Why no "indoor reset"?

JeremyN

My sister is going through a big problem with a new install of a modcon hot water boiler. You may have seen the thread in this forum - It's getting pretty huge...



Anyways, I don't know much about hot water heating systems, but her problems have got me thinking and trying to understand things. The one question I keep having is why do we use outdoor resets to modulate the boiler temperature? From my understanding, outdoor temperature is used an an approximation for what the heat loss from the house will be. Colder outside - the boiler increases water temperature. When it warms up, the temperature goes down in order to maximize the heat absorbed into the boiler water due to higher delta T values.



While outdoor temperature may be the largest factor in heat loss, it's certainly not the only one. Wind, sun, rain, snow and changes to the house itself all affect this number. Even if the other factors are inconsequential, it would seem that someone would still need to tune the system to get the best efficiency while still being able to heat the house on the coldest days.



So why not instead just ignore all of this and look at the one thing that truly matters - the difference in indoor temperature and requested temperature. Surely a thermostat that could transmit this difference to a controller must exist. From there a PID controller knowing the dynamics of the system (boiler, piping, radiation, etc...) could decided on a control strategy that would optimize for efficiency much better than just an outdoor reset would (effectively a P-controller, although the gain is based of outdoor not indoor temp). I'd think with this strategy, it would be possible for a boiler to be firing on a low setting for almost the entire heating season given that it was sized appropriately. I'd imagine it would still need to cycle during the warmer days due to the low end of the modulation limit.



So, with all this said, I understand that there's a lot I don't understand. Are there any control engineers or otherwise seasoned pros lurking out there who might be able to answer this? Is it just a problem with system stability due to the huge time delays in heating systems? Do I have a misunderstanding somewhere about how modcons, hot-water and outdoor resets work? Does this actually exist and I just haven't heard about it?



Looking forward to an interesting discussion!

SWEI

Outdoor reset

does a far better job of anticipating demand.  When properly configured on a well balanced system, we often see variances of only a degree or so across an entire heating season.  Think of it this way:  On a sunny winter day, when the sun goes behind the hill, the outdoor temperature drops like 10 degrees in ten minutes.  The indoor temperature has not budged in that time.  By increasing the water temperature, we offset the loss before you even notice it indoors.  After 15 years of playing with it, I still get a kick out of seeing people's reactions when it works.



It reaqlly works best when paired with indoor reset (aka indoor feedback) which can boost or buck the curve and (in the best systems) learn from that and adjust the ODR curve.  Viessmann and Rinnai offer indoor sensor/terminals for their boilers -- there are others in Europe, but I'm not aware of much else here in the US.

Rich_49

So A.M

is your sister I presume . Very nice lady , terrible shame what happened within her home .  There are many different methods of control of heating systems , problem is until you can get guys who can design , install or replace as in your sisters' case a simple heating system so that it works we have a huge problem .  I would confidently say that everyone that has contributed toward figuring out the mountain of missteps involved with your sister's project would have quite easily performed this job flawlessly and this discussion and that behemoth would never have taken place .  I believe that we have now identified all that is wrong and will now be at the mercy of the original installer to fix it and program the boiler properly . Not very hopeful .  I have local guys in the area that are awaiting her call and are already aware of the issues if needed .

   Here is a link directly to the utmost authority on the answers you seek . Hold on tight now because this is deeper than you thoughthttp://www.healthyheating.com/Programmable-thermostats.htm#! .



 The men and the technology are available . Now if you could explain to me how to convince someone to spend 50,000.00 for this type of system when your sister's contractor tells her he will keep her warm for 22,000.00 I'd be more than happy to listen . Because the promise of a 50.00 gas bill as opposed to a 140.00 one , and promising comfort and fresh air does not seem to make sense for everyone looking for return on investment . There are forces here that are all powerful whose pockets would be hurt by this type of stuff that administer what technologies get approved for use under all matter of Rebate , loan , performance programs .  They are a monster Octupi with massive tentacles everywhere . What you are searching for is not difficult yet in the same breath is very difficult .  Imagine a system that operated at thermal equilibrium , I have , can't get anyone to shell out the dough . So , I have focused my efforts on designing and installing the very best affordable systems that are practical for the everyday consumer , sometimes that is even a hard sell . 

  All in all I think we have done quite a job diagnosing what went wrong at A.M's house from the distances we all are from Massachusetts .  

