control for intermittent input into radiant slab

Royboy

OK, here's one that I'm scratching my head on.



radiant slab heating system plus a wood stove for quicker response heating. cold climate.



slab to be heated with electric boiler and with solar combi-system doing DHW & some space heat (160 sf collectors - won't get real deep into space heat load but should contribute a fair bit during spring & fall).



homeowners prefer to go with off-peak electric rather than propane for heat. utility program offers half-price off-peak electricity 12 hours on weekdays (8 pm to 8 am) and full time weekends, and any electricity used daytime on weekdays is double-price. owners both work days so this jives pretty well with their schedules. the plan is to try to use electric heat for slab only during 8pm to 8am period, put any solar input in during day, use woodstove to bump up temperature more rapidly when owners come home from work.



in real cold weather, if needed, the off-peak control can be overridden manually and some of the high-price electricity used to even out the heat input.



homeowners and I are aware that there will be temperature fluctuations in this setup - more so as the outdoor temp drops, and they feel this is the way they want to go. unfortunately I really don't know how to model a heating system with these sorts of non-standard inputs, so its hard to predict how it will perform.



my main question at this point is how to best control the electric input to the floor when we are running the electric boiler only 12 hours a day. I understand that high-mass systems are best at stable target temps - and I have had no problems controlling them when heat source was always available. but with no electric heat for 12 hours, indoor temp will likely drop some, and then overshoot.



my hope is to figure out how to dampen that oscillation without getting too complicated. I wondered about using some sort of outdoor reset, where the supply temp from the boiler corresponds to outdoor temps. fine tune that curve from experience over time and I'd think that would help a lot.



thoughts on this one??? I know this design raises a number of issues - including the perspective that its not a good overall approach … at this point I hope to focus on ways this approach could be optimized, rather than going a whole other route, like an LP boiler.



Roy

STEVEusaPA

yikes.....

I feel for you Rob.  You obviously know the shortcomings to doing it this way.  We both know they're doing it the exact opposite of the way to provide the most even comfort.  Have you tried to figure out what it's going to cost them to run the electric boiler almost all night to get the slab back up to temperature, vs. the costs running it normally?  Even with modern boiler controls and a 2 degree setback,  the boiler is going to have to run alot to get the slab back up to temperature...then it'll be nice and over toasty at night when they're trying to sleep, then the house cools all day, they come home to a colder house---never enjoying the comfort they should get out of radiant floor heating.  Then on top of working all day, they want to come home, fire up the wood burner?  Seems like alot of work to save some money.

ok...solutions...I

Don't have many.  But was wondering, (and hopefully someone with more experience then me may chime in) what if you went with a large, or very large buffer/storage tank?  Get that water nice and hot before 8am, and do some injection mixing during the day? Maybe even the solar could dump there. Then most likely the only electric your using during the day is for the boiler control and the circs  With ODR and the right tekmar, you could take advantage of the fact that you dont need a minimum temp for the injection mixing with the electic boiler.

Like I said, I hope alot of people weigh in on this, and find you a solution...good luck

Jean-David Beyer

I have radiant in a slab for the downstairs zone of my house.

It used to be heated with an old GE oil burning boiler with a Beckett burner. The house is a little Cape Cod and pretty well insulated. It used to fail every few years when the high voltage transformer quit. I am in New Jersey where the design temperature is 14F. When the transformer conked out, I did not bother to call the service company except during business hours. If it quit on a Friday night, it did get colder, so I would wear a sweater in the house, and sometimes use two blankets instead of one. In two days, the temperature would drop 10F or 12F. Not really comfortable, but endurable.



I guess it mostly depends on how well the house is insulated, and the thermal mass mass of the rediant heating slab. With a slab, you can forget about bumping the heat up rapidly; you cannot get rapid temperature changes. If you have radiant heat in the walls or ceiling, perhaps you can get rapid temperature changes.

cattledog

control for intermittent input to radiant slab

Royboy--



Here are some comments based on my experience with a one zone, high mass, radiant slab plus a wood stove which is used daily in the winter. However, I am in a mild climate and am set up for pretty cool house temperatures when the stove is not running.



1) Sensor location. Control the slab off an in floor sensor. Find the right floor temperature for some base comfort level. Responding to room air temperature can give you wild swings with a slow responding system.



