Air Getting into New Radiant System After Cool Down

danwheeler

Well, if nothing else, I'm sure learning a lot this week. It looks like the Honeywell thermostat I'm using to test the system has a zero degree "deadband" setting... they don't allow you to change that.

BUT, they do have different "CPH" (Cycles Per Hour) settings for the different heating types. Pretty sure I'm still set to electric forced air instead of hot water. Forced air can cycle 9 times in an hour while hot water is limited to 3. So that may be what's going on. I'm going to call a neighbor to have him set it to boiler.

it's an interesting way they've decided to work their thermostats. Not sure I like that.

mattmia2

It looks like it is on for maybe 2 minutes then off for 10 minutes or so? If you reduce the supply water temp it won't heat as fast and will run longer cycles. The first step is to do a heat loss calculation with the slant fin calculator or some other calculator. That will tell you how much heat you need. from there you can figure out what supply temp you need for that output. Ideally you would control the water temp on outdoor reset and maybe a timer for recovery.

EBEBRATT-Ed

Does that stat have a cycles/hour setting?

danwheeler

I had a neighbor go over and change the thermostat from forced air electric to boiler. This changes the CPH (Cycles Per Hour) from 9 to 3. I think this is going to fix the short cycling.

It's interesting that Honeywell approaches the same problem from a different angle. They choose to keep the air temp consistent and adjust the number of cycles per hour while other manufacturers will allow the temp to "swing". I suppose from a human comfort standpoint, it makes more sense to try to keep a consistent temperature. I think Honeywell's style will work best for me because I do notice 1-2 degree temperature differences. I'm comfortable at 74. Not 73 and not 75 hahaha.

Crazy how much there is to learn and know about this stuff and that I can spend a few hours doing research reading forums to find past discussions of people with the same issue, what caused it and how to fix it.

We've come a long ways from making a fire in a teepee.

I checked into outdoor reset for my Electro Mini EMB-S-9 but it doesn't seem to be an option for that model. Can you add outdoor reset to any boiler?

Jamie Hall

The thermostat really shouldn't cycle that quickly. Without playing with the thermostat in question, it would be hard to say what the problem is, but something just isn't right there.

hot_rod

The manual for the thermostat might shed some light on the cycling? I have an anti cycling function built into my boiler that prevents frequent on off cycling, it can be adjusted, I have it set for 10 minutes. It is useful mainly when the room stat is just about satisfied.

Maybe that thermostat has such a function?

You might check with Electro and see if you can upgrade the control board to get the modulation. Usually it just involves some triac relays running the element. Outdoor reset may be available also to help better avoid those on off cycles.
So much can be done with microprocessor type controls. If a perfect heating control logic exists, the boiler would never shut down, just modulate to the load throughout the heating season. I suppose those elements could be modulated like my stove top, it goes from low to 9 is .5 steps.
Reducing cycling may not lower the electricity consumed but save on/ off wear on contactors and controls.

What are your electric rates? Looks like you have a good insulation package, the glass is you big loss. insulated shades can make a big difference if you haven't already. If you know the heat load, energy cost and degree days you could get an idea on what it will cost to maintain temperature.

It might be fun to invest in an infrared camera or attachment for your phone to take a "look" at heat loss areas. There is another thread here on the wall showing an IR look at a home.

danwheeler

@hot_rod and @Jamie Hall see my couple of posts above about CPH on the Honeywell thermostats. Honeywell does things a little different... no "swing", they use a zero degree deadband, they use Cycles Per Hour and attempt to keep a consistent temperature. I think I've got it sorted out now, I had it set to forced air electric which allows 9 cycles per hour. I had a neighbor go over and set it to boiler which limits it to 3 cycles per hour.

I'm not sure if it's fixed the problem because after he did that, it turned on and hasn't turned off since. It's been running full blast for 2-3 hours now and hasn't raised a degree. Not sure what that's all about. Maybe just typical lag and the outside temperature reducing as it gets later.

hot_rod

maybe I missed it in the thread, do you have a heat load number? I think the boiler you have is a 9KW, so 30,690 BTU/ hr. is what you have.

Is the home 700 square feet on two levels? 1400 square feet total? If so, you could put 22 BTU/ sq. ft into the space with that boiler.

