Predicting Future Weather

I have a pretty sophisticated heating system, just like many others on here. It is a radiant floor heating system that uses a PLC to calculate desired temperature in the house, what the temperature is outside, what the in-concrete floor temperature is, and calculates once per minute the best way to achieve that temperature at the lowest cost based on degree day and temperature curves.

That is all well and good, and I am happy with the system, but I am thinking there is one slight problem and there might be a better way to use that problem to save even more money...

The issue is the delay in getting my concrete slab to heat up or cool off. The problem only presents itself when a warm or cold front moves in quickly. Because it takes 12 hours or so for my concrete floor to make a noticeable difference in the house, I was thinking what if we used that delay to conserve energy and create better comfort within the home?

I was thinking, if my heating system's PLC was tied to a weather service that predicted what the weather was going to be like in the next few hours and days, my heating system could make better choices on how it operates. Currently it only knows that it is X amount of degrees in the house, and x amount of degrees outside. The problem arises when the temp outside suddenly drops leaving the boiler cranking to catch up, BUT it takes a few hours for my concrete slab to transfer that heat to the room, leaving my house cooler then it should. At the same time the reverse is true; if a warm front approaches, my house is warmer than it should be. But think of the implications if my boiler could predict what the outside temperature was going to be; it could adjust itself to compensate and average out that temperature over a longer period of time. With an approaching cold front, it would leave my house more comfortable by being warmer, and with a warm front, it would save me fuel costs.

The beauty is: we already have this technology. My heating system is 6 years old, yet it can do this very thing. And we have the ability to provide weather forecasting in digital language. The only real issue is getting a digital weather forecasting service to deliver this information over the internet so my heating systems plc can interpret it and compensate.

I think by leveling out the spikes in my homes heating needs, and utilizing the thermal mass located in my concrete slab, some significant savings can be achieved.

What do you guys and gals think?

This old farmer thinks he'll just stick with looking out the window & pay attention to what these old bones tell me. As far as accuracy, be it digital or not, what do the numbers look like concerning our meteorlogical services? I get Accu-weather sent to my computer on a daily basis & I'd say accuracy hoovers around 50%, plus no matter what temps or conditions they call for I have to adjust accordingly because their #'s come out of the closest airport or elsewhere--not from here on the hill.

Great idea, but i think if you are going to go to that length it should have an adjustable curve for wind speeds similar to the heat curve for outside versus inside temps. Incredibly complex but that is what computers are for.

Kevin

That's one of the drawbacks to radiant heat in a concrete slab. Don't have any good suggestions for you though.

circulate in the slab at cooler liquid temps but stay in relation to heating degree days - just lower the bell curve- run a closer differential for on off cycles
or
circulate the loops more and longer but with less energy [cooler water] to cut the tops and bottoms off the over runs

Fred brings up a good point, you would need a reliable source of LOCAL data to make the system work. Kevin's mention of wind speed [and direction] is also an important variable if you real want such a system to work properly.

My $0.02 - this might one of those cases that more complex is not necessarily better. Have you considered just holding the floor at a comfortable temperature and circulating more often? Some people hold the slab at a stable temperature and install baseboard radiators with a two-stage thermostat. The baseboards react quickly, and prevent a big dip in room temperature while you wait for the slab to heat.

I am doing what Sting suggests because of inadequate piping. 24/7 circulation with temp dictated by tekmar control, injection pump setup. I would like the wind control as it affects the building more NOW then the outside temp does in the short term. Long term, it gets dark and the temp drops, in Maine at least. But the damn wind keeps howling. Overall, I am very happy with what the tekmar did for me, considering the stupidity in my radiant floor.

Kevin

I am having a difficult time explaining this...but let me try again, but realizing that I blame none of you for not quite understanding what I am saying. It is complicated.

