Mercury thermostat (Honeywell) may have caused short cycles and inadequate heating?

SteamBoiler

Background: 2 story home, two-pipe system; WeilMcLain EG-50 series 2 steam boiler with Honeywell PA404A-1009 pressuretrol set to about 1psi cut-in (assuming 1psi additive differential) in basement; in-wall steam radiators; Honeywell 2-wire mercury thermostat similar to the one in link, but heating only, heat anticipator was set to about 1 which is longer anticipation, no timed program, blue and red levers held together at 68F.

This is our 3rd winter in the home, had forced air heat everywhere we lived previously. We noticed that a couple radiators upstairs weren't heating at all or would only come on - barely - if we kicked up thermostat setpoint a couple degrees above where it normally was. Basement was usually warm, utility room with boiler always very warm. In general even the working radiators never got much warmer than warm.

For coolness reasons mostly, a couple days back I removed the old Honeywell 2-wire thermostat and installed a Nest 3rd gen Learning Thermostat (a 3rd unused wire ran up in the cable from the boiler transformer to thermostat location in wall, and I was able to connect C on the boiler transformer to this wire and use it as C for the Nest). I can now see the timing and duration of the calls for heat. When set to 68F, Nest starts call for heat at 67F and stops a little below 68F, thermostat location warms up over time to 69F - in its measurements, the room now ranges from 67F to 69F. During a 40F cloudy daytime, Nest made one 25 minute call for heat every 2 hours approx. We now have heat even on the radiators that I previously had suspected had steam trap problems (looks like previously the boiler wasn't running long enough in each heating call). The basement isn't that hot anymore and the utility room is only hot when the boiler runs. I stayed in the utility room and verified that the boiler was running for the entire duration of the heating call (not cutting out because of high pressure).

It looks like the old thermostat was causing short heating cycles and running boiler frequently. Does above make sense? With the old thermostat I have no idea of the exact temperature at which each heating call started and the duration of the call (never thought about it in those terms till I got the Nest). I might have been able to get the old thermostat to make a longer call for heat by setting a shorter heat anticipation but it is academic now.

A couple rooms are still hotter than the rest and I may need to reduce steam going into those radiators and get a general system tune-up involving checking all the steam traps. Any pointers for steam heating professionals in Union County NJ?

Hap_Hazzard

The correct anticipator setting for steam is 1.2, which is as high as it goes.

I switched to a mercury thermostat about a year ago, and I'm very happy with it. Most of the features programmable stats provide are either unnecessary or undesirable for steam heat, and there's no way on earth I'd want to connect my thermostat to the internet. Enabling 2FA can keep hackers from hijacking your account, but to me that's just more trouble than it's worth, and nothing can prevent it from going down. I subscribe to the old adage, "don't buy trouble."

EBEBRATT-Ed

Check for @EzzyT , @clammy or @JohnNY you can PM them all excellent steam pros. You heat anticipator in the old stat is the problem.

The Nest seems to have solved your problem although Nests are problematic with steam sometimes.

Work on your main vents to help speed the steam distribution and have your traps serviced and checked

SteamBoiler

Hap_Hazzard said:

The correct anticipator setting for steam is 1.2, which is as high as it goes.

Homes and steam paths differ, it looks like in my case I would have needed to use a smaller anticipator setting to run the boiler longer and avoid short cycles. Unfortunately I don't have the data to back this up though intuitively the explanation seems to make sense.

Hap_Hazzard said:

Most of the features programmable stats provide are either unnecessary or undesirable for steam heat

I somewhat disagree. I like the Nest data. The one thing the old style manual anticipators provide is the ability to tune in a final temperature range (limited probably to a few degrees), hence duration of heating call. With the Nest it appears that I am limited to +/-1F range and whatever heating call duration is necessary for this. But I can now imagine a setback at night by at least a couple degrees and recovering in small increments, ditto coming back from vacation to a warm home programming the recovery in small steps on the final day...

SteamBoiler

EBEBRATT-Ed said:

Check for @EzzyT , @clammy or @JohnNY you can PM them all excellent steam pros. You heat anticipator in the old stat is the problem.

The Nest seems to have solved your problem although Nests are problematic with steam sometimes.

Work on your main vents to help speed the steam distribution and have your traps serviced and checked

Thanks for these suggestions. Since we are in the thick of winter I plan to monitor the system and make sure it is consistent and I understand its operation. Will try to do some minor balancing myself and call the pros in spring to get the traps serviced. I don't understand the part about the main vents to speed steam distribution, will browse here to understand better.