If you haven't done a load calc, start here.
https://www.slantfin.com/slantfin-heat-loss-calculator/

danwheeler

I'm not sure I'm qualified to use that calculator I tried plugging in some numbers but I don't really know what I'm doing. A heat loss calculation was done by the supplier that shows a need for 22K BTUs but with an acknowledgement that the windows may require more. I put a few numbers into the calc and got 33K BTUs but again, I don't think I was putting all the values in right.

The main floor with the slab is 440sq feet and the upstairs which is basically just a bedroom and a small bathroom is 305 sq feet so the whole thing is about 750 sq feet.

The Tstat finally just clicked off. Still rock steady at 70 where I want it, I was just a little surprised it didn't get up to 71 after running for several hours today.

I'm doing some research right now on Honeywell's algorithms for their smart thermostats. From the sounds of it, there's way more going on than just a simple on and off based on the set point. They may be doing anticipation and using outside temperature and all kinds of other smart stuff that is not disclosed information.

One thing I keep seeing that I don't understand is "heat load"... I see lots of things refer to 50% heat load but what is that and how would a thermostat have any idea what the heat load is?

Jamie Hall

The problem with radiant heat is -- as you are noticing -- that it is very slow to respond. It's the nature of the beast.

I'm glad you got the cycles per hour thing fixed. I suspect that may have been almost all of your cycling problem, if not all of it. The better Honeywells do have a remarkably sophisticated little program running inside them which takes into account the difference between the set point and the actual temperature, and also the rate at which the actual temperature is changing, and uses that -- with considerable success -- to decide when to activate the heat or turn it off. The cycles per hour setting factors into that little program. It will, given time, get better at deciding when and how much to run -- but only if you leave it pretty much alone. At least in the ones I've examined -- by no means all of the models -- if the difference between set point and actual temperature is mor than a few degrees it just simply turns the heat on and runs until the set point is reached, ignoring the rate of change. Which, of course, often results in an overshoot on recovering from a setback.

Which you shouldn't be doing to radiant heat anyway...

hot_rod

Is the glass wall facing south by chance? That would help with some passive solar, depending on the type of glass.

The heat load calc by the supplies should have the glass in the calculation ? Or was he unsure of the window type? The calculators figure the r value of all the exterior walls and windows, roof and exposed floor.

So a simple calculation, 22,000 btu/hr required/ 750 square feet of radiant floors? = 29 btu/ sq foot required on design (coldest) day. If your only heat is the floor, that may be a tough heat on the coldest days.

Usually anything over 20- 24 btu/ sq ft requires some supplemental heat for those cold days, you just don't have enough floor surface (radiator) to cover high load days. Assuming the entire floor is exposed to be able to radiate heat into the space. Any built in, cabinets etc that go down onto the floor subtract out of that 750. Little if any heat will come out of floors that are covered.

danwheeler

Thanks, guys - @Jamie Hall yeah it did eventually overshoot by 2 degrees tonight after running for several hours at 70 with no increase. I just watched it do another short'ish cycle even after setting Cycles Per Hour to 3. But this short cycle was about 5 minutes which is longer than they were. In theory, at 3 CPH, it should only be able to run for 10 minutes, sit idle for 10 minutes then repeat that two more times in an hour.

I'm learning that these radiant systems are almost like the human digestive system - eat a high protein meal then burn it slowly and release the energy later

Not planning on doing any setbacks in fact, just a little bit ago I was shopping for T87s. If you're not doing setbacks, why bother with a fancy thermostat or even one that you can control remotely? I think they are really cool looking and dead simple. But as a vacation home, I think I will drop it by 3-5 degrees while I'm away, I just have to make sure I set it back to desired temp 6-12 hours before I arrive.

@hot_rod The big window wall faces East. I don't think I'd survive summer there if that window wall was facing South or West. I do have a mini split that I haven't installed yet which will heat and cool AND a wood stove but we were only able to bring 125 amps into the building (hidden inside one of those columns) so with no gas available in the area, everything is electric which means my electrical panel is VERY tight and probably oversubscribed. 60 amps for the boiler, 20 for the mini split, 30 for the DHW, 30 for the dryer, another 60 or so for cooktop and oven.

PC7060

@danwheeler - I used a delay on break timer like this one to address a similar short period on-off-on behavior (aka short cycling) with my Honeywell 9000 series Thermostat. 

