Short Cycling on Vapor System Boilers?

Harry_6

Greetings. I have had a problem on several occasions with boilers installed on vapor systems with orifices, and set to top out at around 8 oz pressure. It appears that the boiler (specifically smaller Utica JE series) is able to pressurize the main pretty quickly, and then short-cycles as the steam is metered to the cold radiators. Once everything is warmed up it settles down, but still, this is not desirable.

In one case I placed the vaporstat at the end of the main with a thermostatic sensor on the pipe which was wired to short past the vaporstat and rely on the boiler pressuretrol set to a couple psi to manage things until the end of the pipe was hot and the vaporstat could take over. In this case the trick worked fine, but other times I've had no luck.

I guess ideally this boiler would have a dual pressure regulator (which I have done elsewhere on a larger boiler - but I didn't install this one, and I don't even know if hi/lo is available). Does anyone have a suggestion for a way to underfire it until it needs full capacity? Or some approach I haven't thought of?

Jamie Hall

If the boiler is doing that -- any size, any make, any pressure, your main venting is woefully inadequate.

Can you tell what type of vapour system this is? Does it use crossover traps (most do) and, if so, are they operating correctly? Where is (are) the main vent(s) located? In almost all cases, it (they) should be clustered where the dry returns join before dropping to the boiler, if it uses crossover traps, or at the ends of the steam mains if it does not. What is (are) the main vents?

Harry_6

Good point. In this case the type of system is kinda irrelevant. It started life as a Mouat, but every radiator valve and trap was stolen, so I inherited piping. And the piping had been messed with by at least two generations of folks who didn't know what they were doing. Returns had been moved, water traps created, unnecessary F&T's, installed, etc. I had to go straighten that mess out first, along with sizing replacement radiators (luckily an old radiator survey existed), new valves, traps and orifice plates. I ended up with a 3/4" low-pressure check valve as the main vent for each half of the system. It looked like there never was a vent at the end of either steam main, crossover or otherwise, unless it was one of the pipes capped long ago. I put a crossover at the end of one main but not the other, so perhaps that might help, but the vaporstat was originally on the boiler and hunted constantly - thus the move. But the current main venting is just those open 3/4" checks on the ends of the dry returns. No thermostatic element at all. You think another crossover is in order?

Jamie Hall

Another crossover surely won't hurt. Check valves work... sort of... as vents, but many of them have cracking pressures which are higher than what you want to run the system at. I'd switch to honest to goodness vents, and I'd pick Gorton #2s.

Vapourstat hunting can be a real problem, particularly if it is placed in the usual on the boiler position. They are sensitive enough so that you get nuisance trips as bubbles form and break. The end of the main idea isn't terrible, but it's a little far from the boiler. The vapourstat for Cedric is on the header (along with the low pressure gauge) and is quite stable and happy there, but you might also be OK with just a pressure gauge snubber like this one: https://www.supplyhouse.com/Winters-Instruments-SSN515-1-4-Brass-SSN-Water-Snubber-10000-PSI
which will damp out the spikes giving you nuisance trips.

Harry_6

Yeah, some of these things I'd considered. I'm using Atlas vertical checks with the plastic check and the spring removed, so the cracking pressure is only ounces (it took me forever to find something that would work - thanks to whomever on the board suggested it!). Their capacity is large, and I didn't want to go with a non-vacuum system, which is what Gortons would require. I thought of a snubber, but the short cycling isn't "hunting", The boiler will fire and I can watch the pressure (on a 20 oz gauge at the vaporstat) rise to 8 oz over about a minute, then it shuts off, and it takes another minute or two for it to drop back down to the 2 oz lower setpoint. I think a snubber would only stretch that out a little.

Now in a perfect world, where I had a low and high fire burner, it could drop down to low and just simmer along. I did this on a bigger boiler (see my "Symphony of Steam" boiler installation time-lapse on Youtube) and it works like a charm on an intact Mouat system. It will warm up and produce steam just as fast as it is consumed, and simmer at about 2 oz indefinitely. Oh, for variable output burners!

