Should I assume all steam traps are bad?

MountainSteam

After a couple years of staggering heating bills to heat my 80 year old house (two-pipe system with a 1989 natural gas boiler).

There are 31 radiators with what look like 2 fairly recent trap models: Hoffman 17C 1/2" angle bronze, and Sterling 750-A 1/2" angle bronze. I didn't check all the traps, but found 4 that are not working: once the radiator is hot, the returns heat up to steam temperature, not just at the base of the trap, but several feet away from the radiator on the way to the dry return.

I see that I can get replacement discs or bellows for these traps for about $60 each. So for 30 radiators, that's around $1,800 total (perhaps less if I get a volume discount).

• Should I assume that the traps are all bad or about to be bad? I assume that if a few failed, they may all fail soon.
  
• Or should I setup some test stations and cycle the existing bellows through a test? That would take time and work, but then I could buy only what I need.
  

I feel like a greenhorn, and I don't want to jump into a $1,800 project when I could approach it with a more discerning plan!

Thanks.

Jamie Hall

I wouldn't. The place I care for has an 80 year old system. There are two traps which are suspect, but the problem was fixed easily by just throttling the radiators with the inlet valves.

However, you seem to have an inordinate number of bad traps. The first thing to check -- the very first thing -- is the cutout pressure on the boiler. If that is over 1.8 psi -- and lower is better, if you have a vapourstat -- that is killing the traps. The next thing to check is whether the trap is really bad. A full radiator will discharge condensate at very close to steam temperature, and the only reliable way to check without taking the trap apart is with an IR thermometer (wrap black tape on the pipes to get a more reliable measurement) or a tempilstik. If the outlet to the trap is more than a few degrees cooler than the inlet the trap is working.

Then, assuming the trap really is not working, try throttling the inlet valve to the point where the radiator just barely gets hot all the way across on long runs.

And if ll that doesn't work, then start replacing traps. But only then.

MountainSteam

I forgot to disclose that when the boiler operates for a while, steam comes out of the feed tank's vent, so I know at least one trap is bad.

Regardless, I'll follow the systematic approach a you suggest and identify the culprit(s).

Sailah

I've seen some really bad fail % but 100% isn't likely.

You are welcome to send whatever you want to me and I'll test them.

You are also welcome to contact me and I can recommend one of our distributors that can help you procure products at a competitive price.

As mentioned, trap will discharge condensate at close to steam temps, temp isn't very reliable for thermostatic traps. Far better to test with the outlet of a trap disconnected and look for the steam.

But the presence of steam in your receiver tank indicates failed traps somewhere.

Steamhead

It might be the pressure, not the traps, as Jamie said. If the steam pressure is too high, condensate leaving the traps can flash back to steam when it reaches the returns which are at atmospheric pressure.

Also, where are you located?

PMJ

I'd pause a long while before spending money on traps on a 2 pipe system. I'd try to spread out the steam production with better control and not ever fill the radiators. You probably have so much installed radiation in that 80 year old house that your radiators don't need to be even 1/2 full to equal the heat loss on an average day. The reality is that with the conventional controls and the boiler running to a pressure stop (even a relatively low one) radiators have to fill way more than is needed for most conditions before the boiler stops making steam. Sounds like in your case way more than is needed - the pressure stop setting must be too high.

Short term just start throttling back each rad with the inlet valves like Jamie said. Then get that pressure down as low as possible so the stop comes sooner.

If you are going to spend money improved control will help the most. Unfortunately there isn't a lot of help out there for this at a price most homeowners are up for. With the standard controls we are stuck with overfill/off as the only cycle. Some of the smarter thermostats have features that help some.

Question - have you been paying the bills in this house for a while and suddenly 2 years are much higher or have you been there just these 2 years?

MountainSteam

The house is at an elevation of 9,000 feet, West of Denver.

About the pressure: The boiler is controlled by a simple thermostat and a basic additive Honeywell pressuretrol (pictures here). The pressuretrol is set for a cut-in of 0.5 psi and a differential of 1 psi. However, during testing, the pressuretrol did not cut-out until 6 psi, although it cut back in at 0.5 psi. So I need to replace it.

As of now, the thermostat controls every boiler cycle: when the temperature drops, the thermostat triggers the boiler, the two radiators in the dining room heat up first, the thermostat measures the temperature rise and turns off the boiler after about 20 minutes. During the cycle, about half the radiators heat to 90%, the other half to 50%, and the pressure rises to about 0.3 psi.