SWEI

Smarter Controls

are routinely implemented on higher end commercial and industrial jobs.  We can look at multiple outdoor air temps (east, west, north sides of building), wind speed and direction, dewpoint, and temperature (with historical data and trends) and more to better predict demand.  Temperature sensing in each room is done by something that looks like a thermostat but is really a programmable user interface and IO device.  Occupancy sensors, daylight sensors, all kinds of fun stuff.



I'm in the process of commissioning my first residential job using some of this.  It's fun, but challenging.  Cost of the hardware less than a Tekmar system.  My time for programming should be under an hour on the next job.  Install labor is no different than any other system.  Still working on a good way to support it remotely when installed by another contractor in another state.  This is not an off-the-shelf product -- yet.

knotgrumpy

:-)

"This is not an off-the-shelf product -- yet."

Go get 'em, SWEI!  We knew you when...

Rich_49

The Wall

has a good smattering of pretty good guys that I guess could commission . I have been entertaining that idea myself . I design and ship packages a lot and was thinking of reaching out to certain individuals whom may be interested in a paid gig like that .  I think I may be insulted . LMAO.

Harvey Ramer

I'm interested

SWEI. Kudo's to you for taking the initiative!! I hope it works great and I hope you tell us all about it.

JeremyN

Cool!

I'd be curious to hear how this works.  Still without the indoor feedback it still seems like calculating the actual heat lost from a building would be an incredibly challenging task and would likely require an iterative process.  And on top of that, I would think it would be hard to prove that you have an optimal setup.  How do you determine when to stop tuning the reset curve? 



I still like the idea of a system totally driven by indoor temperature.  Theoretically a boiler should never need to turn off at a time when the heat loss is greater than the low modulation limit of the boiler.  My thinking is that a boiler that never shuts down is operating at the highest system efficiency possible for the given system as the average return water temperatures would be the lowest.  Am I correct on that?

JeremyN

yep

That's her.  You guys have all been super helpful over there.  I don't know what situation they would be in right now without all the help.  I encouraged her to post due to the help I got a couple years back on the steam wall.  My thanks to everyone on this site!



It sounds like setback is not an appropriate strategy with a high efficiency system, but that seems strange to me.  We're probably a bit unusual, but in our house, we have the thermostat set back for the majority of the day.  It gets set to 55 when we go to bed and only gets turned up when we get home from work the next day.  It's only turned up for 4 or 5 hours a day.  Granted, we have a steam system with a boiler almost twice the size of what my sister has, but I can't imagine - even with the most efficient system - that keeping the heat up constantly would be a better strategy for our case.  I'm wondering if high BOILER efficiency doesn't lead to higher usage in some cases just due to the necessary changes in behavior. 



Tying this back to my original post, an indoor reset would allow for a strategy like mine by allowing the system to recover from a setback quickly.  Albeit at a lower thermal efficiency.

SWEI

Actual heat loss

is a constantly changing thing, but (assuming the indoor setpoint doesn't change) it tracks the outdoor temp reasonably well.  We stop adjusting the curve when the indoor temp is comfortable across a wide range of outdoor temps.



Relying solely on indoor air temps is never going to work as well as knowing the outdoor temp due to the delayed response.  I liken it to driving a car down a winding road:  A single stage boiler or furnace controlled by a thermostat is like driving when you can barely see past the end of the hood, with the throttle bolted to the floor, using the clutch to control speed -- it's nearly impossible to keep the car on the road.  A modulating boiler with outdoor reset gives you an accelerator pedal and the ability to "see down the road."

Jean-David Beyer

calculating the actual heat lost from a building

I am a homeowner, not a heating professional. I did calculate my heat loss because my former heating contractor did not. He paced two sides of my house and came up with a figure that was much more than the existing boiler could provide, even though the existing boiler ALWAYS heated the house enough, and did very rapid cycling so must have been oversized.



I used the Slant/Fin program for doing heat loss and came out with some loss at 0F outside and at some higher temperature (perhaps 60F). I did not know my design temperature at the time. Actually it is 14F.



