Steam Main Crossovers?

MTC

I manage a few smaller apartment buildings, and have been learning as much as I can about their steam heating systems so that I can properly maintain them. These systems have been messed with by far too many knuckleheads over the years, and many things are wrong with them. I'm working to get the owner to make corrections a little at a time, but it can be a fight at times.



Anyway, I'm aware that all the near boiler piping is wrong, there' no chance of correcting it right now, so just looking for some insight into the merits or insanity of a couple other issues in 2 of the systems. Hopefully my sketches of the piping will be large and clear enough to show what's going on. They're very different systems, the first is a 1 pipe in a smaller 7 unit building with 2 small boilers, the 2nd is a 2 pipe in a larger 7 unit with one large boiler. Both systems have newer (10 or less years, I think) boilers. Both have boiler feed pumps, and no significant amount of wet return lines. The 2 pipe system has an auxiliary condensate pump to return condensate to the boiler feed pump from the end of a couple fairly long mains.



My question comes in that both systems have 2 mains, or branches of mains, which meet each other at some point in the system in one format or another. I didn't think this was supposed to happen, and was wondering if there was some reason for it, or if it was a knucklehead "fix" sometime along the way.



The 2 pipe system has 2 mains with a full size crossover between them at about the midpoint, and another one at the end. In the end crossover, the long run of pipe is full sized, then there's a small (1" maybe?) crossover pipe between the 2 lines by the condensate pump. This short, small line also has a gate valve in it (currently open) and a pressure relief valve. All of that makes no sense to me.



The 1 pipe system has a main coming off the header that splits and goes somewhat around the perimeter of the building in both directions. One main ends, drops down into a condensate line, and that condensate line goes over to meet the end of the other main, which also drops into this condensate line, which then returns under that main to the boiler feed pump. Above this point where the two "meet", there is a single Hoffman #75 main vent, the only vent in the entire system, other than the boiler feed pump's vent. The Hoffman seems to, to some minor degree, vent both branches of the main. The branch that goes right to the vent heats up in about 12 mins from call for heat (not start of steam production), the branch that drops into the condensate return and then over to the vent takes about 18 mins from call for heat to get hot. There is no room for additional vents anywhere else on the mains, but might be able to tap another into the condensate line near the end of the first branch, then maybe menorah both locations out and put in several vents on each. It's not ideal, but this is all inside a finished apartment (its a weird arrangement), so our options are very limited. This may help balance the two primary supply lines a bit better and vent them much faster. But is the piping where the steam in the condensate lines meets each other a normal setup? Should it take this long to get heat out to the end of mains? This system is controlled by a thermostat, with one boiler running primary, and a 2nd that kicks in based on an outside override set around 20 degrees. It has been very difficult to get any sort of balance of heat between the apartments. I'll work on getting the owner to replace the guts of the F&T traps in this system as well, though so far I don't think there's too much steam escaping through the pump vent, but I'm sure its long overdue anyway.



There was also no insulation on many of these lines in both systems. I'm working on getting them all tucked in again. We're replacing all the rad traps and the guts of all the F&T traps in the 2 pipe system in a week or so, as it hasn't been done in a very long time. After getting some of the basics taken care of, I'll need to work on balancing the systems, setting up the Heat Timer properly in the 2 pipe system, the vapor/pressuretrol settings in both systems, etc. But I figured it wasn't much use messing with those too much now when the mains are condensing way too much, venting poorly, experiencing trap failures, etc.



I know that's a lot in one post... but the primary thing I'm interested in for now is the crossover piping in both systems. Can anyone shed some light on what's going on here, and if its proper or not? Or any advice for any of the other stuff I mentioned/drew?



Thanks for all your help.

Jamie Hall

The most amazing thing about steam

is that it is so forgiving of oddities... most of them, anyway.



I have to admit that I can't see any purpose for the crossover lines between mains, other than to be a possible pain in the neck.  And to make figuring out how the steam gets to where it is wanted considerably more difficult.  On the other hand, if they are pitched so that you aren't getting water hammer in them, I can't see that they'd do any harm to speak of either.



Both your systems use dry returns, up to the condensate receivers and pumps (of which I am no great fan; gravity being very inexpensive and reliable, but that's another story).  What I do see is a lamentable lack of main venting,  Now there is nothing inherently wrong with venting a system through the dry returns -- vapour systems do it all the time -- provided the dry returns are big enough to handle both the condensate and the air.  If you are going to do that, though, the problem is to ensure that steam can't possibly get into the dry returns, but allow air and condensate into them. 



