Using riser pipe size in basement to "discover" routing on upper floors?

SteamFTW

Three storey 110 y.o. house, HW boiler, original pipe, original CI rads. I was sketching out the supply and return piping to the rads, and all the pipes upstairs are completely inaccessible. The house also has an odd floor plan, so it's not self-evident how the second- and third-floor rads are connected. Fortunately, in the basement everything is visible. So, I thought, "Maybe if I look at the location and size of the risers that I know go directly past the first floor, I can try matching them up with the radiators' EDR/BTU and make some educated guesses as to which feeds what."

Everything about the piping that I can see and measure says "Built by the Dead Men" to me. So, I think it's safe to assume that the pipes were originally sized according to the attached loads and safe flow rates. (But... this would have been a gravity system in the beginning, so maybe I have to adjust what I consider a flow rate threshold, since that didn't become a concern until circulators were introduced. Hold that thought.)

No, I don't expect to answer anything conclusively, but here's my thought process...

10k BTU per gpm (at 20F delta-T)

If I see a 1" riser, and at that location on the second floor there are 2 15,000 BTU radiators, then it's unlikely that they serve either* (and certainly not both) of them. And it's more likely that the 1" riser serves the 4200 BTU bathroom radiator on the third floor.

So...How crazy is this idea?

As always, thanks for your insight!!



*Hmm... 15,000 BTU through a 1" pipe means 5.56 ft/sec, right? While that's not great for a new system design, is it possible that it's close enough that, I ought to consider that the 1" pipe _could be_ serving a single one of the 15,000 BTU radiators? On the other hand, maybe this system has always had a sub-optimal circulator, and therefore there's never truly been 5.56 ft/sec in that pipe? (I'll stop now before I think of another wrinkle.)

Jamie Hall

Problem. If it was originally a gravity system, the pipe sizing will be a very unreliable guide. It didn't have a circulator -- they weren't invented yet. Circulation was entirely by gravity -- thermal convection -- and trying to use the how big a pipe does it take to feed a certain EDR just isn't going to work.

A few other assumptions, however, may help to narrow things down. First, it is unlikely that pipe size was changed in a riser. It may have been changed at individual radiators, but it probably wasn't changed for any lateral runs either. Second, it is very likely that the risers and runs were as direct as possible, given the original floor plan (you may have to do some sleuthing). Third, if there are valves which are functional, and you close one, the rate of temperature rise in a riser will be slower (if a return serves a single radiator, it may not rise at all).

Good luck...

mattmia2

Is there a reason you need to know this?

Tapping on one end and listening on the other with a dowel or screwdriver or stethoscope is an effective method.

EBEBRATT-Ed

As far as I know weather pumped or gravity the rad should produce 150 btu/sq foot of CI radiation.

So I would calculate the btu output of the rads then look at a gravaty pipe sizing chart to figure the pipe size to the rads you think are on the same riser.

@mattmia2 idea may be the best.

You could also (if you want to go to the extreme) Disconnect the rads and stick a hose in the pipe and then go in the basement and see which riser gets cold

leonz

What are you doing????

Are you intent on ripping everything out and replacing it with something else????

Is this system still a gravity hot water system?

Is the open to air expansion tank still in use?

The riser and cool water drop pipe(s) would be close to each other.

SteamFTW

@mattmia2 If I had a nickle for each time someone asked me that, everyone could retire. I fully expect to be tapping on pipe as part of the solution.  Stethoscope: I can put that up on pipe that I couldn't get my ear on!  Thanks!

@Jamie Hall Thank you for your comments! As usual they resonate very well with where I'm coming from, as a "sleuth".  Several things have occurred to me, which will take a little time to distill into a manageable reply.

@EBEBRATT-Ed The hose idea is clever!  While letting that percolate and mix with the idea of tapping on the pipe, I remembered some old houses used to have "speaking tubes".  I might just make a mouthpiece that I can attach to the air vent tapping, treat the piping as a "speaking tube", and see what I can learn.

@leonz "ripping everything out"  Please don't say such grisly things out loud. I'm morally programmed to stand up against violence to old heating systems

-S

mattmia2

You have to remember with gravity the flow was much less and the delta t was much larger.

PC7060

Simple way is to use hand held FLIR scope or similar device connected to phone. The hot pipes show up very clearly in walls and ceilings. 

SteamFTW

@mattmia2 I was just reading on Friday a HH thread from 2002(?), where @Steamhead and @Boilerpro were discussing their experimental results and how they seemed to imply exactly that! I also just realized something I didn't think to notice before: the older the manual/handbook, the more examples, and the wider delta T in those examples. Some just give you the math to figure out your system based on ANY delta t. That says something about how flexible you had to be when doing gravity.

mattmia2

We just use 20 degrees delta t now to have something to plug in to the math, it isn't what we actually end up with in the finished system.

JUGHNE

FWIW, if this helps you in your search, The gravity system I have encountered, stack the "zones".

Many people try to figure out the 1st floor from 2nd floor, that was not the case.
It was perhaps the west side, east side etc, but doing both floors.

That is you find a set of pipes going up thru the floor to the first floor, they usually continue up to the next floor.
That 2nd floor riser may be reduced because of less heat required up there and the taller the vertical riser the more flow is created.....maybe a combination of both.

SteamFTW

@PC7060 Unfortunately in the specific scenario that prompted my post, the boiler has been flood damaged and must be replaced. It’s out of commission. But that’s a great idea!! Almost makes me want to hook up a temporary alternate heat source, refill the system, and try that out. With the boiler out of the way, I might even be able to do each loop one at a time.  The other two loops would get no heat whatsoever, and I could get results faster than if they were all hooked up at the same time.

These are all great ideas. Thank you so much for sharing them!

@JUGHNE Yes, I quickly realized there wasn't an obvious correlation between floors, especially a couple of radiators that seem to be ridiculously far from the nearest riser. I'll also admit it took me a while to rearrange in my mind the textbook "direct return" and see that what I've got here are loops where some branches go directly to a first floor rad, and others go higher up--and that those upper floor risers could form a sort of nested direct return.  With one of the three direct return loops from the boiler, it's the very first branch (and the very last branch) off the "main" that skip the the first floor and go right to one or both of the upper floors. On another loop it's the last and the second to last branches. And the other branch it's another permutation. What I find so...I dunno...awesome(?) is that even a very rough estimate of the total equivalent lengths of what I can see supports the paradigm that you need to add fittings to slow down flow on certain branches in order for the rooms in the house to heat evenly. I realize that I'm basically saying, "Wow, those Dead Men knew what they were doing!", which isn't news to anyone, including me, but it's thrilling to map out a run where there's no obstacle that forced the fitters to stick in 2 45s and a 90, and find out that without those seemingly superfluous fittings and nipples, the head loss would not have favored (or avoided) that branch to the benefit of the occupants of the house. I love that! There are no machines involved. It's human creativity finding a way to pull something off within the limitations set by the laws of physics. Not only did they not have ECM circulators--they didn't have circulators at all!

PC7060

Cold water from domestic supply would show up well.  Also work as a crude but effective leak detector. 

mattmia2

The combinations of fittings where there is no obstacle are much more likely a swing joint to allow sections of the system to expand and contract independently.