My mod-con is very easy to adjust the reset curves. I have one reset curve for a radiant slab zone and a different (hotter) one for my baseboard zone. I thought I could calculate the curve I needed from this, but it was not really satisfactory, though it was better than the factory defaults. Basically, the reset temperatures I calculated were way too warm. I guess my heat loss was quite a bit under my calculations. I do not know if that heat loss program has too big a margin of safety, or if my input data are not accurate enough (probably some of each). My house is an existing one, and there are a lot of unknowns. For example, I do not know if the radiant slab at grade is insulated or not (I expect it is not). I do not know the details of the wall construction. I do know the R values of the Marvin fancy windows. I have some idea how good the insulation is. I knew there was not enough baseboard in the baseboard zone, so I have now almost 5x as much as there was originally.



I then spent about 2 years trial and error. I could have done it in about a week if I could control the outdoor temperatures, but not being God, I had little influence over those. I am glad I did it myself, because If I called my heating contractor out to do it, it would have taken him only a few minutes each time, but I would have had to pay his minimum 1-hour fee. And I do not quibble about the fee. I understand that there are more costs involved than just the few minutes to push a couple of buttons. And the guys from that company do not understand outdoor reset at all. They say it does not make much difference and they did not even want to connect it up when they installed the boiler even though the sensor comes with the boiler. I insisted they connect it.



Now the reset curves are so close that my thermostats, especially for the radiant zone, are so close that they call for heat for extended intervals (18 hours is the record, though I get 12 hours sometimes. And the house holds the temperature set by the thermostat. It is a conventional on-off thermostat, not a proportional control. I experimented with setback in the baseboard zone and gave it up. The reset curve is so tight that if I use 2F setback, it takes over 4 hours to recover from it. I then enabled a "boost" feature so that if it has not recovered in 2 hours, it will boost the supply by 10F. That helped some, but not enough. So I do not bother with setback up there. And is worse than useless doing setback on a radiant slab zone.

JeremyN

Point Taken

Outdoor temperature has useful information to offer the system.  It allows it to look into the future to see what upcoming needs will be.



But to keep with your car analogy, outdoor reset is like a cruise control system that looks at external information in order to set throttle position.  For example, it might detect that you're entering a hill and adjust the throttle by some predetermined factor in order to compensate.  A more advanced version might take into account rolling resistance or wind speed.  In reality, this would be a very poorly performing system as the factors that affect total drag are difficult to quantify exactly and the car would likely either go overspeed or not be able to attain the desired speed.  To combat this, the designers would have to take the same approach we take with heating systems - Use too much throttle in all situations, and then simply cut it completely when we hit the desired speed.  Its a better system than just pegging the throttle and cutting it at the desired speed, but it isn't optimal.



In actuality, cruise control doesn't look at any of that.  It just looks at the speed of the car vs, the speed desired.  Even with varying road conditions, you don't notice the car's speed changing from the set point more than a fraction of a MPH.  This is analogous to an indoor reset.



Now of course this analogy breaks down a bit because the car's speed is almost instantaneously affected by throttle position whereas a house's thermostat takes a good number of minutes to read any difference in boiler output but that can be overcome with some smart engineering. 



My anecdotal evidence:



I bought a simple temperature controller off amazon for under $20.  I use it to control the temperature of a chest freezer I use to ferment beer in.  It drives either the freezer's compressor or, at this time of year, a heating element.  Even without any modulation, I'm able to maintain temps within half of a degree in the face of changing external conditions (It's in the boiler room which probably has 30 degree temp swings depending on on/off).  Granted, a freezer and a house have different dynamics, but I'd imagine scale up wouldn't be an impossibility. 



So indoor reset with external temp fed to it in order to anticipate changing demand?

SWEI

Indoor control

using proportional valves or mixing pairs quite well with outdoor reset.  Think of the outdoor reset as the coarse adjustment and the proportional valves as the fine adjustment.  By far the easiest way to accomplish this is with TRVs.  If the boiler water is a constant 150ºF, the valve has to hunt across its entire range in order to control the room temperature.  Most of the season, the valve is going to be working between perhaps 80.1% and 80.2% closed in order to hold temperature.



If the water temp is preset by the ODR control, the TRV has better authority, perhaps modulating between 45% and 55% closed to accomplish the same thing.



When setting up a system like this, we crank the TRVs up as high as they can go, then set the ODR controls so the indoor temp is perhaps 4-5ºF higher than what the owners find comfortable (or as hot as they might want it when an elderly relative is visiting.)  Between the ODR curve and the balancing valves, adjust the system until room temps are even.  Then turn the TRVs back down to a comfortable level.