Hap_Hazzard

SteamBoiler said:

Homes and steam paths differ, it looks like in my case I would have needed to use a smaller anticipator setting to run the boiler longer and avoid short cycles. Unfortunately I don't have the data to back this up though intuitively the explanation seems to make sense.

To make the boiler run longer you need a higher anticipator setting.

SteamBoiler said:

I somewhat disagree. I like the Nest data. The one thing the old style manual anticipators provide is the ability to tune in a final temperature range (limited probably to a few degrees), hence duration of heating call. With the Nest it appears that I am limited to +/-1F range and whatever heating call duration is necessary for this. But I can now imagine a setback at night by at least a couple degrees and recovering in small increments, ditto coming back from vacation to a warm home programming the recovery in small steps on the final day...

Setbacks are generally not so good for steam systems. It might make sense if you're going to be gone for a week or so, but I wouldn't want to do that to my cats. If that wasn't a concern, I'd just turn the heat back up when I got home and the house would be warm by the time I finished unpacking the car.

SteamBoiler

Hap_Hazzard said:

SteamBoiler said:

Homes and steam paths differ, it looks like in my case I would have needed to use a smaller anticipator setting to run the boiler longer and avoid short cycles. Unfortunately I don't have the data to back this up though intuitively the explanation seems to make sense.

To make the boiler run longer you need a higher anticipator setting.

My anticipator was already at 1.0 of a maximum of 1.2 which is why I thought I needed to set it lower. It is of course possible that the anticipator was bust in which case the home would never get to the target temperature :-( Edit: which is why again I like the Nest, you can't really read off temperatures properly from a needle thermostat.

Hap_Hazzard said:

Setbacks are generally not so good for steam systems. It might make sense if you're going to be gone for a week or so, but I wouldn't want to do that to my cats. If that wasn't a concern, I'd just turn the heat back up when I got home and the house would be warm by the time I finished unpacking the car.

My understanding with setbacks on steam systems is that they force the boiler to run continuously for much longer than normal to get to "normal" heat, which is why I was talking about gracefully recovering from setbacks with the programmability of the new smart systems. Looks like I will be using it soon...

Jamie Hall

I'll chime in here... it is possible that the anticipator was open. That, paradoxically, would have made the cycles longer, not shorter -- and it would have overshot the setting. As @Hap_Hazzard said, it should have been set at 1.2 . But that's water over the dam now.

Recovering from a setback will take almost exactly as much fuel whether you do it in very small steps or in one giant leap. One giant leap will, if anything, take less, as the boiler and piping and radiators won't have a chance to cool off and need to be reheated. But -- take your pick.

I might add that the nice new Nest will not cure the evident heat imbalance problem which you have with your system. It probably won't be much worse, but it won't be any better either. The warm basement and utility room suggests to me that perhaps your mains aren't insulated?

Hap_Hazzard

Well, the good thing about doing it that way is that you won't be there to hear what's happening, so you can believe it's graceful even if it wasn't. If you think it's worth the expense and the time it takes you to set it up, then great. It's good to have a hobby.

mattmia2

The longer cycle is in some ways masking the imbalance in the system. The current draw of your controls will also affect the anticipator setting. Those numbers are the current draw of the control circuit and you are supposed to set it to match. You can adjust a bit form there for longer or shorter cycles. You need to have good venting of the mains/returns, wherever it vents on that particular system so the mains all fill at about the same time and do so quickly. You can use the radiator valves to throttle down the rooms that heat first to balance the system further if it is 2 pipe.

SteamBoiler

Jamie Hall said:

Recovering from a setback will take almost exactly as much fuel whether you do it in very small steps or in one giant leap. One giant leap will, if anything, take less, as the boiler and piping and radiators won't have a chance to cool off and need to be reheated. But -- take your pick.

I wasn't thinking about fuel in terms of recovering from a setback. I was thinking about stressing the system less by ramping up setpoint temperature instead of a giant step (eg when set back by say 5F when on vacation). I hear a lot about setbacks - especially setbacks from smart thermostats - being bad for steam systems. It isn't clear to me why that is, could you please explain?

Jamie Hall said:

I might add that the nice new Nest will not cure the evident heat imbalance problem which you have with your system. It probably won't be much worse, but it won't be any better either. The warm basement and utility room suggests to me that perhaps your mains aren't insulated?

The mains are insulated. My assumption is that the old thermostat was causing frequent short cycles so much of the heat was spent in heating the basement and utility room.

It really does appear that there is an imbalance since some radiators get hot quickly and some get hot late. This afternoon there was a 15 minute heating call (normal 25) and the late radiators barely got warm. Since the system is already implemented, what can I do to balance? I could reduce the steam going to the hot radiators. One other idea is to service all the steam traps, maybe one is open and affecting the downstream radiators?