If I am reading your drawing correctly on the one pipe system, steam can indeed get into the longer dry return, quite unimpeded, in the one pipe system, and the F&T at the boiler is serving as a vent -- and the shorter main has no vent at all, except the F&T.  I have no idea how this thing was originally intended to be set up -- except that what you see isn't likely to be it.  To avoid repiping a whole bunch of stuff, though, if you could add venting at the location of the Hoffman 75, and install new venting where the other branch of that main drops to the dry return, it couldn't hurt.  And also, perhaps, add venting at the end of the shorter main.



On the two pipe system, adding vents at both ends of the two crossovers (if possible) would also help, I expect.



Good luck...

MTC

Good ole buildings...

that have been messed with by too many hands. I'm pretty certain that the crossovers were added later (one looks like a fairly new pipe), trying to balance a system riddled with failed rad and F&T traps. I guess its not worth removing them, only real problem with them other than potential water hammer (I think the crossover condensate drains fairly well into the drip at mid steam main, but not sure - it doesn't seem to hammer), is that its eating up some steam. At least for now, I guess I'll just insulate it and leave well enough alone, but if it seems to create problems later, probably worth cutting it out and plugging those tees? Its a fairly long pipe, either 2" or 2 1/2", don't remember for sure.



Yes, this is all dry returns, only "wet" lines are the pumped lines. I don't have enough experience to have a preference, but this is what we're working with. I assume some guy upsold the owner on them or something, but they are what they are now, I'm not going to undertake removing them, at least not anytime soon. FWIW, I think most of the condensate lines are 1.5" in the 2 pipe system, and 2" in the 1 pipe. But this is all from memory, so...



You are correct, the 1 pipe system has steam in the main return line. The short main (its really just a little branch inside the boiler room off that bullheaded riser from the header) wraps around the boiler room through an F&T and back to the pump. The few rads this serves actually get way too hot and some of their air vents have been removed and plugged, as it has by far the shortest venting route. The other two branches of the large main combine in the condensate return, with no trap, and return with steam to the F&T traps in the boiler room again. Guessing it was a wet loop gravity return system, and they added those 3 F&T traps to convert it to a boiler feed system at some point. There are 2 vent pipes on the boiler feed pump that essentially vent the entire system through the F&T traps.



The issue with this system is that most of the piping other than the boiler room stuff is inside an apartment, and very near the ceiling. I'm not convinced the old Hoffman #75 is actually venting at all, at least I don't hear it vent when system is running. It's not easily accessible and in the tenant's bedroom, so want to figure out what I'm going to do with it and do it all in one shot, to avoid invading/making a mess of her space multiple times. Where the Hoffman is, I think I can probably menorah a couple Gorton #2s. Would this be unacceptably noisy for being essentially right over the tenants bed? Would it be worth tapping the condensate line just after the end of the other main branch, and venting there as well? I can't get a vent on the main itself, no clearance. Oh, btw, I think i said it backwards in my initial post. The main branch that has the hoffman on it is the one that takes longer to heat up at the moment. Its larger piping, but similar overall distance from boiler to vent, assuming the Hoffman vent is working at all.



On the 2 pipe system, both the condensate pump and the boiler feed pump have 3/4" (I think) open vent lines on them. Would there be much added value to additional venting? Since the bulk of the venting seems to be put on that end of line condensate pump, maybe just getting a 2nd vent pipe coming up off of it would be the faster, easier, cheaper, less moving parts route? This system has been far less problematic than the 1 pipe system, and i think most of the issues we're having relate to the steam trap failures, which we're about to replace all of them. I'm hoping this will help balance the system better, but I'm all ears to any further suggestions.



Thanks Jamie, and anyone else who chimes in, in advance

Hap_Hazzard

It looks like

Somebody replaced the wet return on the one-pipe system with a dry return, not realizing they were providing a path for steam to flow unimpeded from one main to the other. I can't see any reason to have a condensate pump when the entire return loop is above the water line.

Jamie Hall

It occurs to me

looking at your diagrams and reading between the lines in your description -- and then reading Hap's post...



Was a good bit of this ... ah... interesting arrangement done when that finished apartment to which you lovingly refer was created?  That might account for the otherwise somewhat inexplicable lack of wet returns and the installation of return pumps...

MTC

That might be the case,

but if so, it happened a long time ago. The owner has had this building for over 30 years I think, and i know that while he was redoing the main staircase, they offset one of the steam lines a bit with a couple 45s to make it work, but otherwise the majority of the lines stayed as they were when he got it.



