Why are gas boilers (for one pipe steam) either on at maximum output or off w nothing in between ?

ethicalpaul

Sorry but I disagree in your presentation. That is a drawing of the state that occurs AFTER the system has been shut down due to pressure. My system in normal use doesn't shut down due to pressure. I'm not sure we're talking about the same thing anymore so I'll exit here.

SlowYourRoll

ethicalpaul said:

Sorry but I disagree in your presentation. That is a drawing of the state that occurs AFTER the system has been shut down due to pressure. My system in normal use doesn't shut down due to pressure. I'm not sure we're talking about the same thing anymore so I'll exit here.

The book describes this as the state BEFORE the system is shut down due to pressure. The sentence immediately after this drawing states: "what we have now is a boiler that's under pressure and a radiator that's in a partial vacuum. It's an ideal situation for steam movement (high pressure to low pressure). The problem, though, is the radiator is full of air instead of steam."

He's very clearly talking about the state of the system before the p-trol cycles. He's saying that this state is why the p-trol cycles. We can disagree about many things, but we should at least be able to agree on what Dan is saying. And the whole point of what Dan says there requires this drawing to be the state of the system before it cycles.

ethicalpaul

If it's before the system has cycled off, then how did the air get "expanded" (lower density/partial vacuum)?

Jamie Hall

Go back and read my bottom line comment just above. Please...

But some bits and pieces. A radiator -- unless its inlet valve is closed -- cannot be at a pressure much below the boiler pressure. There simply isn't enough flow resistance in the pipes to do that. It will be at a slightly lower pressure -- a few ounces, perhaps -- but not more than that -- and then only if steam is flowing. I'm sorry to have to differ with what @DanHolohan said, if that's actually what he meant, but I do have to differ.

Now let's consider the dynamics in a one pipe steam system again. We'll make it simple. Not only one pipe, but one radiator.

Starting from cold. The boiler fires, and after a bit starts to make steam. The system as a whole is at or very very close to atmospheric pressure, and the radiator vent is open. Now as steam is created, the pressure increases -- of necessity -- slightly, and air will start to move out of the vent, since it has to either do that or compress. This will continue, with more and more steam being produced and more and more air being expelled from the vent. The pressure, mind you, is still low, and is controlled by the resistance of the vent to air movement and the pressure loss due to friction in the pipe.

At some point in the process, steam reaches the vent and its thermostatic element closes. Now what? Well, if the radiator can condense steam as fast as the boiler can make it, essentially nothing. The boiler boils, the steam moves, the steam condenses, the radiator radiates the heat, the condensate returns and the process will continue that way indefinitely without outside (e.g. thermostat) intervention.

So... what happens if the boiler is enough oversized that it makes more steam than the radiator can condense? The pressure rises, and the pressuretrol shuts off the burner. Now there is no steam movement, and the pressure in the system will equalize -- very rapidly, at nearly the speed of sound -- throughout the system. The steam condenses, also very rapidly but not instantaneously; the rate is governed by the rate at which heat is radiated from the system. As the steam condenses, the pressure will fall. At this point, one of two things will happen: the pressure will drop low enough that the pressuretrol fires up the burner again, and steam will be produced again, and we are back right where we were before. However, if the pressure drops to at or near atmospheric, the vent will reopen and air will be drawn back into the system -- if, and only if, the pressure in the system is below local atmospheric, even by a small margin (remember -- no flow, pressure in the system as a whole must be uniform). Then when the system fires up again, that air will be expelled from the now open vent at the rate determined by the vent. Eventually, the vent will see steam again and close, and we are back to where we were.

OK. Hopefully, clear enough for one radiator. But what happens if we have two, one of which is in a cold room with a nice fast vent, and the other is in a warm room with a very slow vent?

Up to the point where the vents are closed, there is no change -- except that the fast radiator has been full of steam for quite some time before the slow one gets there. Now the slow vent closes. To shorten the story, the pressure rises, the burner shuts off, and the steam collapses and if the pressure drops low enough the vents open and the air come in. The system fires again. Again, the fast radiator gets steam faster -- that's why we have the fast vent on there -- and delivers more heat. Eventually the slow radiator also gets there -- but will have delivered a smaller fraction of the heat of which it is ultimately capable than the fast one, since it has filled with air less rapidly.

By allowing the boiler to cycle, in short, we have maintained control of the heat output of the radiators. The action of the vents allows air in between cycles, and that air must be released less rapidly in the radiator with the slow vent, which will, therefore, deliver less heat.