Finally, if I force the boiler cycle to continue by shutting the valves to the dining room radiators (so that the thermostat thinks the house is cold), the pressure stays put around 0.6 psi, and nearly all the radiators heat up on the first floor, but many are still cold on the second floor. If I close the valves to the first floor radiators, the pressure goes up to 1.4 psi, the second floor radiators are hot, and soon steam comes out of the feed tank vent.

Sorry about this long narrative, but I suspect these facts point to failing steam traps on both floors, which I can investigate systematically.

MountainSteam

PMJ, About the bills: I've been through 2.5 winters, and the bills have been consistent. We saw a drop when we insulated the top floor outside walls, but otherwise, it's consistent.

Jamie Hall

Hearing the 6 psi... yeah, I'm afraid you have some failed traps. They can take that once in a while, but not on a regular basis.

So... the first thing to do is to take the pigtail off the boiler and clean it out thoroughly, and check that the opening in the base of the pressuretrol isn't clogged. Reinstall. Better, put on a T, a couple of nipples and 2 90 ells and reinstall with a 0-3 psi gauge, so you can see what is actually happening.

On closing valves: remember that two pipe steam has a terrific advantage in that it doesn't mind partly closed valves. Throttle those two dining room radiators, but not all the way.

Also, you don't mention it but I'm going to bet that you don't have main vents, or if you do that they are either stuck shut or inadequate. A radiator on the first floor and one on the second over it, for example, should start to heat (maybe not fill, but start) within a minute or two of each other -- and unless the house is immense, all the radiators should at least start to heat within about five minutes of each other.

Sailah

Jamie Hall said:

Hearing the 6 psi... yeah, I'm afraid you have some failed traps. They can take that once in a while, but not on a regular basis.

I would beg to differ on this point. Most large universities that run 2 pipe steam are running at 5-15PSIG continuously. These guys have 1000's of radiator traps. They all see MUCH higher pressures than the Heating Help crowd. For better or worse they operate in that range and we can't distinguish any difference in longevity just from operating pressures.

All our cage units are rated to 25 PSIG. The only difference between our regular cage units and those we install in 175 PSIG ultra capacity steam traps is the solder in the diaphragm. At those higher pressures we use a high temp solder but the actual components are the same.

And to go further, all our cage units are what we call balanced pressure thermostatic units. Which means when we fill and seal them, we pull 26" vacuum on them. So when there is 47PSIG in the trap body, the same pressure is inside the cage unit and it is balanced. We do this mainly as our cage units will work in vacuum systems whereas other manufacturers will not because they fill at atmosphere. It's also less stressful on the cage units and they last longer.

I wonder if there is anything to the elevation he is residing?

We test at sea level I wonder if I would get the same results at 9000' with a Hoffman diaphragm if it isn't a balanced pressure unit?

jumper

For eighteen hundred dollars I'd make my own orifices. You may be able to buy washers to fit into intake of hot radiators and see if that warms up the cold ones.

Steamhead

Yup- that's the wrong pressure control for this system. You need a Vaporstat, which will keep the pressure below 1 pound.

Two-pipe steam systems installed in large houses built around the time yours was, are generally Vapor systems. Vapor refers to the extremely low pressure (ounces!) these systems were designed to run at. These systems were the Cadillac of heating in their day, and are still some of the best out there currently. But when working on them, you need to consider the system as a whole. Basically, Vapor systems used the return lines to handle both condensate (water) and air leaving the radiators, venting the air at one central point in the basement and returning the condensate to the boiler.

Every manufacturer of Vapor equipment had its own twist on the basic Vapor concept. There was Hoffman (like Jamie has), Dunham, Webster, Illinois, Sterling, Trane and many others (and these were just some of the ones using thermostatic radiator traps like you have). Some of these companies are still in business, like Hoffman and Sterling. So we'd want to know whose equipment was originally installed on your system. If you still have any of your original radiator shutoff valves or any devices in the piping around the boiler, see if you can find any names or trademarks on them. Better yet, take some pics and post them here. With this info, we can give you a better idea of what you're looking at and what to look for.

You are correct that the traps have to be working properly for the system to run efficiently. The function of a trap is to let air and water get past the trap, but close against steam. Some systems use traps at the ends of the steam mains as well- these might be radiator traps piped to vent air from the mains into the dry (overhead) returns, or they might be larger float-and-thermostatic traps which also handled the condensate from the steam mains. Look for these traps.