The lines he changed are obviously newer, even though by now they're fairly old. Both the main, and the return that hangs under it with those old school pipe hangers, look to be quite old.



I'm gaining a good bit of technical knowledge of steam, but have little practical everyday knowledge at this point. I don't know how these systems would have operated back in the day. Could it be that they used a dry return back to where you see a couple F&T traps in the boiler room, then just dropped to the floor there and back up through a hartford loop to the equalizer, and that was the only wet part of the system? Or would that not be enough condensate storage for the system to operate correctly? Perhaps issues with that were the reason for the switch over to the boiler feed pump?



I didn't think it was overly uncommon to have a 1 pipe system like this... its sort of a hybrid system where most of the time the condensate flows in the same direction as the steam. It seems wasteful, sure, to have that steam go through the dry return, but that's what gives a gravity system the ability to refill the boiler, correct? Though its in many ways better to do so on wet drips to reduce steam losses...



Long story short, couldn't tell you for sure, but the system has def been piped this way (outside the boiler room) for a good long time, but may well not be original.



Given the situation, and the fact that the owner isn't going to go for a major repiping or anything like that, is there a good solution? I think the only way I could stop the dry returns from being full of steam is with some sort of linear F&T trap, if such a thing exists, due to the space restrictions. Or maybe if there's some sort of small drop loop type trap that could be used, but I don't understand much of that side of things. I still couldn't put it at the main ends, but could maybe get it in 35-40 feet out from the boiler where the return goes through a little storage room. This would remove the steam from a substantial portion of the return, if possible, which would free up steam for the radiators that aren't heating sufficiently...



So with this mess that I'm dealing with, my plan of attack right now is to add more main venting as much as possible, and insulate the mains as much as possible (the apartment I "lovingly" describe will present restrictions on both of these, but any improvement is something, i suppose). What's the next step?



The strange thing is that the ends of those two mains feed up several stories to similar layout stacked apartments. The top floor overheats, has Danfoss TRVs on the rads set to like 1.5 or maybe even a little less, to try to get the heat to go to the floor directly above the steam lines, 2 stories down from the apartment that overheats.



The thermostat sensor is in that top floor apartment, which I know shouldn't have the TRV in the same room, but so far that arrangement has been the closest to getting some balance. We're about 10 degrees different between floor 2 and floor 4, when things are going well-ish. Sometimes worse than that.

Jamie Hall

It's much more common

for a one pipe parallel flow system to have a drop to a wet return at the end of the main -- or wet returns and the ends of the mains.



It doesn't seem to me, though, that given the constraints you have, that repiping would really gain you much.  What does concern me is that temperature and steam feed difference between floors.  More main venting is going to help there.  Slowing the vents on the radiators on the top floors will also help.  And insulating the dickens out of every pipe which carries steam, whether it qualifies as a main or a return!

MTC

Thanks for the insights

I love learning about this stuff, and you guys certainly seem to know your stuff



For the 1 pipe, I'll do what I can with venting and insulating, then see where it goes from there. One more question: owner has it set up on a setback thermostat. I have heard that these can wreak some havoc on steam systems, particularly in the recovery period when pulling back up to the higher temperatures. We're trying locking the temp constant for the weekend to see how it goes in the field, but wondering if anyone has any more input on that? Should these be avoided?



For the 2 pipe system, I was just reading a bit in Dan's "Greening Steam" book, which references Gerry and Steve's "Balancing Steam Systems" writeup, which I also have. Anyway, I was viewing this 2 pipe system as not needing any additional main vents as the lines all run through F&T traps to the condensate and boiler pumps, which have open pipe vents on them. In looking at their research, however, it seems that the B&J cage units I'm putting in the Hoffman 17C traps have a higher venting capacity than the F&T traps! This seems counterproductive to the balancing efforts, so I guess I need to add main vents before the F&T traps as well? This is new to me, I thought that a system like this was adequately vented through the condensate receiver vents, but it seems that the F&Ts are the weak link. I've got 5 in this system: 1 Hoffman at .5CFM, 2 B&J at .75CFM, and 2 Dunham Bush's, which aren't on this list. All of those are at 1oz. The cage unit for my 17C pulls 1.3CFM at 1oz. Seems crazy that those little cage units can vent so much faster than the F&Ts, but that's what I'm reading...



Thoughts? Other than that its Friday night and I should be out doing something instead of on here...

Jamie Hall

Crazy it's not

those little cage babies can flow an amazing amount of air with practically no loss -- which is why they are used as crossover traps on vapour systems.  They need to be followed up by really big main vents in those cases!