If the boiler didn't cycle, the output of the radiator over time is not controlled -- or controllable -- by the vent, but only by the size of the radiator. This is not desirable.

It is all, also, fundamental physics. Think it through.

SlowYourRoll

ethicalpaul said:

If it's before the system has cycled off, then how did the air get "expanded" (lower density/partial vacuum)?

from the book: "when steam hits the vent's float, the alcohol/water mixture flashes to vapor. That pops out the bottom of the float against its base, driving the needle into the vent port and closing the vent. While this is happening, the steam is condensing inside the radiator and shrinking to about 1/1700th of its original volume. This condensing process creates a slight, naturally induced vacuum inside the radiator. That vacuum draws more steam toward the radiator. Now try to visualize this: in the presence of the slight vacuum, the air that wasn't vented on the first cycle expands until it completely fills the radiator. The air does this because it's a gas. A gas will always expand to fill the space allowed for it."

that's right out of pages 176-177 of The Lost Art Revisited.

Jamie Hall

OK, folks. I give up.

I'm not a Biblical Fundamentalist, either.

SlowYourRoll

Jamie Hall said:

OK, folks. I give up.

I'm not a Biblical Fundamentalist, either.

Jamie, i will go through and read everything you wrote a little later on. sorry i didn't respond yet.

KMSNYC

What would happen - theoretically speaking - in a one pipe system where the air vents closed on heat and opened only when they cooled, which sort of describes their basic operation? In other words if they didn't close because of pressure. Can't quantify this at the moment but blowing air into an air valve (at room temp) held vertically in the correct position, air blows through the valve as expected - but if blow quite hard the valve shuts off. it takes a fair amount of pressure, I would say, to do that. I suppose the manufacturers publish data on this ?

KMSNYC

Also - I believe that heat overrides everything else on an air vent (unless the pressure is so great as to destroy the valve). In other words a hot (enough) valve is always closed. But a cool valve is not always open - if there is enough pressure in the radiator even if that pressure is "cool" the valve will close, it seems.

jumper

Presumably steam pushes air down? So below vent there is always steam?

SlowYourRoll

Thanks for your explanation, @Jamie Hall. i think i understand your explanation and Dan's explanation now, and how they differ from one another. the two of you are obviously more knowledgeable about this stuff than me so i doubt i'll have much luck trying to get to the bottom of it. and from a practical standpoint, you two might have differing reasons for setting the p-trol to 1.5/0.5, but if you both ultimately recommend the same settings than i guess it really won't matter in the end when i get a new boiler in a few months. i guess i'm just frustrated (not at any of you) cause i was really hoping to gain a good working knowledge of steam, and already there's this confusion about one of only two control loops in the operation of the system, the other being the thermostat control loop (i'm not counting LWCO or things like that, they are also control loops but don't control the normal operation of the system). basically i was hoping to move away from "my boiler takes in water and then some magic happens and then my house gets hot." oh well. anyway, thanks for taking the time to write everything out. i do appreciate it

cgutha

How can one properly size a boiler When the house is designed to keep warm at 30 degrees below zero (design temperature), then in a winter like this most of the time it is twenty or thirty above? Sure the total EDR of the building is the same, but the heat load is only one third of design temperature. Obviously, the boiler is oversized in these conditions. a modulating or duel heating system would be nice.
(i will now go back and read the rest of the discussion)

PMJ

cgutha said:

How can one properly size a boiler When the house is designed to keep warm at 30 degrees below zero (design temperature), then in a winter like this most of the time it is twenty or thirty above? Sure the total EDR of the building is the same, but the heat load is only one third of design temperature. Obviously, the boiler is oversized in these conditions. a modulating or duel heating system would be nice.
(i will now go back and read the rest of the discussion)

Bravo - quite right. In average conditions all boilers are way oversized 2-3 times anyway. And unused EDR downstream of what is currently needed has no effect on anything. The original coal boilers were quite large, and relatively small modulation in the middle of their capacity range would be a more significant percentage change in filled EDR - most of which was quite obviously not used on an average day.

With regard to the current discussion the whole practice of totally filling design day EDR enough to close vents at any time on an average day makes little sense to me. Spaced cycles even with very big boilers avoids pressure and totally filled radiators and closed vents altogether and runs the system much more like the original design.