My system circulates water 24/7 now, and the water temperature is adjusted every minute depending on what is going on outside. If the temperature outside drops, the water circulating goes up, yet if the temp outside rises, then the water circulating in my floor drops. It is all based on the degree days and how much heat my concrete is slab is losing. Most of the time it is not a problem, it is only when you get cold fronts or warm fronts that cause a problem.

lets say that today is Monday and it will be 30 during the day, and 20 at night, but tomorrow, Tuesday a warm front is moving in and it will start out at 20 but increase to 45 during the night and remain that way for most of the day on Wednesday, then by Thursday it will go back to normal temperatures by being 20 at night and 30 during the day. So in simple terms, during Tuesday-Wednesday it is going to get warmer...

Right now my heating system is pretty dumb, it only knows what is going on RIGHT NOW. It has no idea that in a few days it is going to get warmer. If it did know, then it could tell itself..."don't keep the floor so warm on Tuesday night" and let the slab lose some heat and by not running, save me some money.

Now if it was going to be really cold on Tuesday-Wednesday, the system would know that, and could compensate by increasing the slab before hand so that the highs and lows would be flattened out. That would work if creature comfort was important. What I would probably do however, is have the cold fronts programed out so that only the warm fronts would be adjusted for. Yes I would be a bit chilled until the system gets the floor warmed properly, but it would save me money.

On a single home basis, the savings would not be extensive, but if you consider as a whole what my county, state and the entire Northeast would save collectively by having predictive heating systems, the savings would actually be substantial.

I think this would actually work. As I said, the technology to do this is here already. It really does not matter if the weather service was completely right or wrong, as long as they provided my heating system with the knowledge that a significant warming trend was about to take place and for how long, I would save money.

I like that.

(And yes Klook, to be wind compensating would even be better).

Commercial buildings with central heating and cooling systems have adaptive weather prediction control systems. Snoop around the commercial control web sites of companies like Johnson Controls or Honeywell. Learn how they do it. My county school system contracts with Johnson Controls to provide such a system. It's a service they provide to many customers. I think the control system contacts a central weather prediction computer via a Internet connection. The system knows the school holiday schedules and adjusts accordingly.

I understand
now you must understand - sometimes the answer is not the one you want

http://www.degreedays.net/

Calculations using HDD have several problems. Heat requirements are not linear with temperature, and heavily insulated buildings have a lower "balance point". The amount of heating and cooling required depends on several factors besides outdoor temperature: How well insulated a particular building is, the amount of solar radiation reaching the interior of a house, the number of electrical appliances running (e.g. computers raise their surrounding temperature) the amount of wind outside, and what temperature the occupants find comfortable. Another important factor is the amount of relative humidity indoors; this is important in determining how comfortable an individual will be. Other variables such as precipitation, cloud cover, heat index, building albedo, and snow cover can also alter a building's thermal response.

an outdoor temperature reset control [ such as described above] can be programed to ramp up or down in consideration of the load - but nothing save your human intervention and intuition will do what you ask

But there is you answer - you can shackle yourself to the control panel in your boiler room and become the ECM of the system --- hummmm - I think there were men employed in servitude for this task over the last 300 years or more - maybe this isn't a new question of an old problem - just one I couldn't possibly throw enough money at for a proper solution

Hey, an oil furnace with a thermo would solve all that--Set the thermo & in accordance with house temp. --the furnace would kick on & off as needed. There ya go, problem solved---OHHHHH, now comes the cost part--hmmmmm, depressing. BUT, if the world were to take on this idea--what would the cost be per unit ???? Come on sting, we could split one cost wise---wait--that wouldn't work

Thanks Yanche, that was exactly what I was thinking of and looking for, but yes it makes sense that it would be on a bigger institution that naturally would save more money then a residential home.

Our local school is pretty cool; they knew oil would never be cost effective so they installed a biomass boiler. It takes about 20 cords per week of "clean chips" and makes hot water. They system is so efficient that they have pex under the sidewalks of the school to melt the snow and ice off instead of salt...the boiler is that large. All in all they said the fuel cost is half of that of the older school which burned oil. Now electrical costs; that is a different story. Because every class room is "digitally plugged in", the electrical consumption is twice the old school. They were going to install a windmill (because this school is atop of a huge hill), but the cost was too much and they were already over-budget.