In my humble opinion an F&T is not a good substitute for real main vents; they just don't flow the air.  In fact, there was a recent thread suggesting using the cage type "traps" like the 17C as main vents -- something which I think would work well in some situations.



Setbacks are a genuinely vexed subject, and you will find more opinions here than you will find authors.  The general consensus, though, seems to be that for steam (and gravity hot water) systems anything much more than 3 to 4 degrees is too much for a daily setback.  "Your mileage may vary"!

MTC

Good to know

Here I thought those traps actually functioned fairly well as vents, but guess not. One of those mains has a plugged fitting not far from the trap, so it should be very easy to get some vents on. I'll have to take a closer look at the other, but I'm sure I can figure out some way to get it done. Thankfully, this building has so far had far less issues than the 1 pipe system, so its not as urgent. I'm mostly concerned with getting the pipes insulated there, and then tweaking the heat timer a bit, as I'm sure the insulation will change things on that end.



Thanks for the info on the setback. Yesterday I had the owner (I wasn't on site) lock in the temp for the weekend, so we'll get a report from the biggest problem apartment on monday when i go in and see if it seems to help at all, maybe I'll reprogram with just a few degrees setback if it seemed to help, and see how that goes.



Jamie, do I remember hearing you say that you've installed inlet orifices in your system? I've read a lot about those, and everyone you talk to seems to have some differing opinion (I think primarily based on what they're trying to sell you, ie more steam traps), but they seem like a good option for balancing, possibly even eliminating the need for traps if the system is oversized enough that you could trottle down to less than 80% of radiator's condensing capacity... How have they worked out for you? How difficult was it to get them all sized properly? Did you just get blanks and drill your own, or have Tunstall size them for you? Were their sizes right if so, if not, how did you go about choosing your hole sizes?



I'm trying to sell the owner on them for the 2 pipe system, if I can sell him that they'd both balance the system and make his traps last a lot longer, it might just happen. Worst case, afterall, is that you just have to go around and pull them all back out and you wasted a relatively small amount of money on them compared to all this other stuff.



Thanks again! You've been very helpful.

Jamie Hall

I don't think it was I

who installed orifices; I do remember reading that thread though.  What I have done, since I am blessed with the original Hoffmann radiator valves which are meant for exactly the purpose of controlling the flow to the radiators, is adjusting them -- which takes some patience -- so that each radiator gets just the steam it can condense, and no more.  Which accomplishes exactly the same thing as orifices, but without quite so much hassle.  And no, I don't know why it wasn't done when the system was installed...  Whatever, it makes the traps redundant, though I haven't actually removed any of them!  Depending on the valves you have on the two pipe system, you may be able to pull the same trick yourself.  Otherwise, Tunstall will do them quite happily -- or you can, if you have the time and inclination.  Not hard, just fussy.



I would note, however, that the trick will only work if you are controlling the pressure to a very low -- half pound or less -- level.

MTC

Oh, ok

sorry for the confusion on that. Thanks for your thoughts anyway though!



I've read that the orifices work at different pressures, you just have to size them for the pressure you're going to use. But it also just makes sense to me that higher pressure, combined with the necessary smaller orifices and the higher velocity necessary to get through said smaller orifices, would be potentially noisier, as well as have a higher potential to be clogged with rust, etc. So very low pressure would probably be ideal. Its running right now on a pressuretrol, so would have to get a vapor on there if i wanted to try for that low. something to think about, i suppose...



This place has a hodge-podge of valves, only maybe a few that i'm unfamiliar with that might have some sort of orifice system in them. All of them are subject to tenants messing with the valves, so no way to balance them that way. If I had orifices based on EDR, then tenants could turn it down further, but until they're roasting, they'll always turn whatever setting I put on there up higher and throw things off again.



Hopefully just getting the steam out of the returns will take care of most of the balance issues anyway, but I'm still so intrigued by the orifice concept... seems so brilliant - no moving parts, nothing to break or replace...

MTC

back to the 1-pipe system...

Ok, so I'm thinking about adding a lot of venting to the mains in both of these buildings. Initially I didn't think the 2 pipe system needed it with all the open pipe vents on the condensate tanks, but since learned that F&T traps vent rates are far too slow for those to be effective. So I'm working on selling the owner on 5 Gorton #2s on the ends of the two mains there right before the F&T traps (3 on larger, 2 on smaller main).