Jamie Hall

Don't give up, @SlowYourRoll ! It think one of the problems we have in thinking about steam is that we tend to lump all steam heat together, and treat it as though it was all the same.

Which it isn't. Other than using steam as the heat transfer medium.

One pipe steam in its usual configuration (radiator vents) and the related, much earlier, two pipe with air vents steam (very rare now) is in terms of control and management a completely different critter from standard (1.5 to 2 psi) two pipe steam, which is actually remarkably different from low pressure differential steam (vapour systems).

And vacuum assisted systems, either ones such as the Paul system and its friends or two pipe vacuum assisted, are different yet again -- and within them there are very significant differences between natural vacuum systems and induced vacuum systems.

Solutions -- or understandings -- related to one of them are more or less inapplicable, if not flat out incorrect, when related to others.

About the only thing which is common among them is that the total installed radiation must be capable of meeting the building heat loss on the design day -- and the boiler must be capable of meeting the demand of the radiation. Beyond that -- particularly in legacy structures where the present design day heat loss may be much less than the original design day heat loss, and hence installed radiation capacity -- it becomes a matter of control strategy -- and, as I say, a control strategy which works splendidly well for one type of system may work poorly, if at all, with another.

PMJ

I must agree that there are different steam system designs with different approaches. Fortunately, whenever someone comes here looking for help there is only one system to talk about...theirs. And so if there are particulars about that system that need to be related to the laws of physics about moving steam around those specific things are easily separated out from all the noise and discussed.

The boiler must easily be able to meet the heat loss of the structure on design day. The boiler really does not need to be big enough to fully fill all the installed radiation, and most of us have a lot of extra. And about that, the higher the percentage of the total time the radiators are actually condensing steam, the lower the percentage of the EDR in them you actually need to use for any given demand, with continuous steam supply obviously using the smallest amount. The more you spread out the supply, that is the higher the percentage of the time you can get your radiators doing anything, the less of the EDR you ever need to fill.

The fundamental point I am making is not affected by the type of system. Continuously supplied steam uses less of the installed radiation by default. Continuously filling more radiation than is required for the demand would obviously just overheat the structure. So with a continuous supply rate matched to the demand the filled amount of EDR required for that demand is the smallest amount it will ever be. Intermittent operation increases the required max fill EDR amount to compensate for the zero supply times. Pressure is so bad for this project because once at pressure the heat input rate is basically maxed out. Once there, more pressure doesn't get you much more rate into the rooms. What it does get you is bigger overshoots and longer zero supply times. Bottom line, the higher the percentage of the time you can get the radiators condensing steam at all, the smaller the range of temperature they will oscillate through will be, and the less full they ever need to be. Pulsed operation reduces the amount of EDR needed and used. A smaller percentage of the radiator is hot a higher percentage of the time. One way to look at continuous steam supply is simply as an infinite number of infinitely short pulses. Once one starts thinking about the supply in shorter pulses one realizes how adjustable this whole thing can really be. The good news is that it really doesn't take that many more pulses to get away from pressure and make a big difference.

Put simply, the best and most even heat comes from using less of the EDR more of the total time, and certainly not from using more of the EDR less of the total time. Pressurized operation does the latter. It is really too bad so many systems are no longer balanced well enough to take advantage of PWM.

Jamie Hall

Sorry about the noise, @PMJ . I'll shut up.

PMJ

Jamie Hall said:

Sorry about the noise, @PMJ . I'll shut up.

Really didn't mean it that way @Jamie Hall . I do apologize. Plenty of my stuff can be considered noise too I'm sure.

Any comments about the specifics? I'm interested in the truth about those and much less about being right. Not being wrong about some things could only happen if I write nothing. Anyone pointing those things out for me does me a favor and speeds me along my way to real understanding. I give you credit for much of that over these years.

KMSNYC

If my understanding is correct of what PMJ is saying (and it may not be) is that in practical terms, what one would want is a larger number of "smaller" boiler cycles where the boiler cycles between a cutout at a relatively low pressure (whether that is 1 psi or 1.5 or whatever) and a cut in at say 0.5 psi . The "pulses" of steam that result will allow the radiators to "absorb" and radiate the heat better since they are given a greater period to condense the steam. Something like that ?

PMJ

KMSNYC said:

If my understanding is correct of what PMJ is saying (and it may not be) is that in practical terms, what one would want is a larger number of "smaller" boiler cycles where the boiler cycles between a cutout at a relatively low pressure (whether that is 1 psi or 1.5 or whatever) and a cut in at say 0.5 psi . The "pulses" of steam that result will allow the radiators to "absorb" and radiate the heat better since they are given a greater period to condense the steam. Something like that ?

