I have a 12 year old radiant floor system that is heated with two 60 gallon water heaters. They have reached the end of their useful life and I am planning to replace them with an electric boiler. I have done the heat loss calculation and arrived at a 30,000 BTU/hr requirement. (roughly equivalent to the output of the current water heaters). My question is do I need to multiply the BTU/hr requirements by a factor if I am using a timer to take advantage of time-of-day rates and only running the boiler between 11PM and 7AM? The current system does seem to be doing the job, but I want to be sure I am sizing the replacement boiler correctly.

I guess some of this depends on what your floor temp is and what the storage tanks are charged to. IE: 1 BTU will heat 1 pound of water 1 degree F. if you have 120 gallons of storage and you charge it to 180F and you need say 90F for the floor that is 960# X a 90F drop or a storage capacity of around 86400 BTU or about 3 hours of heat at your calculated load...

In order to get you over the 16 hours between 7AM and 11PM, you are going to need a LOT more storage capacity(roughly 5X+?) or about 600 gallons to supply that load for 16 hours... Once you determine the ammount of storage you need, you would want to size the boiler to meet the heat load and fully charge the storage in the 8 hours of off-peak power you have available... For 16 hours you would need 480,000 BTU to charge the storage + the load for those 8 hours, 240,000 for a total of 720,000 BTU divided by the 8 hours to charge it, or a 90,000 BTU/HR boiler...

Sorry, I guess I wasn't too clear - I have a slab on grade system where the storage is mostly the slab itself - i.e. the tanks are pretty much a wash. At 11 PM when the heat comes on they are at 75 degrees, so the first hour or two is spent heating the water in the tanks to 120 at which point the circulation starts. Then at 7 AM when the power is shut off to the tanks, they are at 130 and the water circulates for an hour or two before it is expended and the water temp drops below 100 degrees. Essentially the slab stores the heat for the day.

Removing all of that discussion, my question boils down to: if my heat loss calculations says I need 30K Btus/ hr to heat the space, does that mean that if I used 0 degrees F as my low temperature design point, then on a day when the temperature remains steady at 0F would I need to run and deliver 30K BTUs for 24 hours to keep the space heated? And thus if I am only running the heat 12 hours a day, I would need to be able to deliver 60K BTUs/hr?

“If my heat loss calculations says I need 30K Btus/ hr to heat the space, does that mean that if I used 0 degrees F as my low temperature design point, then on a day when the temperature remains steady at 0F would I need to run and deliver 30K BTUs for 24 hours to keep the space heated?”

Yes, you will need to get the slab surface to the required temperature that will generate 30K Btu/hr and maintain this temperature for 24 hours in order to continuously generate and replace the 30K Btus/hr lost each hour by the building. This is equivalent to a heat loss and heat gain of 720K Btu/day.

“And thus if I am only running the heat 12 hours a day, I would need to be able to deliver 60K BTUs/hr?”

No, if you were to deliver 60K Btu/hr, you could overheat the building. 60K Btu/hr is equivalent to a heat gain of 1,440K Btu/day and your heat loss is only 720K Btu/day. If you only deliver the 60K Btu/hr for 12 hours, your daily heat gain and heat loss would be the same. However, the hourly heat gain would vary both above and below the required 30K Btus/hr during the day. How much variation you would actually see would largely depend on the thickness of the slab, how well insulated/sealed the building is, and the variation of the outdoor temperature. A thicker slab providing less variance and perhaps being acceptable to you. But why not just deliver 30K Btus/hr...or really, deliver what the thermostat actually calls for real time since the 30K Btus/hr is only the design heat loss and not the actual hourly heat loss?

30K sounds low. What program are you using? The answer to time depends on alot of factors only an accurate hygrothermal sim model can answer. You'll need to hire a pro.

No, a simulation isn’t at all needed...only a well designed HR system, a well insulated/sealed building and a good thermostat...

only a well designed HR system.

Exactly what a whole building simulation does vs in getting it wrong and associated cost or a redesign at that point. You need a model that brings in local TMY3 weather files and alot of other data. Moisture content in concrete alone has drastic effects of Thermal Conductance W/mK especially before free saturation (or capillary water region) in the 0-12 % by volume range it goes up ~ .08-.20. A "well insulated" building is not all that is needed. A good hygrothermal design temp/humid-stat can help/is.

Most static thermal calculators use a static simulation the OP has already performed. Some are better than others and the ones that consider moisture are better especially in moist climates and concrete.