The 1 pipe system really doesn't have a great place to put vents other than where the one is shown on that first picture. If I maximize this location's 3/4" tapping with 4 Gorton 2s, I'll have a good bit of airflow, unfortunately shared between the mains and therefore possibly not very well balanced, but I guess we'll find out.



My question is, since the current system has the returns all full of steam back to the F&T "master trap" in the boiler room, how effective are these vents going to be? I guess the F&T is restrictive enough that the air might favor the radiators...



If you look at that first picture, where the left branch of the main hits the main vent location, then the return runs back directly under it, i was thinking it would be nice if I could stop the steam somehow along there. The whole bottom half of that image is inside an apartment, don't think i could do much in there. But about halfway down that stretch of return headed back to the boiler from the main vent location, it crosses over a storage room. Is there some sort of steam trap device I could install in the return line that wouldn't require lowering the remainder of the return, and still allow the condensate to pass through? I know that's a tall order, but I feel like it would be way better than filling that additional 35 or 40 feet of 2 1/2" return line with steam. Maybe a drip loop? I don't know much about those types of setups...



Thanks for your help!

MTC

Finally got the 2-pipe system

all cleaned up today. I had already gotten most of the mains insulated a while ago with insulation that was stored on site but had been removed over the years. I need a little more to finish up the job, but after insulating, I have already turned down the heat timer setting by 1.5 "letters," a substantial decrease.



Today we replaced all the radiator traps and rebuilt all the F&T traps (except one, which wasn't in yet, should be done tomorrow). Hopefully this will be another big step in reducing unbalance in the system, increasing efficiency, etc.



A few odd things/notes of the day:



1. One of the radiators, which did not have an air vent added to it, and appeared to not have had one at least for a long time, was about half full of water. Was quite the surprise (fountain) when I popped the trap lid off with the impact gun, and quite the scramble to get it back on before the bedroom carpet was totally destroyed with rusty water. Guess that was one of those rare failed closed traps... we wrapped the trap body in towels, then took the cover back off and removed the trap element as fast as possible to allow it to drain, which it did quite effectively as soon as the element was out of the way. How would that much water build up in there w/o having a vent on the radiator to let air out/steam in? I guess maybe there was one before, but the plug was painted along with the radiator, so it didn't look like it...



2. 3 of the 4 F&T traps we rebuilt today had no thermostatic elements in them. They were plugged, and just acting as float traps. The B&J rebuild kits should add a significant amount of venting to the mains (there are no main vents as of yet) compared to what was there. The 1 that actually had a thermostatic element in it was full of mud - and by full, I mean full all the way to the top and backed up into the drip from the steam main. Nothing was going through that trap. I was in frantic mode trying to get all the traps done in one day, so didn't get a chance today, but i suspect that the line from the F&t to the return is not pitched properly, causing the back-up. This means that in the 4 we did today, there was zero main venting. We'll see what the last one (which is inside the boiler room, not far enough away to be a significant main vent) looks like tomorrow, but basically the system had no main venting. It was simply venting through the radiator traps to the returns, and then out the condensate tank vents. I imagine that if all the radiator traps were somehow closed upon system shutdown, we could have ended up with a substantial vacuum formed, as there was no venting whatsoever on the supply side. Crazy.



3. If you're changing radiator traps, and don't have an impact gun, GO BUY ONE! We just got a harbor freight cheapo one (under $40), but it probably saved $300 in labor just today, let alone the potential costs of broken pipes/fittings/etc from trying to wrench off those seized on suckers by hand.



So, will be interesting to see how things go with the new traps. We discovered there are no strainers either, so might take on adding strainers before the traps sometime over the summer. Most weren't too gunked up, other than the one that was completely full, so might not be totally necessary... any thoughts?



The guy that we bought all the parts from, who's a member on here as well, didn't seem to think that additional venting beyond the F&T traps and the open vents of the condensate tanks after the traps, was necessary. The new radiator traps are very high CFM capacity, not sure what the F&T replacement thermostatic element ratings are, but I'm guessing they're similar to the radiator traps. This seems to me like steam would favor the close to boiler radiators, and not get to the end until most of the rads along the way filled. Any thoughts? The venting at the condensate tanks is more than sufficient, but are the F&T traps enough venting to allow the air to get through to the condensate tanks and balance the system?



Now that this system has had most of the maintenance done to it, guess its time to rebuild the 1 pipe system's F&T traps, and try to increase main venting there. I've already insulated almost all of those accessible lines again already, and reduced the setback thermostat differential settings a good bit, and this seems to have helped the imbalance problem a good bit already, but its still far from ideal.