Not quite.

Pressure, any pressure, can only be generated AFTER radiators fill totally and vents close. Prior to filling any radiators the pressure at the header may be only an ounce or so. My max and many others here is like is 2 inches of water. By the time you have reached 1psi your radiators are already way too full for what I am talking about.

Picture how full your radiators would need to be to heat your place if they were filled to the same amount and heating continuously 24/7. This is what would have been happening in the original design with a coal fired boiler. The radiators could be nowhere full and there would be no pressure at all - certainly not enough to do anything with a pressuretrol or vaporstat.

If you were to switch the firing to timed/spaced pulses you can run more like the original setup with only partly full radiators and no pressure at all. The vents would never close at all. It will not work well in one pipe systems that are not really well balanced with steam basically arriving at all radiators at the same time.

There is sadly no off the shelf control for this. It is DIY only so few are in a position to do it.

SlowYourRoll

I should probably post a bit of a follow-up...

@ethicalpaul and I went through that section of The Lost Art Revisited, and i'm fairly certain now that there's a big problem with that section. i can go through and detail where it seems to fall apart later (when i get a chance next week) if anyone is interested, but in the meantime just disregard probably everything i said in this thread, since i was basically just reiterating what the book said.

PMJ

I am interested.

Jamie Hall

I will be interested in what you all come up with -- but I have said all I'm going to about it.

SlowYourRoll

okay, this is all in the context of pages 174-177 of The Lost Art Revisited, so if you have that book it'll be easier to follow what i'm saying.

first off, this part is not in that section but will aid in the discussion...the pressure inside the radiator is going to be just slightly above atmospheric up until the moment the air vent closes. it can't be at exactly atmospheric because there needs to be some pressure difference to force air through the vent, but since the vent is open there's no way to build up anything but the smallest bit of extra pressure in the radiator.

so at the moment the steam hits the air vent, any trapped air inside the radiator is at roughly atmospheric pressure. the book goes on to state that at this point the steam collapsing in volume as it condenses creates a natural vacuum, and this vacuum has the effect of expanding the trapped air. that corresponds to the figure i posted above that shows "expanded air" occupying most of the radiator.

the book then states that there are two things holding the air vent closed. the first way is that the the alcohol/water mixture flashes when the steam heats it to about 180deg. the second way that the book mentions is that the pressure inside the radiator is enough to hold the air vent closed. the book notes that if you blow hard enough into an air vent you can feel it close. this second way is where things seem to break down in the book's analysis.

the book argues that this "expanded air" is being pushed by the steam into the air vent and holding it closed. the problem is that "expanded air" is essentially air at a lower pressure than its starting state, which in this case is roughly atmospheric pressure. but for pressure inside the radiator to push the air vent closed, you'd need that pressure to be greater than atmospheric.

so the way the book describes it can't happen. you could have "expanded air" taking up too much real estate inside the radiator, or you could have trapped air become condensed enough to push the air vent closed, but you can't SIMULTANEOUSLY have expanded air taking up too much room AND pushing the air vent closed.

(and as @ethicalpaul pointed out and Dan hints at when he says you have to blow "hard" into the air vent, it actually takes quite a bit of pressure to close the vent. that kind of pressure would not be seen in most residential heating applications.)

(my best guess as to what goes on is that the trapped air occupies its largest volume at the moment the steam hits the air vent and closes it, and once the air vent is closed the radiator can finally build up pressure, and so this trapped air would shrink in volume until the radiator reaches its peak pressure)

(as far as figuring out how/why this ended up in the book, Dan cites a passage from a book given in 1900 about how the trapped air would act like an air spring. i'm not sure what other references went into the preparation of this book section, but i think it's possible that maybe earlier air vent designs were susceptible to radiator pressures holding them shut, but that maybe subsequent designs figured out how to minimize this issue, to the point we are now where it takes quite a bit of pressure to hold an air vent shut. it may be that "drop away pressure" really was important at one point in history based on air vents of the time, and the term "drop away pressure" stuck around longer than the actual engineering problem it solved. this part is really all just speculation though. most of the air vent patents are classified in USPC 236/61 through 236/66, so if anyone wanted to follow the development of air vents over time, that'd be the place to go.)