One additional clue is that the 2 slow radiators are the second ones in their respective rooms (upstairs bedrooms). Wouldn't this indicate this is not a mains venting issue?

Thanks for your inputs.

SteamBoiler

mattmia2 said:

The longer cycle is in some ways masking the imbalance in the system. The current draw of your controls will also affect the anticipator setting. Those numbers are the current draw of the control circuit and you are supposed to set it to match. You can adjust a bit form there for longer or shorter cycles. You need to have good venting of the mains/returns, wherever it vents on that particular system so the mains all fill at about the same time and do so quickly. You can use the radiator valves to throttle down the rooms that heat first to balance the system further if it is 2 pipe.

Thanks for your input, I am a little disillusioned with the mercury thermostat mainly because of the lack of data on heating call timing and duration and when it starts and stops, so will use Nest for the foreseeable future (as I said it appears to leave the room in a 67F-69F range and stops the heating early enough to let the room warm up to 69F).

I will definitely check the vents. I do have a 2 pipe so I will try to throttle down the rooms that heat first and get hottest.

mattmia2 said:

You need to have good venting of the mains/returns, wherever it vents on that particular system so the mains all fill at about the same time and do so quickly.

One additional clue is that the 2 slow radiators are the second ones in their respective rooms (upstairs bedrooms). Wouldn't that indicate that those branches are all filling quickly? Unfortunately I don't have any way to trace the branching.

mattmia2

if there are bad steam traps, steam could be getting in to the returns and preventing those radiators from venting. it could just be that they are further along the main.

in theory if you can keep the boiler pressure low you can meter the steam in to the radiator to just match the amount it condenses with an orifice plate or the valve so the steam never actually reaches the return and you don't need radiator traps.

there are numerous ways the main could vent in a 2 pipe system. they could have their own vents or there could be steam traps on the mains that connect to the returns and they vent through the returns.

the radiators that eventually heat could be heating only when the boiler produces relatively high pressure and compresses the air in the radiator enough to let steam in. producing that higher pressure wastes fuel and can damage steam traps if it is too high.

Jamie Hall

One thing which is really important to remember on two pipe systems is that the dry return must be vented, and not only vented but have bigger vents than the steam mains themselves. As @mattmia2 just noted, there is a possibility that a trap somewhere else on the dry return is leaking steam. There is also a possibility that that particular bit of dry return is not able to vent. You'd really have to trace the piping to make sure.

I agree that the Nest does give you (and Google) the ability to satisfy curiousity about how long the heat calls are, but it doesn't give you anything about how long the boiler actually runs if it cycles on pressure. I would find the 2 degree temperature swing from it annoying (the mercury T87s like the one you junked which are on all the places I care for have around a half degree swing, if that), but if that's OK... just be sure that all your passwords are robust and up to date.

SteamBoiler

Thanks for your response, much appreciated.

Jamie Hall said:

One thing which is really important to remember on two pipe systems is that the dry return must be vented, and not only vented but have bigger vents than the steam mains themselves. As @mattmia2 just noted, there is a possibility that a trap somewhere else on the dry return is leaking steam. There is also a possibility that that particular bit of dry return is not able to vent. You'd really have to trace the piping to make sure.

Unfortunately, it appears that there are no main vents at all on my setup. There is a return Trane direct trap and a return vent (green vent to right of trap in 2nd picture). Not sure how functional they are. I have started an attempt at rebalancing the system by closing 2 early hot radiators' supply vents 50% but today's a warm day so no calls for heat this afternoon.

Jamie Hall said:

I agree that the Nest does give you (and Google) the ability to satisfy curiousity about how long the heat calls are, but it doesn't give you anything about how long the boiler actually runs if it cycles on pressure.

Here is a screenshot of today's Nest datalog. The last heat call was for 23 minutes which is about normal for Nest at this time of year (as part of Nest setup and boiler pressuretrol counter-clockwise screwdriver tweak I verified physically previously once that the boiler wasn't cutting out). If the boiler were to cycle on pressure, the Nest heat call would be longer than normal since it would wait for the temperature to increase inside the home not knowing the boiler is cut out.

I agree that a mercury thermostat gives one the ability to fine tune a temperature range for a specific home with the anticipator (I said as much in an earlier post) but I do think a narrow range comes with more frequent and shorter heat calls, and no real indication of when a system is bust unless one starts collecting their own heat call logs. I wish Nest would give more control over temperature range instead of a +/-1 degree range and let the power user deal with tradeoffs of heat call duration vs call frequency.