2) Temperature oscillations. With the electric boiler you don't really have to worry about short cycle times, and a pulse width modulated set point controller like the Tekmar 150 or a controller with 10 or 12 cph using the floor sensor could keep the overshoot in control. The low temp which the floor will get to when not heated is unknown, but like JDB says, the drop may not be that bad in 12 hours. I have had control success in my system using an on/off controller with a .5F differential. These guys are hard to find, and I use a Heat Link 46543, but they are not very common. Azeltek may also have something. I control the overshoot on long runs from cold by running the controller through a repeat cycle timer which clips the input during long runs.



3) Outdoor Reset. If you control from the slab temp and not the air temp, I think that for comfort,you are more likely to need to adjust the slab temperature in response to outdoor conditions than the water temperature which is heating the slab. Again, with electric there are no major efficiency gains with the lower water temperatures. Hotter and cooler circulating water temperatures may lead to faster and slower response times of the slab, but the degrees/hour response time as a function of water temperature is currently unknown, and the outdoor temperature may not be a big factor.



Regards

Royboy

thanks guys

Steve - yeah, plan is for a 120 gal tank as a solar preheat for the domestic hot water, and I plan to be able to pull space heat off that tank. solar will go to domestic water heater first and then to 120 gallon tank when water heater is satisfied. if I can figure out how to control things, I think I could electrically top up that tank at night, then draw it down into the heating system before the solar would get around to wanting to heat the tank. devils in the details on that one, and I don't have them worked through yet ... ;-)



cattledog - many thanks for your thoughts. I am not really up to speed on injection mixing and will need to change that to fully understand your concept here, I think. perhaps Tekmar might help me with that - or I'll give Siggy's book a try.



is the idea with the Tekmar that as the slab sensor approaches the target temp, it reduces the temp of the supply to the slab and thus minimizes overshoot? that makes sense conceptually, and would seem to be even better if the response curve can be customized to address this situation with extended "outages".



can outdoor reset then adjust the target slab temp? - it seems like that would be desirable if its a straightforward thing to do.



your other suggestion as I understand it is with a thermostat with a tight differential, like the HeatLink. I like the simplicity compared to injection mixing (as best I understand it) but my gut sense is it wouldn't do that well at avoiding overshoots.



am I am more or less understanding your ideas?

STEVEusaPA

responding....

one quick question, then more thoughts.

What is the homeowner's ultimate goal?  If you do find a solution for what you're proposing, what will be the overall cost/convienience (burning wood to help), etc of operating the system this way, as oppose to a LP mod/con with outdoor reset, combined with the solar.  Seems to me this would be the most cost effective.

 Was just wondering if this was completely figured out for the homeowner?

Some comments on the slab sensor.  The tekmar will control the temperature of the water (by injection) needed to satisfy the the t-stat (set to slab).  The reset curve selected will determine what water temperature is needed.  The tekmar t-stat will let you select min/max temps, and cycles.  This may help you fine tune your zones, and run times.  Also the boiler control has occ/unoccupied modes.  As far as the total control package tying it into the solar, I'm sure someone can help with suggestions.  Maybe start with a drawing (wallies love drawings), and hopefully suggestions/help will arrive

Royboy

good question

"homeowner's ultimate goal" - don't know that there's a single bottom line. comfort. low-cost. eco-friendly. easy. the usual mixed bag.



I just laid out all my concerns to them with this electric approach and we're going to meet to reassess in the next day or 2. will see if they want to continue on towards an electric boiler with some known issues (of unpredictable magnitude!) due to the off-peak interruptions & resulting complex control mechanisms. or flip over to an LP unit which they started out preferring not to use.



one rub is that I pretty much steered them down this path based on my personal experiences with our other local electric utility - where their "off-peak" program is interrupted for only 5 hours at a time and slabs seems to be able to coast through that no problem. didn't have experience with this other utility's program but assumed (wrongly as it turns out) that it would work fine with slabs, too. big oops.



but at least I'm being straight with them about the issues I have come to see & they are good clients and appreciate that.



I will draw this out at some point. solar tank is a 2-tank system, with space heat being drawn off the first larger domestic preheat tank, and solar input being directed to 2nd tank as a priority and when that's satisfied, back to the preheat tank which can contribute to the space heat side. I'm pretty comfortable with that setup and controlling it - have done it on other installations & it works fine.



well, I'm an advocate for simplicity when possible, but got myself into some complications here and at least I see I'll benefit by learning some about injection-mixing, Tekmar controls, etc.



thanks for your time & comments, Steve. I'm not done with this one by any means, but it sure is nice to be able to come here and know that lots of people are kind enough to have my back ...



Roy