The central air venting location is also important. If that vent is not working or too small, the system will struggle to get the air out of the way of the steam, wasting fuel in the process. Find that vent. Some systems like Trane used several vents instead of just one, so it's important to find them all.

Post some pictures of what you find. We love to see this stuff!

EBEBRATT-Ed

This system doesn't need air vents. He mentioned a condensate tank with an open vent pipe which is the best vent you can get. I agree with @Sailah the pressure is not killing the vents. Unless there is a return line trapping water no vents are needed.

get the pressure control fixed and a good low pressure gage so you can see what is going on.

I would try to find out why the second floor is sluggish. check the firing rate against the connected load, check for missing pipe insulation and sagging pipes that could hold water

after that is done if the boiler is producing dry steam and the system won't balance I would think about TRVs

How are your windows and insulation??

Is the boiler water line steady and not tripping the low water cut off??

How many times does the boiler cycle/hour on a cold day??

Steamhead

I'd bet that condensate tank is unnecessary, as many are. And we still need to vent the steam mains.

PMJ

Hey guys doesn't point E need to be below the waterline? Above it don't I see steam mains directly connected to dry returns - or am I seeing this wrong. He says point E is pressurized and hot - no kidding - as drawn isn't it still the supply main?

If the lowest radiators in the place can gravity return then the return tank/pump isn't needed.

My system is also 2 pipe, same vintage, and just slightly less EDR (1000). I only have one big loop (supply and dry return) around the perimeter of the whole house. There was a vent on the supply main at the farthest and lowest end of the loop away from the boiler where main and dry return drop and join below the water line. I removed it altogether - all the air goes out through the rads and the dry return just fine. Those crossover traps are fine if they are working. If not they will mess a lot of things up. I'd start there.

Also, @Mountainsteam it doesn't look like there is an automatic damper on the flue on that monster. If not, consider getting one. Huge losses without that. It's not so much the straight up loss of heat up the chimney; it is the cooling effect on the boiler and the delay that causes in time to steam on next fire. Biggest single improvement I made on my system. My boiler is still warm to the touch after 24 hours off. With a wide open flue it was room temperature in 3-4 or maybe even less.

MountainSteam

Lots of questions! I'm flattered at the attention.

Insulation
The main floor of the house is not insulated - just a 12" stone wall, then a 1" air space and some lath and plaster. There is no way to insulate it without removing the lath and plaster, which is OK in some rooms, but not the main ones which have priceless moldings. The rest of the house is well insulated with all new windows.

Boiler Cycle
The typical boiler cycle, controlled entirely by the thermostat, is 20 minutes. There are two steam supply main loops. The steam reaches the end of the short one within 7 minutes and the end of the longer one in 10 minutes. On the day I tested, it was 17 degrees outside and the thermostat was set at 65: under those conditions, the cycles started every 1 hour & 40 minutes.

The boiler water level is pretty steady during operation.

For a lot of details about the cycle, see here).

Pressure and Pressuretrol
About the pressure: Every radiator gets some heat with the boiler at 0.3 psi. The only reason I let the pressure rise was to test the pressuretrol, which was set to cut-out at 1.5 psi. So I think this system is capable of being tuned for operation at 1 psi or less.
I did make sure the pressuretrol's pigtail is clean and unobstructed.

Main Vents & Vapor System
It seems to my untrained mind that this used to be a vapor system as @Steamhead suggests. For one thing, the house was built by a wealthy lady (she fell in love with a cowboy and wanted to give him a ranch), the house is solid, well designed, so it follows that the steam system would have been a quality one.

Next, the mains vent into the dry returns with cross over-vents (see here) and then plunge to a central point. Two of the condensate returns were piped to a feed tank with a pump triggered by the boiler water level instruments. The third return, which is from two radiators in the library that are lower and can't feed into the other returns, feed to the same central point mentioned above.

If I ignore all the copper stuff installed in 1990, it seems to me that it looks a bit like the vapor systems described in Holohan's "Lost Art" book, on page 238, and that the 1990 installer pulled out all the vapor equipment, but didn't notice or care about the library return.

@PMJ, your question about the lowest radiators: they are borderline for a gravity return. Perhaps the tank & pump are not needed, but is there a benefit to removing them? It gives me peace of mind to think about that big tank full of water when I leave the house unattended.

