covered squeaking should not be and issue. The A pex is forgiving for kinks and takes an expander 1960 fitting nicely.

    My go to for remodel and threading tube was the Viega FostaPex. It had the advantages of a full bore inner pex, plus the features of the aluminum outer layer to limit expansion and hold bends nicely. Now special fittings required like some PAP. It does require a stripper tool. It is also available in 20' sticks if you have short runs.

    If it had a A pex inside it could be expanded, a nice upgrade. Although some of the B pex guys are allowing expander fittings now?
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • Zman
    Zman Member Posts: 7,561
    edited August 2018
    I would say Pex A or Pex-AL-Pex. It really just depends on the installation and personal preference.
    I would be surprised if your existing system was ever steam. Much more likely it is a converted gravity system.


    "If you can't explain it simply, you don't understand it well enough"
    Albert Einstein
    CanuckerIronman
  • Ironman
    Ironman Member Posts: 7,367
    We generally would use PAP for that kind of job, but I have all the press tools and such which makes it easier. You can get compression fittings for it that don't require pressing.

    Any of the three types of tubing are reliable if done correctly, but I would caution about one area: if you're gonna do this you have to do the entire system. You can't leave part of it with large gravity piping and have part of it PEX. Water takes the path of least resistance and that would surely be the larger piping leaving the PEX with little or no flow.
    Bob Boan
    You can choose to do what you want, but you cannot choose the consequences.
    ZmanCanucker
  • Jamso
    Jamso Member Posts: 9
    Thank you all for the replies. A clarification on my setup is that (as I mentioned previously) I was told it was probably steam at one time by someone that came to clean and inspect the boiler....and so I sort of had that in my head as being the likely truth without any real knowledge of the past setup or of boilers and radiators in general....however based on reading the post from Zman being surprised that it was ever steam and from what I've seen on the system I would be inclined to doubt the steam as well....partly because I think all the valves on the radiators are likely original and they are of the hot water variety vs steam....i.e. they are cylindrical in shape vs the more ball shaped valves for steam....internally they have another half or partial cylinder that when turned one way the large gap in the cylinder leaves the path fully open and when turned the other way the cylinder closes off the gap except for a tiny hole. (however in my case even when closed there seems to be a pretty significant amount of flow on several of the radiators) From my reading on steam vs water valves etc I've concluded all the valves on my radiators are water valves. Hence I would guess ZMAN is most likely correct in guessing it is a converted gravity system. Currently there are two pumps on the on the supply path (I'm assuming that's part of the conversion from gravity feed) In that case would that change anyone's opinion on converting to pex?

    Ironman, thank you for the advice on the partial conversion or actually the advice against it. I might hold off on the overall conversion for awhile and just replace some of the smaller 3/4 inch pipes to one radiator that is currently disconnected with short runs of pex.

    I'm a little partial to using pex-a as that is what is now on most of my potable plumbing lines. I preferred working with it over pex b of which i have some as well. I found the pex b crimp tool a little difficult to get into some of the tight spaces, but found that the Milwaukee expansion tool didn't need to get into the tight spaces it just needed to get near it and then you could bend the pex out a bit into more open area expand it and then quickly bend it back and insert it on the fitting and no need to use a gauge or calibrate the tool. However I didn't want to use pex-a for this project based solely on the fact that i already have the tool and like working with the tubing, i want to use the right stuff and I see i can use compression fittings for PAP so was assuming i would do that if PAP was the way to go. I have plenty of wrenches.

    When I actually do this, I'll probably end up using PAP. (unless of course someone has some new and compelling argument to use something else) Based on some other threads as well as this one it doesn't seem that pex-a is a bad or totally incorrect choice for this, but it appears that PAP is still basically preferred, and i'm thinking the extra expense which is not huge would be worth the piece of mind of not worrying about the potential noise issues. I guess my conclusion from my research is that PAP is definitely not wrong for this, but it's possible that other types might not be quite as good, so go with the sure bet.