SlowYourRoll

also, if anyone does actually want to flip through the old patents for air vents, just go to the USPTO Advanced Search site, choose "1790 to Present" from the "Select Years" drop-down menu, and then paste this in as your search string (without quotes): "CCL/236/60 OR CCL/236/61 OR CCL/236/62 OR CCL/236/63 OR CCL/236/64 OR CCL/236/65 OR CCL/236/66"

that's the spot for thermostatic air-relief valves. there's also 137/197-202 which is the spot for "diverse fluid containing pressure systems > fluid separating traps or vents > for discriminating outlets for gas." Also class/subclass 251/11 is "valves and valve actuation > heat or buoyancy motor actuated." and then class 237 "Heating Systems" covers more general inventions where an air vent can be a part of the system, so 237/68 is for air discharge in a steam system.

SlowYourRoll

if you wanted to search all those places the search string would be "CCL/236/61 OR CCL/236/62 OR CCL/236/63 OR CCL/236/64 OR CCL/236/65 OR CCL/236/66 OR CCL/137/197 OR CCL/137/198 OR CCL/137/199 OR CCL/137/200 OR CCL/137/201 OR CCL/137/202 OR CCL/251/11 OR CCL/237/68"

that's like 1300 patents prior to 1950 so that would take awhile to flip through them all, but anyway that's how you'd do it

KMSNYC

There are 2 different methods (as far as I know) that air vents use to close. The first is the one explained above - a water / alcohol mixture inside a "float" or "bellows" that expands when heated. when it expands it pushes up a "needle" into an opening that closes the opening. The Hoffman vents use this method. They are supposed to do this at 180 F. I think Vari-Valves use this technique also. The other method is by a bimetallic element that expands (or contracts) when heated. (I say expands or contracts because the Gorton valves use this method but the Texas Instrument bi-metallic element inside actually contracts when heated according to what their rep told me. In any case it moves when heated (210 deg F for the Gorton) and causes a device to push up and close the vent hole. The round shape of the vent when viewed from the horizontal gives away this bi-metallic method although there may be some of this type that are not round. I think there is a little ball in the Gorton that gets pushed up to close the hole. If these valves are functioning properly they will always close on heat which overrides other factors. They all use gravity to open - the heat gets less, the float falls and the vent hole opens. The purpose of the "float" is to close the vent hole in case of an inrush of water. I say "inrush" because this is not designed to prevent a small amount of spray from exiting the vent in case there's a bit of water trapped in there or very wet condensing steam. In my understanding , this does not happen often these days but in the days when people used to push the pressure up a lot and for other reasons (instant conversion to a hydronic system by a stuck low-water filler float will do it) pipes and /or radiators could occasionally become flooded and this float system would prevent a scenic, but potentially dangerous, Yosemite-style geyser from occurring in the living room. It's a safety thing. When one blows (hard) into the radiator connection end of these valves they will shut. That is your breath pushing the float up the same way an inrush of water would do. Turning the valve upside down also does this.

There is a pressure range associated with when the valves will open properly. It's the "operating pressure" or "drop-off" or "drop-away" or "re-open" pressure. I am not 100% certain that these are 100% the same exact thing, but I think close enough for practical purposes.
If the pressure is within the drop-off/operating pressure range, the valves will open when expected - that is, when things cool off. In the case of Gorton vents, if there is < 3 lbs. of pressure in the system and the vent is cool enough (such as initial air pressure upon startup of a cycle when the steam is “trying” to push the air out of the system) the vent will open. If pressure is > 3 lbs. then it’s ½ psi “drop off” pressure is relative; that is if system is running at 5 psi and it drops to 4.5, the vent will open (as long as it’s not hot). That’s largely theoretical since most system run at much lower pressure, ounces not lbs.

later I am going to crack open a few of these vents (old ones I have) and examine their guts & will report back. I imagine more than one person here has already done this.

One-pipe steam system, which are the only kind I've had experienced with, are (almost all) "atmospheric" systems, meaning they are meant to let air out when the steam is on its way, pushing the air ahead of it, close when the steam gets there, and re-open again when they cool off, letting air back in. That is the major part of what they do where air venting is concerned. They are not considered vacuum systems, primarily, and "vacuum systems" are distinguished as a generic type from "atmospheric" systems in the literature.

