Let me try this again, this time with a 1" supply pipe. Surely that will cause trouble!

Nice experiment Paul. I'm just a lowly homeowner/engineer who doesn't have any experience with steam heat, but I have read LAOSH. As I understand it, the whole question of near-boiler piping size is an attempt to regulate the exit velocity of the steam as it leaves the boiler.

Dan talks about this question on pp 42-43 of LAOSH, and he quotes early steam guru Ara Marcus Daniels from 1928 saying "Steam boilers are not provided with means for assuring dry steam…hence velocity of flow through outlets of such boilers should be relatively low—not over 15 feet per second—if the carrying over of water with steam is to be avoided." Dan then looks at specs of some older boilers and lists the exit velocities for them. All but one are below 15 fps.

Then Dan notes that a book from 1936 says excessive steam velocity will interfere with the backward flow of condensate. And you showed that in your video with the condensate film on the walls of the sight glass being blown upward by the steam velocity. Dan follows that book quote with a table showing the maximum recommended velocity for different size riser pipes on a 1-pipe system, based on condensate flow. The max recommended velocity for a 3" riser is 29 fps, and Dan observes that Ara Marcus Daniel's 15 fps max is only half that, which is very conservative.

Dan then looks at the specs of some modern boilers. The 7 modern boilers he choose have exit velocities ranging from 25 to 37 fps, in some cases more than double Daniel's 15 fps max. Dan then observes that this means most modern boilers have exit velocities that exceed the max velocity against which condensate can drain, meaning that the headers and equalizers must separate out the condensate before it gets out into the system.

But it seems what your video showed is that, if no water droplets are being generated inside the boiler by surging, the exit velocity doesn't really matter. As Dan showed, modern boilers already have exit velocities above the condensate backflow limit, so they're already pushing the condensate up the walls even with correctly-sized near boiler piping. And as you showed, even higher velocities don't matter if there are no water droplets leaving the boiler in the first place.

It sounds like this is another instance of a very conservative "rule of thumb" (don't exceed 15 fps) having been established back in the day when boilers may have been more likely to generate water droplets due to poor water quality, maybe different boiler geometry, maybe less control over water lines levels, etc. Apprently modern boiler mfrs have learned that they can safely exceed 15 fps but are still trying to keep velocities reasonable enough so that if there is surge, keeping the velocity down minimizes the effects.

It reminds me of the Mythbusters episodes about dropping a penny off the Empire State Building. They built a small vertical wind tunnel to find the terminal velocity of a penny, which turned out to be around 65 mph. or about 95 fps. The Ara Marcus Daniels' 15 fps limit is presumably the terminal velocity of a water droplet, so if the surrounding steam is traveling slower than 15 fps, the droplet will fall, and if higher than 15 fps, the droplet will rise. But as with many old heating "rules of thumb," this may be conservative.

What I take away from your video is that, if the boiler isn't surging or producing water droplets some other way, excessive steam velocity isn't a problem. So the important factor is having water clean enough, and the water line low enough, that water doesn't get carried out of the boiler exit in the first place.

Dan mentions seeing many steam guys using anti-surge tanks. I assume like this:

Does anyone ever recommend those to homeowners with surging problems? That might be less costly than replacing all the undersized/badly installed near-boiler piping. Because as you showed, if you have clean dry steam, pipe size doesn't matter much.

I would appreciate if someone clarified what good water quality means. If it means no oil on the water surface and a PH above 7 then I have it in my 63-03 boiler. The next term I seek clarity on is surging. If it means the water line fluctuates more than +/- 0.25” when the boiler is operating then I have no surging, but my water line is approx 2.5” from the bottom of the sight glass. Does this mean I have significant carryover? I do get a fair amount of rust despite using 8-way. I am using distilled water to refill after draining the rust once a month. Could the carryover be caused by suspended rust particles?

It is important to note that none of these conditions are causing any heating issues.

Lastly, if undersized takeoffs cause carryover due to a high steam velocity, then Bernoulli tells us that oversized pipes should lead to a lower carryover at the same pressure. In my case I have a 3” single takeoff to a 2” header. The spec is 2” takeoff and 2” header. The 3” takeoff gives me a 44.4% greater cross sectional area, consequently this should result in a velocity that is 44.4% of the velocity with a 2” takeoff. Yet, despite a 55.6% reduction in exit steam velocity, I get wet steam- steam that is not 100% vapor.