I have mine set to hold the boiler on for 17 minutes after the thermostat call for heat drop.  Works very well and there won’t be any risk of temp swings with a high mass system like yours. 

https://www.digikey.com/en/products/detail/littelfuse-inc/TDUB3000A/7931401?s=N4IgTCBcDaICoBECqAhAzABiwQRAXQF8g

danwheeler

PC7060 said:

I used a delay on break timer like this one to address a similar short period on-off behavior with my Honeywell 9000 series

Thanks, so does that mean you run your circulator pump(s) continuous?

Did you check out the CPH setting on your 9000?

mattmia2

Note that "deadband" is the minimum difference that you can set between the heating and cooling setpoint on a heat/cool thermostat.

For example if the deadband is 3 and you set the heat setpoint to 72, the lowest you could set the cooling setpoint to would be 75.

danwheeler

mattmia2 said:

Note that "deadband" is the minimum difference that you can set between the heating and cooling setpoint on a heat/cool thermostat.

For example if the deadband is 3 and you set the heat setpoint to 72, the lowest you could set the cooling setpoint to would be 75.

From what I've read, that depends on who you ask. Seems that is a contentious topic. If you ask Honeywell, "deadband" is NOT what you just described. To Honeywell, deadband is what some call "swing"

To Honeywell, temperature differential is what you described.

And really, the word "deadband" is better suited for a range that has no effect on thermostat operation. "Deadband", meaning a range (band) in which the thermostat does not react (is dead). Same thing as swing.

IMHO, as a layman, Honeywell's use of the term makes more sense than using it to desctibe the required difference in set point between opposite modes of operation.

danwheeler

Confirmed my short cycling problem is resolved now. Cycles are running longer and less frequent. I can see the cycles getting longer as it gets colder outside tonight. Meanwhile, temp is staying spot on at 70 with no swing. Well done, Honeywell.

PC7060

danwheeler said:

Thanks, so does that mean you run your circulator pump(s) continuous? Did you check out the CPH setting on your 9000?

The 9000 series does not have a setting for CPH. It relies on selection of the appropriate system under control. I have a very low head converted gravity system with 19 radiators.  The pump is set to low and moves between 4 to 8 gallons per minute when boiler is operating.  

Your experience with concrete construction mirrors my own. I have learned make sure I am on site to make sure the concrete slump and the rebar match the drawings

Jamie Hall

Sounds like maybe you're on your way. As I said, it takes the better Honeywell thermostats some time to "learn" the appropriate constants for some remarkably sophisticated math to optimize the system. Oddly, the better they are, the longer they take (they take more things into consideration to fine tune).

And Honeywell's use of the term "deadband" is correct. The same term is used in any control system to define the maximum variation in the controlling variable without affecting the controlled variable.

The cycles per hour setting is a little illusory -- it really doesn't set a specified number of cycles per hour. Rather, it affects the deadband in some systems and some of the math in more sophisticated systems. The newer practice of calling for system type is much less confusing -- and does the same thing.

EBEBRATT-Ed

Some of the terms used in controls has always been an issue. Back in the days when pneumatic controls were popular and there was Honeywell, Johnson Controls , Barber Coleman and Powers were the big 4

Some called it "differential" others called it "throttling range", "dead band" etc

To me "dead band" is the dead area between the heating and cooling set points so the heating and cooling equipment can't fight each other.

mattmia2

EBEBRATT-Ed said:

To me "dead band" is the dead area between the heating and cooling set points so the heating and cooling equipment can't fight each other.

I think that is how the 20 year old Honeywell Chronotherm I have uses it.

danwheeler

Oh man, that’s confusing well I guess as long as the person you’re talking to knows what you mean, it’s all good.

Just got to the cabin after being away for a day and letting it run, this thing seems like it is really dialed in now. I bumped it up from 70 to 74 this morning at 8am and it had no issues getting there with it 20F outside. No overshoot either.

I wish I could take everyone here out for a beer, I really appreciate all the help. It’s great to have gone through all these issues, now I know so much more about the system and will be able to fix it if it breaks in the future.

hot_rod

I wish I could take everyone here out for a beer, I really appreciate all the help. It’s great to have gone through all these issues, now I know so much more about the system and will be able to fix it if it breaks in the future.


That's what DoorDash is for. Plenty of dashers in my area, Rudolphs too, at this time of year.

danwheeler

Hi all, again,

I've been away from the house and new heating system for a week or so now. It's been cold, in the single digits at night and it is working and keeping up, although it's definitely working hard.

I've got a new rattle. I can't quite figure out where it's coming from but it's either the pressure relief valve, the air separator or the pressure gauge all in the same area (see pic below) The rattle increases in speed and frequency with the pump speed.

