Why does temp swing 5-6 degrees from set point?

Ken_40

about the rest of the house since it will determine if the problem is house-wide, or merely one room.

You should be correct regarding getting the rad that impacts the 'stat being key to the ultimate solution, but with steam, many logical progressions are anything but...

And no. 5 degree swings are not "normal" for steam systems, but in a few instances, unavoidable; e.g., a rad is grossly oversized for the room it heats.

Charles_8

Is this a venting problem?

Single pipe steam system. Pressure is very low (doesn't move the needle on an 0-15 psi gauge) but house heats well except as described further. Thermostat is in the 14x14' living room, mounted on inside wall about 8' from the radiator, originally an old-fashioned mercury switch beehive. This radiator is very long (24 segments) but only about 18" high, ubetween the two front storm windows.

When getting up in the morning, and turning up the heat from (say) 60 to 68, it was necessary to "inch" it up just barely until the boiler kicked on, and even then the room would reach almost 70, long after the thermostat satisfied. Then I could set it to 68 and although the temp would "swing" several degrees, the overshoot was much smaller thereafter.

I just installed a new Honeywell RTH110B digital thermostat. It has a nonadjustable 2 degree hysteresis (on at 65, off at 67 for example). The room temp still overshoots 5 or 6 degrees on every cycle. Set to 65, boiler off at 67, continues to rise until 70, over time gradually fall to 65, boiler on, repeat...

Is this a venting problem (should the living room radiator be vented larger or smaller), or is it just unavoidable due to the large thermal mass of the radiators and pipes? I'm guessing that I'm stuck with it but would be interested in your experiences...

thanks
Charles

Ken_40

Always...

Put the smallest vent orifice in the room that affects the 'stat.

What about the rest of the house?

Do those areas also have 5-6 degree swings?

Charles_8

I'm not sure about the rest of the house. On another thread I had posted that my vent valves are all messed up (improper function and unknown or wrong sizes) and I've got some new ones on order.

I figure I can fix the rest of the house once the living room/t-stat control loop is working right

Is 5 degrees an expected swing for a home steam system or should I be able to get that down to 2-3 degrees which I would like much better?

-Charles

ttekushan_3

Ken, you said

> about the rest of the house since it will

> determine if the problem is house-wide, or merely

> one room.

>

> You should be correct regarding

> getting the rad that impacts the 'stat being key

> to the ultimate solution, but with steam, many

> logical progressions are anything but...

>

> And

> no. 5 degree swings are not "normal" for steam

> systems, but in a few instances, unavoidable;

> e.g., a rad is grossly oversized for the room it

> heats.



quite appropriately,

"but with steam, many logical progressions are anything but..."

Like redoing the venting and actually putting a larger vent on radiator closest to the tstat. If the radiators are all oversized but fairly balanced, the system will overshoot like that after a deep setback. The larger vent in the tstat room will shut the boiler down before the others have a chance to overheat.

Trial and error, tho'.

ttekushan_3

Ken, you said

quite appropriately,

"but with steam, many logical progressions are anything but..."

Like redoing the venting and actually putting a larger vent on radiator closest to the tstat. If the radiators are all oversized but fairly balanced, the system will overshoot like that after a deep setback. The larger vent in the tstat room will shut the boiler down before the others have a chance to overheat.

Trial and error, tho'. Sometimes counter-intuitive-ness rules.

-Terry

Charles_8

Sizing radiator?

What exactly is a "grossly oversized" radiator? Is there a table somewhere that shows the proper size?

This house is in far northern Maine where the temps in winter are frequently -20F overnight in February. But it's not always that cold - late fall and early spring, for example, might be a balmy +30F. It's normal to be still heating a little all through May, too. Does such a large temperature range make it difficult to size a radiator properly?

I've shut off the unused rooms. The bedroom and bathroom are comfortable (although I have not checked their temp swings yet). The living room, as noted, is small, about 14x14 feet, but has three large double-glazed outside windows plus french doors to the (now unheated) dining room I don't use.