Sting: Thanks for the explanation. I love my radiant floor system and all that it entails but I am really not looking for it to be perfect, in fact far from it. Right now it is like it is on cruise control...it is set to within a degree of what you set the thermostat for, but like with cruise control in a car, that is only good on flat, straight sections of road like the interstate. I live in Maine, so just like our roads, our temperature fluctuates and there is no need for me to maintain a perfect 55 mph knowing a half mile ahead there is a nice long hill I can coast down and save a little fuel. It is the same thing here. If my heating system knows there is a hill to coast down (warming trend ahead) I can tolerate going a little slower (being a few degrees cooler) to sip less heating fuel costs.

The one place I could trim a little fuel, but don't is on the lower extremes of things. My system has a default mode that shuts down the minute by minute calculations once the outside temp hits -10 degrees below zero (f). At that point, it just pumps 100 degree water through my floor since it is so cold outside. I could program the plc to default at -20 degrees or below and save a wee bit of fuel by doing the minute by minute calculations, but I am not sure I would save that much fuel at temps that low.

I'll research the bigger systems, send some emails and wait until the service reaches the residential market. It will soon for sure.

I submit = something is amiss with your outdoor reset control set

is the outdoor temp probe on the north side of the dwelling? Out of the direct sun? exposed to the wind chill? Is the wire broken that runs from the probe to the ecm?

Did you install the system? Is the radiation matched to the load? Is 100 degree water the design temp for the radiation to carry the load - or did the last guy just pick that number in the lottery?

someplace in these archives Yanche has placed a link to several great heat loss calculators that predict the correct circulation temperatures by system design - that may help

but

It Depends

nosmoke...as a tinkerer & programmer i can definitely see the fun side of this project as well as the theoretical benefit if you can make it work. one issue that comes to mind is you will have to have some sort of offset from the general area weather forecast for your local micro-climate.

as mentioned outdoor wind speed plus temperature plus indoor humidity would probably help quite a bit in adjusting the comfort level. also the direction the upcoming weather is traveling from would impact your local micro-climate as well.

seems to me you will be tweaking the prediction models quite a bit to dial it in as you learn why one particular forecast or another didn't have the desired effect. depending on your perspective...continued fun or a PITA....

Sting wrote:is the outdoor temp probe on the north side of the dwelling?

Yes

Sting wrote:Out of the direct sun?

Yes

Sting wrote:exposed to the wind chill?

Yes

Sting wrote:Is the wire broken that runs from the probe to the ecm?

No

Sting wrote:Did you install the system?

Yes, but my Brother-In-Law helped and is the instructor for the Maine Oil Dealers Association and knows radiant floor heating systems well

Sting wrote:Is the radiation matched to the load?

Yes, every pex loop is controlled by a flow control valve and is dialed in to get a 15 degree drop in temperature from entering my concrete slab to exiting; the number he told me what I needed. No loop is greater than 200 feet in length. The circulators are also three speed circulators for greater control.

Sting wrote:Is 100 degree water the design temp for the radiation to carry the load - or did the last guy just pick that number in the lottery?

He told me 100 degree water was the default temperature for -10 degree weather or below.

I talked with another radiant floor expert this summer and he said my metering system is not needed, and that I can actually program my Munchkin Boiler to do everything that metering pump does (but he did not call it a metering pump, he had some other name for it. Basically it is the circulator that takes the water in my main boiler loop, and gets it down to the target temperature before going into my concrete floor loops). But he said when I get my coal/wood boiler installed, I will need it to drop down my high temp water so I should keep it for now.

The only other thing about my house that is different is that under my concrete floor, I have 400 tons of rock for added thermal mass isolated from ground temperatures by insulation.