    Thanks again......any additional opinions, advice or personal experiences are still welcome and encouraged.....i might not end up doing this for another year. It's summer now, but winter comes quickly in Duluth, MN so I don't want to have my heating system all torn apart once it starts getting cold.
    Zman
  • Jamso
    Jamso Member Posts: 9
    I have two more questions regarding my radiators in the event that anyone is still paying attention. Most of my 16 radiators have 3/4 inch valves and 3 have 1 inch valves. In other strings where people have asked about the size of pex needed most answers have seemed to say that 1/2 inch pex should be more than sufficient for most radiators and 3/4 only being needed for very large radiators....i.e radiators larger than ones generally found in homes. I was thinking that since I am replacing a few old broken valves that I could possibly get 1/2 inch valves to go with 1/2 inch pex and not use reducers etc.....however it appears that my radiators have bushings with 3/4” in input so I would either need an additional bushing to size it down which I think would look weird and end up setting the supply and return further out in the side and mess up the spacing for the pipes going through the floor. Other option would be to replace the bushing in the radiator. The radiator bushings are painted over and I would expect pretty much frozen in place so I’m guessing that replacing those bushings would be more trouble than it’s worth.....so...

    Is this what many people find.....radiators with 3/4 valves that they end up running 1/2 inch pex to with a reducer somewhere in the mix? If my lines are mostly 3/4 now....will 1/2 inch for most make sense with possibly using 3/4 for the 3 that currently have 1” pipes. Or is it likely that 1/2 inch would work all around and the current 1” is overkill.

    Other question is.....since one of my desires in this project is to get rid of the large iron pipes hanging from my ceiling would it work to replace that big 2.5”pipe with 1” pex fed from the 1 1/4 copper supply on my boiler with half of that 1” pex running to one side of the house and the other half to the other side of the house with 16 1/2inch reducing tees placed along the way. The boiler Is central so 8 tees would be on one side and 8 on the other. Kind of like 2 very long manifolds. I’m thinking that would end up being quite a bit less pex running around my basement ceiling. Would 1” likely suffice for the main runs?
  • Zman
    Zman Member Posts: 7,561
    The flow rates are dictated by the Btu's required. BTU/Hr = GPM*Delta T * 500.
    If you start with a room by room heat loss, you can then figure gpm's required. Once you have that, designing your piping keeping the flow rates under 4 ft/sec is a good idea.
    "If you can't explain it simply, you don't understand it well enough"
    Albert Einstein
  • Jamso
    Jamso Member Posts: 9
    I believe I've seen answers like that on other threads, but I'll have to admit I'm not sure where to start in finding any one of the variables mentioned in that equation.
    Is there a good website to reference that will help me find or calculate these values?

    i.e. doing a room by room heat loss....is that something I can do myself based on calculations of room size, exterior walls, r factor of insulation in the walls etc...or is it something that I actually measure i.e. with a thermometer over a certain amount of time...or is that something that I need to have done by someone else i.e. an energy assessment of my house?
  • Canucker
    Canucker Member Posts: 722
    You can have it good, fast or cheap. Pick two
  • Jamso
    Jamso Member Posts: 9
    I used a couple different calculators that have different ways to apply variables etc. So with some rough calculations based on not knowing the exact r value in my walls, having single pain windows with storms etc... I calculate about 120,000 btu/hr on a bad day when it is -20 outside. I think to get GPM I need to know delta t....? How do I find that is that the delta t per radiator ie if supply is 160 and return is 140 that’s a delta t of 20....or is it more like the difference of inside and outside temp?
  • Ironman
    Ironman Member Posts: 7,367
    edited August 2018
    You'd really be better off doing a home run system with a manifold(s). But, here's some very simplified guidance:

    1. Your system was designed around a 20* at design conditions. That means a 20* difference between supply and return water temps.
    2. Under those conditions, and at 4 fps,
    A. 1.25" pipe = 14 gpm.
    B. 1" pipe = 8 gpm.
    C. 3/4" pipe = 4 gpm.
    D. 1/2" pipe = 1.5 gpm.
    3. With a 20* delta T, 1 gpm = 10,000 btus; 2 gpm = 20,000, etc.
    4. These sizes assume about 6ft. of head per 100' of piping and fittings. Remember, that's actual round trip feet of piping.
    5. The attached radiation tables show how much output per square foot of surface area a radiator gives. Multiply that X 150 to get the btu output of the radiator. Then select the tubing size that will give you the required gpm for that rad. This is simpler than doing a load calc and will suffice for your situation.
    6. Keep in mind that if you do a direct return piping system like you're considering, that each branch that gets farther from the source will receive less flow even though they may all be the same size. Balancing valves will be needed on each branch. That's why a homerun system with a manifold is preferred.


    Bob Boan
    You can choose to do what you want, but you cannot choose the consequences.
    Canucker
  • FranklinD
    FranklinD Member Posts: 399
    @Jamso - welcome from another Northerner...I’m right across from you in Superior! I too have an old home with a converted gravity system, and plan to convert the large old mains to PEX next summer, so I’m reading this thread with great interest :-) Always nice to see another local on the Wall.
    Ford Master Technician, "Tinkerer of Terror"
    Police & Fire Equipment Lead Mechanic, NW WI
    Lover of Old Homes & Gravity Hot Water Systems
  • EYoder
    EYoder Member Posts: 60
    It takes a pretty large radiator to need more than 1/2" Pex (my opinion). Home run Pex to a manifold is so nice and simple.
    Canucker
  • Jamso
    Jamso Member Posts: 9
    Ironman....thank you....that information is super helpful. I’ll have to measure some of my larger radiators to confirm needed size but from memory (I’m in Texas at the moment) most of my radiators would be around 4500-5000 BTUs....so 0.5 gpm and 1/2 tubing should do it. Also thanks on the advice for the manifold...I’ll have to look into the logistics of running 32 pieces of pex out of my boiler room...I agree with Eyoder that ther is a simplicity to it.

    That being said, could you clarify what you mean by 6’ of head per 100 ‘ of pipe. Is the large pipe from the boiler consider head pipe? Also, for a manifold if I had 16 outlets at 5000 BTUs each that would be 80,000 BTUs or 8 gpm. Seems like a 1” manifold would be just enough for that case.....but if after confirming all my radiator btu outputs I end up with more in the 95,000 range....I’m guessing I would need a 1.25” manifold....correct.


    FranklinD...glad to hear from someone else in the area....I’ll keep you posted on my findings, work and outcomes. As you know winter is not too far away so I’ll likely tackle this next summer after I get all my ducks in a row.
  • Ironman
    Ironman Member Posts: 7,367
    Head is the resistance to flow from the pipe and fittings. The 6' number is a generality for how we size new piping. It doesn't apply to your present gravity flow pipes.
    Bob Boan
    You can choose to do what you want, but you cannot choose the consequences.
  • Canucker
    Canucker Member Posts: 722
    I did exactly what you're looking to do at my house. I used uponor pex-a with the oxygen barrier for my conversion, manifolds in the mechanical room. I have 20 cast iron radiators supplied from there and it doesn't take much room, if you plan it out. I would suggest insulating the piping that you end up running together as it clicks when it runs against the others in a chase. I notice it when the indirect satisfies and it returns to space heating or another zone starts up when one has been working. I have insulated a lot of the piping and it quieted right down, the only ones left clicking are the ones I haven't gotten around to yet.
    You can have it good, fast or cheap. Pick two
  • Jamso
    Jamso Member Posts: 9
    Thanks again all....I appreciate the continued comments.

    Canucker...may I ask what kind of manifolds you used? Brand? Material? Main pipe and outlet size? Valves.....no valves? Etc.....