Also @PMJ, thanks for the tip about the vent flue damper. I will act on that.

Pictures
@Steamhead, I put all my pictures and narrative in one place to communicate to my boiler man:
Click here for pictures of the cross over vents and the possible location of the ex-vapor system. The tabs across the top link to other aspects of the system.

Thank you all for the feedback. I'm meeting with my boiler man in a couple of weeks and feel better prepared already.

Steamhead

The one crossover trap (picture on right) looks like a Sarco. Not sure what the other one is. Can't tell who made the valve- there's writing on the handle but the pic is too small to read. Not sure if those elbows on the library radiators have anything in them or not- if not, is it possible the traps were installed in a basement or crawl space underneath?

The radiators are the "small-tube" type which first came out just before World War 2. That dovetails with the age of the house.

In order to make the system heat faster, it would be a good idea to replace the crossover traps with ones that will pass more air. Measure the length and diameter of your steam mains and we can tell you what you need.

MountainSteam

@Steamhead ,

The longer main is 2.5" diameter, but at the end of the main, after the last riser, it tapers to 2", and then to 1.25", and then to vent B1 in the picture. The shorter one is 2", and leads to vent B2.

Sorry about the poor quality of the inlet valve pictures - I didn't think about focusing on them when I was taking pictures. The picture of the one painted valve is as good as it gets, but I'm attaching a blow up of the most common valve.

As for the library radiators, those elbows go directly into a concrete slab. The two radiators' return is combined into one in the concrete slab, and then emerges in the crawl space, and on to cross over vent B3 and then to "point E" as shown in the attached pictures and also here.

So perhaps you are on the right track, and the original installers had a "master trap" (cross over B3) for both library radiators. The library radiators are the among the worst performers. Could it be because the pressure coming from point E makes it's way towards the radiator and prevents the venting of the radiators? Either way it would be easy to (1) replace the elbows with traps and (2) re-route the return to the feed tank.

Inlet Valve:

Cross over vents B1, B2 and B3:

Point E:

Steamhead

Now we see why the library radiators don't work so well. You and @PMJ are right.

Since Point E is above the boiler's waterline, steam is coming at it from the B1 drip and backtracking along the return from the library radiators. It's looking for the crossover trap at B3 which is a point of lower pressure. Both pipes coming together at Point E must drop below the waterline individually and tie in at Point F. This way, water will fill them up to the boiler's waterline and the steam will not cross over between the two pipes. The return from the library rads will no longer be pressurized and hot, and those radiators will vent properly thru crossover trap B3.

I'm sure the previous boiler's waterline was high enough so that water filled the area of Point E. Obviously this has changed.

All the crossover traps are piped in reverse, i.e. steam is coming up thru what would ordinarily be the return connection. If you end up replacing the traps, they should be repiped in the proper orientation.

Let's see some pics of the condensate tank and how it ties into the system. From the diagram it looks like all it does is pump water from the dry returns back to the boiler. With a Vaporstat controlling the pressure, the tank and pump should not be needed, since the boiler pressure won't be high enough to keep the water from returning by gravity. .

PMJ

@Mountainsteam ,

Impressive documentation I must say.

So the point E problem being above the waterline is the biggest issue. I would first fix that and then see where you are. A lot will change. Second, I assure you a damper will save a lot.

On venting I take an unconventional approach. So what to do about the crossover traps I would pause if the system were mine. The idea of those traps is to vent the main. Our systems our quite similar so I will mention this stuff and you can do with it what you will. My logic goes this way:

Vents are designed to let air out and not steam. The trouble is, though, that they also let air back in. So it ends up that their primary function is to let air out quickly that THEY just let back in. And they do this over and over and over. Your original coal system did not do this. The fire was lit, the mains filled with steam, and they stayed that way as long as there was a fire. So this in and out air business was introduced with on and off fire. When I realized this I set about trying to make my system run not letting the air back in. As I did so things just got better and better. Fully vented systems do run ok. They are not more efficient, and the heat is not nearly as even or as quiet as when you push the air out one time on start up and then keep it out. Pushing air out every single time you fire your boiler - even if it takes only a few ounces of pressure - is wasted effort. I have no vent on my steam main at all. The last thing I want to do is to start sticking room temperature air back into my main immediately every time the boiler shuts down which is what those crossover traps will do.