The living room radiator is very short to stay beneath the tall windows. It's about six feet long (24 segments) but only about 15 in. high. I don't think it's ever been fully hot end to end... so maybe it is too big? Making it smaller by removing sections might be a real adventure since it's probably been together for 50 if not 100 years!

-Charles

ttekushan_3

Leave them alone

I'd work at it from the venting balancing angle. Look at the library at "its all in the venting" and pick up a copy of Gerry Gill's "Venting Steam Systems using a Venting Capacity Chart"

Given the region you're in the radiators are probably right. Proportional venting will get you close to your goal, but overshoot on such deep setbacks are unavoidable. But you should be able to get the overshoot to 2 degrees from deep setback. So set it 2 degrees lower.

Please don't change the system to conform to this week's or this year's lifestyle. You or the next owner will pay and pay and pay....

-Terry

Vermonter

How about the old thermostat?

Did it do that with the old thermostat?

Charles_8

Yes, in fact it was even worse with the old thermostat, which is one reason I changed it... The "beehive" had about 3 or 4 degrees hysteresis which added that much more to the problem, since the boiler would not cut off until the temp had risen that much above the setpoint.

The new one has 1.5 to 2 degrees built into it which has helped a bit. I'd like to find a thermostat that has even less, or adjustable, since the feedback loop is so heavily damped (by the mass of the radiators) that frequent cycling would not be an issue.

Altering the heat anticipator settings on the old thermostat should have helped, but I didn't want to make a large change from the recommended 0.3 amp setting for fear of burning out the internal heating element.

The new one doesn't have an explicit heat anticipator, but there is a setting for Heating cycles per hour: 2,3,4,5, or 6, with the longer ones supposed to be for steam and the shortest for electric heat. I've tried the shortest and the longest and it didn't make any perceptible difference (still overshooting 5 or 6 degrees).

-Charles

Vermonter

Maybe a different thermostat?

I can't find the manual for the thermostat you installed, do you have a link? Does the thermostat have any kind of anticipation or is it just on/off?

I've had really good luck with an Aube thermostat, model TH148LE-P. It was designed specifically with hydronic in mind. It has built in proportional interactive adaptive logic that has kept temperature swing in my house inside 1 degree of the setpoint. It also programs itself to come on early so that the house is at setpoint by the time programmed.

It did take about a week and a half for it to "learn" my house and it was a little uncomfortable during that time. Yours may be similar.

chapchap70_2

If you bought a new thermostat anyway, why not try raising the anticipator settings above the 0.3 amp settings on the old one that are recommended for most boilers. I think one boiler manufacturer recommends you set the anticipators for 0.5 amps to make their system work best. What is the worst thing that can happen? You already have another thermostat.

Charles_8

An excellent suggestion!

Thanks for the suggestion, I'll give it a try with the old one.

I'm also thinking about creating my own anticipator setup - a small resistor in series with the 24 volt loop mounted right beneath the temp sensor (which is just what a heat anticipator is). It would have to be the right type to cool down at about the same rate the radiator is still heating up when the thermostat is satisfied - probably a small metal-film or wirewound resistor. It takes a few minutes (3 or 4?) for the room temp to stop rising after the boiler cuts off.

The only problem I see with this setup is that large steps higher in set point would take longer to stabilize, since after a few minutes of heating time, the system will be cut off by the anticipator. But it'll start again in a couple more minutes and eliminate the large overshoot I have now. That's basic control theory, of course - underdamped loops overshoot, overdamped loops take longer to reach equilibrium, and so-called critical damping is the best compromise.

The link on Honeywell's site doesn't work either. I plan to contact them for more technical info than the brief installation leaflet provides. I'm not even sure how the "number of cycles" setting works? Does it deny further requests beyond that number per hour for heat, or what? Doesn't make sense to me so far.

-Charles

Vermonter

Do you have a good electric tester? If so then take the guesswork out of the anticipator setting and see how much current is coming through the thermostat. That's the "correct" anticipator setting. I'd start there before I did anything else.