OK, step back for a second. Yes, I agree that getting a real pro to create an accurate model would be desirable, but I'm pretty much at the end of the earth. Getting a qualified plumber takes 3 to 4 months. I am sure there are no "pros" let alone qualified amateurs within 2-300 miles of my location. I'm not designing the system from scratch, it has been working for 12 years, and for the most part the house has stayed warm. Since there are two water heaters with 4500 watt elements, I know they are generating about 30,000 BTUs per hour. I have run the data on the house through numerous on-line heat loss calculators and they all come back with a heat loss calculation between 27,000 btu/hr and 34000 btu/hr. I know I am not taking into account various factors that a professional would use and as a result this calculation is probably off by 10% or more. However, I empirically know that the 2 current water heaters are doing the job. Chances are good if I just replaced the water heaters with a boiler with the same capacity I will achieve the same results. My question and concern was not how big of a boiler I need to keep the house warm 24 hours a day, but rather how do I take into account that I am only running the heat source 8+ hours a day? (I have been doing that now for the last 2 years. 95+ % of the time the floor has maintained temperature throughout the day, but when the outside temperature has dropped on a few occasions below 0'F the heat has not been able to keep up and the floors never get to full temperature on those days. (but we are only talking about 3 or 4 days each winter). I think sailawayrb is right, putting in a larger boiler is not going to solve the problem as the slab can only hold so much heat. The answer seems to be that when the temperature drops to a certain level, I need to run the boiler on those days even though I am paying more than double for each kwh?

Kitdavis, when I said required surface temperature, I forgot to tell you how to determine that... First calculate your required upward heat flux which is defined as the required heat gain in Btus/hr divided by the heated surface area. So if we use 30K Btus/hr and 1000 sf of heated slab, the required upward heat flux would be 30 Btus/hr/sf. The required average surface temperature is the room indoor temperature plus half the required upward heat flux. So if we use 70F for the indoor room temperature and 30 Btus/hr/sf for the required heat flux, the required average slab surface temperature is 85F. I should also add that the slab will also give up heat as downward heat flux and the boiler needs to supply this too. Typically, a HR heating system is designed such that this downward heat flux is between 5 and 10% of the required upward heat flux.

In terms of the boiler size and given your heating approach, Ronmar’s calculation methodology is correct. What I can’t answer is whether you will be happy with the actual room temperature variations that you will experience given your heating approach. As you indicated, you could always run the boiler full time if needed on the days this may be problematic. So I would recommend sizing the boiler for 30K Btus/hr (and perhaps plus 10%) if that is indeed the correct climatic design value.

The thing about a high mass concrete slab is that the surface temperature doesn't change fast...which can be good and bad. If you have a well insulated/sealed building and use outdoor reset to modulate the supply temperature, all should be well. If you don't have a well insulated/sealed building, don't use outdoor reset, and the outdoor temperature varies significantly, you may experience significant indoor temperature variation.

sailawayrb - that is very useful information. (and I am currently completely violating the results ;-) The system was "designed" and implemented by a local plumbing firm. I am sure it was one of the first systems they ever implemented. (I had to search for quite a while to find someone that even understood the concept at the time). I am sure that there were no serious heat loss calculations involved other than some chart that said "plan" on xx btus/hr/square foot or something similar. They made two implementation decisions that did not work well that I have since corrected. Originally they installed a room temperature thermostat - with the result that by the time the room air reached 70' degrees enough heat had been put into the slab that 4 hours later the inside temp would be 85 or 90 degrees. I swapped the thermostat for one that sensed the floor surface temperature and corrected that issue. The second, more serious mistake was not having a way of stopping circulation when the water in the tanks dropped below the desired input temperature. The first winter we had a cold snap that resulted in the outside temperature dropping to -10 outside, and when the tanks ran out of hot water, the system just kept pumping cold water through the piping which lowered the floor over night to well below 60'F. I implemented a second thermostat connected to the pump that sensed the water temps and turned the circulation pump off when the tank temperature dropped below 100'F. However. using your formula above (which I have every reason to believe is perfectly correct) I should need to heat the slab to 82'F to achieve a 70' room temperature. (The resources to design the infloor heat system may have been lacking, but my contractor knew how to create a well insulated tight house). However, I have the sensor for the floor literally taped to the surface of the porcelain tile in a closet and I set it for 74'F If I test the floor with a infrared thermometer when the system is running at various places in the house, I get a good average reading of 74F. However, the room temperature remains a pretty steady 70-72'F. I either have a house that is much better insulated than I think or something else is off that I don't understand. As someone else pointed out I am heating 1400 sq ft with 30K Btus which is on the low side of what it seems is the norm. Mathematically it doesn't seem like it should work, but I have 12years of evidence that it does,

Heat flux is a directional vector denoted by an angle most often not 90 deg, not up & down or normal to the surface, and not just dependent on temp. Thermal Conductance (BTU/Hr, ft-F) or W/mK I gave values above. Those values have further implications due to liquid transport coefficent (ft2/sec), Water Content Lb/ft3, "Bulk" Density Lb/FT3, and some other factors (soil properties, etc). Anyone using hand calculations these days to determine this don't walk run for the hills away. WUFI used to certify Passive Designs is the best modeling software for this and there are others. It does take a pro to run. I understand the shortages all across the nation, bad designs, need, that is why I am still in business and it’s growing.