It seems to me that what people miss is the huge difference between getting the air out the first time the system starts up and getting it out on all the subsequent cycles. Night and day different. It takes so long to heat all that pipe up that getting the air out is quite frankly a non-issue on the first cycle. I know because I only have one 1/2" vent in the entire system and I never even hear anything coming out. On all cycles after that, however, venting becomes a big issue, but only if you have insisted on letting all the air back in every time. I have shown conclusively on my own 1926 2 pipe system that letting the air back in is totally unnecessary and detracts significantly from the overall performance. I have found that the less air that gets back in each time the boiler shuts down the better the performance in all aspects - how efficient, how even, how quiet.

I will leave it there. It seemed like maybe you were going to do some of this yourself. I can help with it if you are interested in trying. Going down the ventless road with a contractor is unlikely. Too risky for them to break with convention and I get that. There are plenty of folks here who can help you with vents and traps.

Jamie Hall

Well... I think I differ slightly from @PMJ . In that I believe that the crossover traps really are needed -- indeed, are vital. In many of the older systems, one had the crossover traps, and one master vent on the dry returns where they connected together at the boiler. Sometimes this was an open connection to the atmosphere. Sometimes this was a more or less ordinary vent. Sometimes it was a vent which sealed against a vacuum.

What many more recent systems have -- and what may be required for them -- is vents on the steam mains. But, if you have crossover traps those vents are not only not needed, but you don't want them at all.

Keep the crossover traps. If you want the system to drop into vacuum, as @PMJ suggests, well and good -- make the master vent a vacuum vent. Or vents.

As to the problem at point E... it's not uncommon for what had been a water seal to become dry because a replacement boiler was mounted lower than the original. Since you have that, it might be worth wandering around the rest of the system to find out if there were other water seals or pipes which had been wet returns which are now dry; they may be giving subtle problems.

PMJ

Jamie Hall said:

Well... I think I differ slightly from @PMJ . In that I believe that the crossover traps really are needed -- indeed, are vital. In many of the older systems, one had the crossover traps, and one master vent on the dry returns where they connected together at the boiler. Sometimes this was an open connection to the atmosphere. Sometimes this was a more or less ordinary vent. Sometimes it was a vent which sealed against a vacuum.

What many more recent systems have -- and what may be required for them -- is vents on the steam mains. But, if you have crossover traps those vents are not only not needed, but you don't want them at all.

Keep the crossover traps. If you want the system to drop into vacuum, as @PMJ suggests, well and good -- make the master vent a vacuum vent. Or vents.

As to the problem at point E... it's not uncommon for what had been a water seal to become dry because a replacement boiler was mounted lower than the original. Since you have that, it might be worth wandering around the rest of the system to find out if there were other water seals or pipes which had been wet returns which are now dry; they may be giving subtle problems.

Jamie, I am missing what would happen if they weren't there? Are you saying that steam would never fill the mains? I think it would just fine on initial warmup. Once you have a main with no air in it and realize you can keep it that way your thinking changes. If you want to do the back and forth air thing there too ok but I don't see the point. Took me a while to adjust myself to picturing a main always full of steam and always filling rads if slowly even with the burner off. Any source of air to the main with the burner off kills that forward motion.

It is quite possible I'm missing something but if they are vital why don't I need one on my system? I ran for a long time with no vacuum and just one wide open pipe on the dry return. No connections at all between main loop and dry return loop except through the rads. No main filling problem then either - just a backwards running system with the burner off with collapsing steam being chased by air back to the boiler.

Jamie Hall

Well... in a very real sense you are correct in wondering why bother with crossover traps, but it depends on the system.

First off, most systems are not sufficiently tight to maintain a deep vacuum for any length of time, and they will only do that if the only vent or vents they have seal against a vacuum -- which most don't. So it is prudent to assume that after an hour or two of the system being off, there is going to be air in it (note that in the days of coal, this wasn't necessarily true -- if the fire was kept going, there was always some steam in the steam mains. But that's a different story).

Now it would be quite true that the capacity of the crossover traps to allow air to cross into the dry returns would be irrelevant if one assumes that the radiator traps -- which are of similar size -- could do the job equally well. However, in many vapour systems, that isn't true. On systems with traps, it' not because the traps can't do the job -- they can -- but because the inlets are throttled down (orifices or valves) so that only enough steam at the operating pressure -- a few ounces -- can get in to fill that specific radiator. On systems with no traps -- depending only on orifices or throttling valves, the same applies. For these systems to heat evenly, the steam must arrive at all the radiators at as close to the same time as possible -- and in order to do that, any air must leave the main as expeditiously as possible. Via the crossover trap.