JStar

Robertshaw makes a 700 series High/Low gas valve which you can trigger with another vaporstat. It requires combustion tuning after installation.

Harry_6

I may just have to think in terms of using High/Low regulators on Mouat and similar when the boiler is of any large size. Additional cost, but the result is so much better.

PMJ

Harry_6 said:

I may just have to think in terms of using High/Low regulators on Mouat and similar when the boiler is of any large size. Additional cost, but the result is so much better.

How about thinking about a PWM approach instead. Less expensive, less complicated, more effective, more adjustable.

I have a Mouat system with a big boiler. Those systems should be running at max pressures way below anything a vaporstat could do anything with anyway and it can be done easily no matter how oversize the boiler. I have at least 2 times the boiler that is needed and get it all done under 2 inches of water pressure at the header. Just spacing the firing out a little is all it takes.

PMJ

Max pressure I ever see on my Mouat and big boiler.

Conditions during this photo:

-30 degrees outside
-56 minutes into a call
-system currently doing about 8 minute burns and 20 minute
waits or just over 2CPH
-photo is right at the end of a burn and vent on dry return is
open
-Magnehelic shows difference between header and dry return.



Next photo is right at the end of 20 minute wait in deepest vacuum. Header still higher pressure than dry return. System pushes steam to rads always. Much more efficient.


Not complicated. No PLC needed. Why does everyone spend so much time wringing their hands about big boilers and pressure short cycling? Why is anyone running any pressure at all? I don't get it.

Jamie Hall

Just for fun... pressure cycling -- if it is a low enough pressure -- is a form of PWM. That said, @PMJ (mostly to rattle your cage), have you experimented with hooking your controls up with outdoor sensors? To vary the PWM settings? Seems to me there might be an opportunity there...

PMJ

Jamie Hall said:

Just for fun... pressure cycling -- if it is a low enough pressure -- is a form of PWM. That said, @PMJ (mostly to rattle your cage), have you experimented with hooking your controls up with outdoor sensors? To vary the PWM settings? Seems to me there might be an opportunity there...

I agree it is a form of PWM - just far less efficient.

I have taken a different approach trying to keep sensors to a minimum. Now I only use a preheat sensor on my most remote rad in the garage. So every burn runs until that is satisfied to solve the cold start problem. From there I have two timers (in the PLC but don't have to be). From the satisfied preheat close I run 2 more minutes and partial fill everywhere. The next burn is initiated when the same sensor opens plus 4 minutes on another timer. I do this in a PLC where it is easiest by far but this could be hardwired with a delay off/delay on timer for those terrified of PLC's.

The cycles now adjust themselves. As it gets colder and steam condenses faster the waits get significantly shorter. Because the wait is shorter the time to the next preheat close from fire is shorter too. In this way the CPH just increases gradually all on its own. Calls go on longer and longer. I haven't touched anything in 65 days now.

Here is a photo of my PLC display showing the length of the current call and the length of the last run or wait. They change themselves to the conditions. The radiators very gradually increase in fill the longer calls go on never getting anywhere near full. Average call length in moderate conditions is 75 minutes. In very cold conditions they will be 3-4 hours. We haven't had any really cold yet but if I have to I will up the 2 minutes a little and/or reduce the 4 minutes a little and then I will know all conditions are covered.

Harry_6

Ah, just what I was hoping for. Spirited discussion between experienced individuals!

Jamie Hall

Nope. Sorry. @PMJ and I have been at this one for several years now -- I say that using a vapourstat as a form of PWM control works as well as his approach, and he says that his approach if far more efficient (but hasn't managed to convince me). And we remain friends.

PMJ

Jamie Hall said:

Nope. Sorry. @PMJ and I have been at this one for several years now -- I say that using a vapourstat as a form of PWM control works as well as his approach, and he says that his approach if far more efficient (but hasn't managed to convince me). And we remain friends.

This is a very accurate assessment of the standoff.

In @Jamie Hall 's defense (and others here I am sure) , without having actually experienced the performance difference myself I doubt I would believe what I am saying about this either.

I do wish more could experience it. I have done it both ways and it really isn't close.