Regarding vacuum left in these otherwise atmospheric systems here is a passage from the Xylem/Hoffman book that may be relevant:

"One-pipe systems that operate at atmospheric pressure to pressures of about 2 – 3 psig use open type vents. As the steam in the system condenses on the off cycle, these vents allow air to be drawn into the system. This air is vented on the next firing cycle. One-pipe systems which operate in the vacuum range during a portion of their heating cycle have been in use for many years. These systems operate well when coal-fired. When enough heat is present in the fuel bed, the system operates at above atmospheric pressure. As the fire diminishes; the rate of steam generation decreases. Check valves in the radiator vents prevent air from being aspirated as the steam condenses and a gradual vacuum forms in the system. The decreased pressure allows the continued production of steam at a lower temperature. The radiators remain warm over a longer period of time, resulting in good temperature control. The system, commonly called a “vapor vacuum system” does not lend itself well to gas or oil fired boiler operation. During mild weather the firing cycles may be too short to allow the steam to completely purge the system of air. On the off cycle, this air will expand as the system drops to a vacuum, often to the extent that it will creep into the mains. Repeated short cycling in this fashion will result in an air-bound system with poor heat distribution. For this reason, the use of vacuum type radiator vents with one-pipe oil or gas fired systems is not usually recommended."

KMSNYC

https://documentlibrary.xylemappliedwater.com/wp-content/blogs.dir/22/files/2012/08/TES-375B-Steam-Heating.pdf

ethicalpaul

FYI you can edit your post (which I do often) (edit: ahh I see you found it)

On the off cycle, this air will expand as the system drops to a vacuum, often to the extent that it will creep into the mains. Repeated short cycling in this fashion will result in an air-bound system with poor heat distribution. For this reason, the use of vacuum type radiator vents with one-pipe oil or gas fired systems is not usually recommended."

This concern is kind of ludicrous, as if it's strange for air to enter the mains of a one pipe system. My one pipe system is "air bound" every time it fires up. Then the air gets pushed out by steam at 1 inch of water column. How could it be worse to be under partial vacuum in the mains than it is to have full atmospheric pressure in the mains?

KMSNYC

I think what they are talking about is a kind of system they refer to as a “vapor vacuum system" which is specifically designed to take advantage of (relative) vacuum that can form in the radiators, as described. That kind of system uses air vents with check valves that do not let air back in at least not immediately. In my limited experience I have not encountered that type of system, and according to the article they only work properly with coal and / or can have problems with modern boilers. I suppose if one wanted to speculate such a system might work with an oil or gas fired boiler that adjusts its fire level like a coal burner. I posted the article because it discusses how the partial vacuum can form. Under normal circumstances in a one-pipe atmospheric system the air lock won't happen. I am not certain, but I surmise (from Dan's books) the way it could happen is if the pressure in the system were high enough to keep the air vents from re-opening but we are talking about a higher pressure than most systems use (consistently > air vent operating pressure) esp. when they cycle off. He says keeping things at low pressure esp. the cut-in prevents this. I s'pose this means if a system were to cycle between 4-5 psi it could happen and he says he has encountered it.

PMJ

ethicalpaul said:

FYI you can edit your post (which I do often) (edit: ahh I see you found it)

On the off cycle, this air will expand as the system drops to a vacuum, often to the extent that it will creep into the mains. Repeated short cycling in this fashion will result in an air-bound system with poor heat distribution. For this reason, the use of vacuum type radiator vents with one-pipe oil or gas fired systems is not usually recommended."

This concern is kind of ludicrous, as if it's strange for air to enter the mains of a one pipe system. My one pipe system is "air bound" every time it fires up. Then the air gets pushed out by steam at 1 inch of water column. How could it be worse to be under partial vacuum in the mains than it is to have full atmospheric pressure in the mains?

Quite right @ethicalpaul.

I do this every cycle under vacuum in my two pipe and have for many years. After delivering every last bit of residual heat in the boiler enough to make steam after the burner goes off steam finally ceases flowing into the rads with vacuum increasing the entire time. All during this process air already in partially filled radiators and the dry return(all initially at atmospheric pressure) expands as the pressure drops, eventually back filling the rads completely and on into the steam mains.

On every next firing all this "expanded" air is a lot easier to "push" back into its original atmospheric volume with the system vent closed than it would be to push the same volume of air at atmospheric pressure out of the system with the vent open. The difference in effort at low non-pumped vacuum levels is not dramatic, but I say for sure that the lower pressure the filling steam finds ahead of it makes the flow easier not harder. It most surely does not block the steam and prevent it from flowing as is suggested here.