But wait, that is not all. My 3” takeoff is 25” tall from the boiler cover to the header, which is at least 10” higher than the min (24” from the boiler water line). I point out this because I thought that a lower exit steam velocity and a taller takeoff would result in negligible carryover, all else being equal. Alas, that is not the case. I know this might come across as a contrarian viewpoint but it is what it is- a healthy mix of empiricism and stoicism.

I had an oversized Burnham that died after 34 years. It had a drop header and it too operated with the water level nearly at the bottom of the sight glass. It needed to be blown down every other week but no heating problems whatsoever. I am beginning to believe there are elves in every boiler.

My opinion is the header is there for a similar reason as the equalizer and Hartford loop.

The Hartford loop is to keep the wet return from siphoning out of the boiler. In order for this to work you need to break the siphon with a vent, the equalizer. Since we have that there anyway, we might as well pipe a header into it that will drain any water that blows out right back into the boiler rather than it having to make the long journey down the main(s) and back. Water leaving the boiler rapidly via the mains is also bad, so if we can slow it or prevent it…..

None of it is needed if everything is correct, but it all helps when things aren't correct.

I've personally watched an improperly piped boiler (no header at all) blow a ton of water up into the main and shutdown on low water. Would this happen if it had a proper header and equalizer, or would it all return quick enough to prevent the LWCO from tripping? @ethicalpaul Have you tested that yet? I can't remember.

I didn't see where @PhilKulkarnihad any problems at all?

Can you point out where because I keep missing it. The amount his water moves in the gauge glass seems perfectly normal, I think he was asking for confirmation more than anything.

The only question I have is where his water level is. Does it start out normal and drop to that, or is it always low?

Great video. What's next?

I just remember the Weil McLain video with the glass piped boiler that is still (probably) circulating on you tube it's probably 15-20 years old. That thing was throwing a lot of water.

It is a mystery I have seen horribly piped boilers with no kind of water treatment or blow down work absolutely fine. You stare at them and say it can't work but it does.

The ones that are piped right usually have no issues that can't be tweaked out.

But when they are bad you don't know where to start.

According to @ethicalpaul video we should try fixing the water quality first and if that doesn't work attack the piping.

I have seen boilers piped right that you couldn't keep the water in them until they were skimmed so maybe its true.

OK, so let's take @PhilKulkarni 's case. He says he has a 63-03 boiler, and with a 3-inch riser he's still getting wet steam. According to the following table, worst case velocity for a 63-03 is 14 fps for a single 3-inch riser. That's even below the probably-conservative 15 fps target, and yet he still says he gets wet steam. So there has to be some other factor like chamber geometry, water line height, etc, and/or a combination of the above, that makes some boilers more likely to generate wet steam than others, even at low exit velocities. For whatever reason, Paul's boiler doesn't seem to have that problem even with a high water line and high exit velocity.

It looks like this is one of the videos you mentioned, Ed. He lets the pressure build to 5 psi, and even with the burner off, when he opens the valve, the superheated boiler water flashes to steam and throws massive amounts of water up into the clear pipes.

That is a dramatic demo, but it doesn't really represent normal operating conditions which are more or less steady-state. During the steady-state portion of the video around 5:30, you can see the boiler water frothing through the glass window, and with the water line about halfway up the sight glass, the steam in the clear pipes appears dry. You can even hear someone say "that's dry steam." So again the question is, why do some boilers produce nice dry steam like that, and some boilers like Phil's produce wet steam even with oversized risers? Is it just water quality, and some people cannot or do not get enough oil out?

LOL. Sometimes boring is good. I was imagining what would have happened if one of the glass pipes on that demo Weil McLain boiler had cracked.

That was the classic Iron Fireman system. @gerry gill based his system on that.

and youtube started playing your other videos. is your foundation pyrobar or are they just large structural clay tile?

pyrobar is a block made out of gypsum. the blocks behind your laundry tub looked bigger than concrete blocks. pyrobar was commonly that larger format although i think structural clay tile was also made in that format too(structural clay tile is a clay tile block that is laid like concrete block but is usually different dimensions. that glazed block that was used in halls and bathrooms of mid century schools was a glazed form of structural clay time that was usually in similar dimensions to concrete block)

also the old style of concrete laundry tub that had the faucets come in through the back of the tub before there were cross connection prevention rules.