My question and concern was not how big of a boiler I need to keep the house warm 24 hours a day, but rather how do I take into account that I am only running the heat source 8+ hours a day? (I have been doing that now for the last 2 years. 95+ % of the time the floor has maintained temperature throughout the day, but when the outside temperature has dropped on a few occasions below 0'F the heat has not been able to keep up and the floors never get to full temperature on those days. (but we are only talking about 3 or 4 days each winter).
The answer seems to be that when the temperature drops to a certain level, I need to run the boiler on those days even though I am paying more than double for each kwh?

Ok, I thought you had changed the loads. The time delay question I can accurately determine but most use 24 hours and a lag time of 6-12 hours.
Seems like all you need is space heaters for those few days/yr the HR is not cutting it or pay the price on running it more. Stick with what you know, other than that without the proper design tools your guess is as good as any.

Kitdavis, it would appear you have a well insulated/sealed house...perhaps even less than 30K Btus/hr. At 30K Btus/hr, your upward heat flux is 21.4 Btus/hr/sf so your slab surface temperature to maintain a 70F indoor temp would be 81F. Please keep in mind that 30K Btus/hr is at the design outdoor temperature and you presumably don’t have anywhere near that high a heat lost the majority of the time. So your actual slab surface temperature required to maintain a 70F indoor temperature should be significantly less than 81F the majority of the time. Perhaps use your IR gun to create a time history table of slab, indoor and outdoor temperatures. The equation is straight out of John Siegenthaler's “Modern Hydronic Heating” and is used by all reputable HR design software. Anyhow, I believe you!

Yes, you most certainly want to use a thermostat that uses both slab and air temperature feedback when you have a high thermal mass HR emitter like you have. You should also use outdoor temperature feedback to modulate the supply temperature to the slab. Most boilers have this feature and it is called outdoor reset.

By the way, we have free software on our website to allow one to determine the relevant heat loss design parameters of an already existing building. This is often more accurate than any heat loss analysis, so you might find this useful:

Borst Existing Building Energy Usage Analysis Software

And with the relevant heat loss parameters known, you might find this software useful for optimizing and selecting the best heat source:

Borst Integrated Heating System Performance Software

Thanks for all of the responses. I feel more comfortable with my selection for the replacement boiler now. (I'm strongly considering the Thremo2000 Mini Ultra 9) One final question, replacing the hot water tanks with a boiler is going to require a complete reconfiguration of the system. Obviously, the mixing valve etc will be removed, but many of the other parts can be reused. Since the system is 12 years old and getting a plumber in these parts is challenging, are there some of the parts that I should just replace rather then reuse? In particular, I am thinking of the Grundfos circulation pump, the Spriovent, and the mechanical parts of the automatic Honeywell zone valves. Thanks again.

Have you considered the NextGen 8 or 12 kW?

A very interesting product, but of course the nearest dealer is in Quebec, only 900 miles away from Cape Breton. I don't think I would want to pay for that 15 hours of travel bill ;-) - I'll contact them and see if they have someone a little closer...

Yes, it is currently the only true HR appliance available in US. However, these HR appliances have been very common in Europe for decades. We have been installing these for maybe two years now and we highly recommend them for locations where an electric boiler makes operational cost sense. NextGen acquisition cost is about $1500. I suggest that you contact Dennis Schramel, NextGen project engineer, and see about having him work out a deal to direct ship to your [email protected]. Please feel free to tell him that you were referred to him by Borst Engineering.

I emailed them after your last message and I have already received a response from him. Certainly says a lot for their customer service!

Yes, their customer service is first rate. We have not experienced any issues with the installations we have done. I had lots of technical questions that Dennis answered impressively well. Dennis even sent us a set of spare parts in the unlikely event we ever do have a customer issue requiring timely service and they paid for the shipping. It is hard to find this level of engineering knowledge and customer service these days.

There are many HRF appliances available here and in Canada. Thermolec is my favorite Canadian electric boiler. There is no harm in up-sizing to a 10-15kW boiler.

http://www.thermolec.com/en/productview.aspx?type=product&id=58

Off-peak savings easily outweigh the incidental cost of operating at full rate during the really cold design conditions you refer to.

Thermolec doesn't currently offer a HR appliance, only conventional electric boilers. A HR appliance is plug and play and includes all the required HR components such as pump, expansion tank, differential pressure bypass valve and other advanced HR control features. The NextGen HR appliance has off-peak too.