    There seem to be quite a few different types out there stainless steel, copper, ones with just straight copper outlets, ones with pex outlets, ones with valves already on, etc. They seem to be mainly 2 to 12 outlets....some with 16 but not very many. Was thinking about getting two copper ones that have eight outlets and sweating them together (actually 4 of them, 2 for supply and 2 for return). 1” seems to be most common... would a 1” copper manifold be big enough for 16 outlets each requiring .5 gpm or should I go to 1 /4? It looks like Watts makes them in pairs with balancing valves on supply and standard full flow on the return....might be a bit pricey but possibly worth it to not need to separately purchase and sweat on 32 valves.

    So suggestions or recommendations for brands, types, sizes of manifolds?
  • Canucker
    Canucker Member Posts: 722
    @Jamso No brand, they are made of steel pipe, ball valves on supply and return of all radiator lines. Boiler loop is 1 1/4 inch as are the ground floor and second floor zones. The basement is 1 inch. All the supply and return lines are 1/2 inch pex-a except for a couple of the largest rads, which are 3/4 inch. The manifolds were built by a former contractor, I merely repiped the supply lines as they were not oxygen barrier originally. I have a thread on here about it somewhere.
    You can have it good, fast or cheap. Pick two
  • Jamso
    Jamso Member Posts: 9
    Canucker thanks for the info. I read your thread about Aquapex in a closed loop which is what I m assuming you are referencing. Crazy issue with the aquapex on the boiler system. Sounds like overall you ended up having done what I am hoping to do. At the moment I have mainly one big set of pipes so I was thinking of replacing that with one large manifold...but it sounds like you have multiple manifolds. Do you happen to have images of the manifolds and boiler?
  • duffy_4
    duffy_4 Member Posts: 79
    I have done quite a few home run systems here in chicago when people want head room back in there basements. Have used both Veiga and watts PAP tubing ,like pap better less expansion ,have mostly run 1/2” tubing after following ironmans chart that he posted.have to allow for adding bsmt heat somehow since it was heated previously from all the old steel pipes.




  • Canucker
    Canucker Member Posts: 722
    @Jamso my set up is very similar to @duffy pictures, mine are just done with threaded steel pipe not copper. I'll see if I have any pictures of the original system then you can see what I started with
    You can have it good, fast or cheap. Pick two
  • Jamso
    Jamso Member Posts: 9
    Thanks for the pictures and additional info and interest from all.
    I'm back home again and have done some actual measurement of my radiators. My biggest is a column type. It has 3 tube columns that are 26 inches tall and has 35 of those. Based on the chart that Ironman posted (thanks again for that) and assuming a 20 degree delta T. I calculate that one radiator to be 19,687.5 btu output...or basically 20,000 btus. Based on the numbers sent by ironman it seems that 1/2 pipe can handle about 15,000 btus, So for that radiator I'm guessing I need to go up to a 3/4 line. Any problems with my math or conclusions there?

    The next biggest in size is another column type, also with 3 tube columns that are 26 inches tall, but it only as 25 columns. That one I calculate at 14,062.5 btus. It's close to the limit, but I'm guessing 1/2 tubing would suffice here. Again, does that sound correct.

    Then third biggest one is a tube type with 7 tube columns that are 17 inches high and it is 25 columns so I get 11,250 btus for that one, so I'm thinking 1/2 would be sufficient for that one and all the other radiators I have.

    Agreed? Other factors I should consider?
  • Canucker
    Canucker Member Posts: 722
    Pipe size is decided by the heat loss. I'm assuming that the output numbers you posted are with 180F water supply temp? If the room doesn't need the full rad output, you won't necessarily need larger than 1/2 inch. All the rads in my house are bigger than yours and only 3 are supplied with 3/4 inch. Unnecessarily as the heat loss calculation for those areas is less than the smaller pipe can carry. I guess the short answer is, unsurprisingly, it depends ;)
    You can have it good, fast or cheap. Pick two