This is, of course, the "secret" to systems which had an open pipe on the dry return, often to the chimney -- the throttling on the radiation prevented any live steam from getting into the dry returns. Just to be on the safe side, it was not uncommon for these systems to have what looked like a huge radiator on the dry return -- it wasn't; it was a condenser, which ensured that any steam that did get into the dry return was condensed and returned to the boiler.

At this point, things can get a little complicated, and the various types of vapour systems begin to diverge -- often rather radically. One of the problems was that with coal firing, it wasn't always possible to ensure that the fire was controlled, and the boiler pressure would rise enough to back water out of the boiler and into the dry returns. A bit tough on the boiler, and also on the heating. Many systems had a variety of gadgetry to ensure that if the pressure differential between the steam main and the dry return did start to rise above the desired level -- 8 ounces for a Hoffman equipped system, for instance -- the gadget actually allowed steam into the dry return, to control the pressure differential and make sure that the water didn't back out. These systems had their vent or vents on the dry return, near the boiler, placed so that when desired steam from the boiler could hit them and close the vent. On such systems, one doesn't want a vent anywhere else -- it would defeat the purpose.

So -- I suppose the real bottom line is that crossover traps aren't needed, if the radiators aren't throttled or if the system can maintain a deep vacuum between cycles. Otherwise, they do no harm and can help.

MountainSteam

@SteamHead,
I understand your conclusions about the library return and point E. I'll measure the vertical drop from the library line at vent B3 to the NWL when I return to the house, but for now, based on other measurements and photos, I measure it as 36", so I suppose it would be enough if the boiler operates at less than one psi (as I hope it will).

As far as the feed tank, it is something of a spaghetti bowl, and it is hard to get a good picture of it because the brick wall of the old coal chute (with a big auger feed still present) is within 4 feet of the boiler so I can't get back far enough to get the whole thing. But if you click here, you will see document with three pictures:

1. The current return tank and most of the piping network
   
2. Same, but with a sketch superimposed, which clarifies the pipe network and includes elements that were not visible in the picture.
   
3. One more sketch showing the entire network with some notes about the water flow.
   
4. My imagined fix (which involed adding traps to the library radiators so that the return could drip into the tank). You can ignore this slide.
   

I hope this helps.

@PMJ,

I'll have to digest the concept. If I understand correctly, the mains feed the radiators, which let the air out and trap the steam. Meanwhile, the end of the mains is simply piped back to the boiler? And once the radiator's traps are open and the condensate enters the return, doesn't air enter the radiator, and then the inlet valve and then the mains?

PMJ

I see what you are saying Jamie and I agree there are lots of different vapor systems.

I would, however, experiment with plugs and/or caps there as long as I had to remove and replace anyway. To me it is all about that initial fill which I don't think needs those traps. After that cycle the one thing I do know is that nothing is better than keeping air out of that main altogether. My system is 90+ years old and is pretty tight. You don't need a perfect seal. My cycles are 20 minutes apart. The vacuum is typically 50" of water and still deepening when the boiler fires again. Any amount of vacuum helps a lot.

PMJ

MountainSteam said:

@PMJ,

I'll have to digest the concept. If I understand correctly, the mains feed the radiators, which let the air out and trap the steam. Meanwhile, the end of the mains is simply piped back to the boiler? And once the radiator's traps are open and the condensate enters the return, doesn't air enter the radiator, and then the inlet valve and then the mains?

In my case both the steam main and dry return are complete loops all the way around. They slope away from the boiler and at the far end both drop to the wet return joining below the water line. So the boiler feeds the loop both ways. I think this is better than the one way mains I see mostly but the far corner of the loop is still a dead end so to speak from which air (or steam) must flow out through the rads. But I believe the ended main loop like yours is fine too. My experience is that if the steam flows fine with a full main then air will flow fine too the same way through the rads and out the dry return. I did it this way for years with no vacuum.

And yes, you are right, with a vented dry return the air flows back through the traps, rads, and into the mains. This is what I mean by running backwards. So I just close my one vent in the dry return when the boiler shuts off which stops all that. The collapsing steam induces a vacuum in the system. The dropping pressure drops the boiling point so more steam production continues in the boiler than would other wise. I end up with a very slightly higher pressure at the header than the dry return and steam continues to flow slowly from the mains to the rads as opposed to the other way if the dry return vent was allowing air back in.