I thought it was the pressure gauge so I replaced it but the sound remains. I notice that any tapping of the pipes 2-3" away from the pressure gauge sounds like it's coming from the gauge when my ear is right up to it so it may just be acting like a stethoscope due to its cupped shape. So, I don't *think* it's the gauge because why would a gauge make noise anyway?

That leaves the pressure relief valve or the air separator. It sounds louder towards the pressure relief valve than the air separator.

I did read somewhere that the pressure relief valve (am I using the right name? It's the thing to the right of the pressure gauge) should be downstream from the pump. Mine is upstream from the pump. Could this be causing my rattle? Not sure why it just now decided to start rattling. I backed the screw off until it released some water then returned it to its position but that didn't help. Thought maybe there was an air bubble in there.

Next issue is that my boiler is making that "steam" sound or like a coffee maker and every once and a while it makes a tiny little squeak. When I get my ear up real close to the boiler, the coffee maker noise almost sounds like a welding arc noise.

Anyone have ideas on the rattling noise? And should I be concerned about it still making the coffee maker sound and the squeaks?

thanks,
Dan

hot_rod

The circulator pump may have a check valve in it. Being directly at the discharge they sometimes rattle. If it is a multi speed pump, try changing the speed setting to see if it stops or changes

Air bubbles can also cause them to chatter. Is the system quiet from the air noises now?

danwheeler

The pump is quiet, the rattle is definitely coming from the area circled in the picture.

I'm not hearing any air noises anymore. Nothing obvious anyway. It seems to be well purged unless the squeaky/steamy boiler sounds are indicative of air.

I had to add some water/pressure to the system. I've been watching it with a camera while I've been gone and when it shuts down and cools off, the pressure drops below 12 PSI. How far below, I'm not sure since the gauges that came with the system seem to bottom out at what shows as 8PSI on the gage. I'm not sure they are very accurate at the low end so I don't know how low it's going. Hopefully not negative pressure.

hot_rod

It’s normal to see pressures decrease at it cools, if it drops to zero….
might be nice to add a good accurate gauge somewhere. A 30 psi gauge allows you to see small changes

don123

Hi, does this type of boiler require a fixed water source as in a fast fill S1156F like any other type boiler? I may be way off. just learning.

hot_rod

A controversial question. A fill or feeder is used to maintain system pressure. If a system is leak free, at some point it does not need additional fluid added.

A tiny leak, for example would show as a pressure drop and need for more fluid

Always wise to add a low water cutoff on systems that do not have a feed system. It prevents component damage should the system leak down 

danwheeler

hot_rod said:

A controversial question. A fill or feeder is used to maintain system pressure. If a system is leak free, at some point it does not need additional fluid added.

A tiny leak, for example would show as a pressure drop and need for more fluid

Always wise to add a low water cutoff on systems that do not have a feed system. It prevents component damage should the system leak down 

Thanks, Bob - I was kind of wondering about this... it seems like a tight system shouldn't need a feeder but I don't want a leak or other blowout to keep pumping, especially if I'm gone.

What do most people do here? Should I run my thermostat W wire through a pressure switch so it doesn't call for heat if below, say 10 PSI?

hot_rod

A low water cutoff would be better. System pressure could drop without a leak being present, temperature related.

A low water cut off has a probe at a level above critical parts, it only protects when the fluid has left the room

danwheeler

hot_rod said:

A low water cutoff would be better. System pressure could drop without a leak being present, temperature related.

A low water cut off has a probe at a level above critical parts, it only protects when the fluid has left the room

got it. Can you recommend a good LWCO?

hot_rod

This was the brand and model I used for years. Simple to install and wire.

danwheeler

thanks, @hot_rod

From the diagrams i've seen and youtube videos I've watched, it seems like those get installed maybe 6-12" above the top of the boiler? So if you have a leak, you're probably going to have water everywhere before it kicks in but saves you from ruining your boiler?

On a radiant system like mine, does the same philosophy apply? Just install it somewhere above the boiler? Should this be inline with the stream of pumped water? on it's own branch? or doesn't matter?

just wondering if it makes any difference to have the sensor piece hanging down into the fluid flow.

thanks,
Dan

hot_rod

yes it protects the boiler and circulators from running without water. it could also send an alarm contact to a phone or devise to notify you.
i believe you want it installed horizontally so an air bubble doesn't trap at the probe. The installation manual should help with that info.