Eastman

I support PMJ's cycle control theory.

PMJ

Eastman said:

I support PMJ's cycle control theory.

I appreciate the support.

It runs a bit against some of the conventional wisdom I know and takes some sinking in. Air is the enemy. Positive pressure is the enemy. Cycles are not the enemy. Push air out once and try to keep it out. Every little bit helps.

Jamie Hall

In principle, so do I. In fact, by creating an hermetically sealed system, using water/steam as the heat transfer medium, one could achieve results similar to the very best variable temperature hot water systems, but without using pumps (the principle is exactly the same to so-called "heat pipes", which use heat transfer media with vapour pressures very close to atmospheric at the desired transfer temperatures). The devil is, as usual, in the details. In this case, the not so minor detail of maintaining moderate to high vacuums over any extended period of time --and maintaining that system integrity over many years of operation.

PMJ

Jamie Hall said:

In principle, so do I. In fact, by creating an hermetically sealed system, using water/steam as the heat transfer medium, one could achieve results similar to the very best variable temperature hot water systems, but without using pumps (the principle is exactly the same to so-called "heat pipes", which use heat transfer media with vapour pressures very close to atmospheric at the desired transfer temperatures). The devil is, as usual, in the details. In this case, the not so minor detail of maintaining moderate to high vacuums over any extended period of time --and maintaining that system integrity over many years of operation.

That's just the thing here Jamie. My piping is 90 years old. I have done nothing special to create 50+ inches of vacuum. Original valves everywhere. Most importantly, because I cycle 3 times an hour and have an average burn time of about 8 minutes I don't need to hold vacuum more than 12 minutes or so to take advantage of it. All the vacuum would be gone in say 4 hours without a burn. But I am firing into 50+inches every time. It takes 3-4 minutes into each burn for the new steam to get the system back up to atmospheric so in total the system is in vacuum about 60% of the total elapsed call time and all for free. And remember - my calls go on for hours on purpose. Long calls = even heat. Even at half this vacuum and half this time in vacuum I am saying it is way better than vented.

Quite honestly it was Igor who finally convinced me to put an air compressor on to look for leaks this past summer. I was too afraid for fear of damaging something. He was right though and the biggest culprit was the huge original shutoff gate valve over the boiler. A little tightening of the packing and voila - double the vacuum, more even heat yet and just plain dead quiet operation.

More people should try this - especially 2 pipers.

jumper

Do traps necessarily involve extra pressure drop? Isn't "heat pipe" the ideal?

Jamie Hall

Yes and yes. Traps do create a pressure drop, although it is very small (I imagine @Sailah could come up with the exact numbers). And yes, the heat pipe concept is ideal. The problem with the heat pipe -- for us knuckle draggers -- is that the medium we are using is water, and most of the time we would like to have both the source temperature (vapourizing or boiling the water) and the sink temperature (condensing in the radiation) a good deal below 212. Which means maintaining the system at a variable vacuum, while at the same time maintaining the vapourizing rate at the level required to heat the space. There are other media, of course, which could be used -- but most of them are either toxic or combustible or both (or have environmental problems -- the various chlorofluorocarbons -- the Freon family -- have some very desirable properties; ask the heat pump boys -- but some rather unfortunate ones as well).

The approach used currently -- the boiler on part of the time and off part of the time -- isn't particularly elegant, but it does have an advantage for most systems in that it is simple and reliable -- and it accomplishes the desired effect, which is that the radiation provides the needed amount of heat to the space, and the boiler provides the needed amount of steam to the system and does it with only two simple controls: the thermostat and the pressure control on the boiler. With properly sized components, and proper controls, it works -- and misses the theoretical maximum efficiencies by very small margins. One runs into cost/benefit problems!

I admit to being intrigued by @PMJ 's work and almost tempted to try it -- all it would require would be swapping out the Gorton #2 main vent for three Hoffman 76's -- but I seriously question how long it would take to recover the investment...

PMJ

Though I feel like pushing up my estimate of how much my methods improve efficiency Jamie, for me it is about improved comfort and that is surely significantly improved.

Jamie Hall

PMJ said:

Though I feel like pushing up my estimate of how much my methods improve efficiency Jamie, for me it is about improved comfort and that is surely significantly improved.

Aye, well... my trustees are vaguely interested in comfort -- but a lot more interested in dollars!