Minimal Radiant Heating Options
Radiant heat gets brought up here a lot. The consensus among experts is generally that using it in all your floors of a well insulated house is not worth the expense. But the idea of small patches of radiant has been brought up. For example – some of the kitchen, under the dining table, and the bathrooms. Heat source aside, what are some good ways to do this assuming a basement and hydronics being used? Warmboard over a portion of the floor with the rest of the subfloor being ordinary OSB subfloor? This seems like it would keep the subfloor level in the transition areas between heated areas and non-heated areas. I like the idea of the heat distribution being hydronic, allowing for a switch to an A2W heat pump in the future or using solar with a normal hot water tank as a large buffer.
In my current home, I installed a small amount of heated floor under my dining table and I love it.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I'm really, really happy with what I did in my new house. So happy I want to tell the world!
On the second floor there are three bedrooms and two bathrooms. It is plumbed as two zones, with floor heat in the bathrooms. There is only one thermostat for the entire floor, in the hallway -- but it's a two-stage thermostat. Stage one is the floors in the bathrooms, stage two is the bedrooms. With the thermostat set at 70F stage one comes on at 70.5F and goes off at 71.5; stage two comes on at 69.5F and goes off at 70.5F.
I log the run time for each zone. For the months of December through February, the bathroom floor heat was on an average of over twelve hours per day, while the bedroom heat was on an average of just 36 minutes per day. So except on the coldest days the bathroom floor heat is enough to keep the entire floor warm. It keeps the floors nicely warm most of the time, and the bedrooms are comfortable. No exotic control mechanism is needed.
I don't think this would work in a house with high heating loads, my house is tightly sealed (1.4 ACH50) and I have triple-glazed windows. Total design temperature heating load for the second floor is 8300 BTU/hr.
This is exactly what I am thinking. Switch the heat from the bedrooms to maybe the kitchen and/or dining room in my case, but same concept. How did you lay the pex? Maybe I'm over complicating it, but it seems the thickness of the radiant tubes would require either using something like Warmboard or adding a second layer of subfloor across the whole house to make up the thickness. Staple up from below would be very cost effective I would think, but probably not ideal for a heat pump system since I think it requires higher water temps.
There's a couple of ways of doing this. In my house, I transition from hardwood floors to tile going into the bathrooms. At the entrance I have a threshold which is a 3/4" step up from the hardwood floors and flush with the tile on the interior, so the tile floor is 3/4" higher than the hardwood. This is really common for tile.
I've also done it where I had to have the tile flush with the hardwood. In that case I ran pieces of 2x4 along the sides of the joists, about 1-1/2" below the joist. Then I put a piece of subfloor across the 2x4's, stapled the tubing to the subfloor and filled to the top of the joists with a sand mortar mix. I put cement backer board across the joists and installed the tile flush with the wood floor.
If it's new construction you can use smaller joists in one section to drop the floor.
Warmboard has to have a subfloor underneath it, it's not structural. So you need to either elevate the floor where you use it, or drop the subfloor.
Sounds good, those are relatively simple options!
I was thinking of the Warmboard-S. It's 1-1/8", and replaces your subfloor. Then you could use 1-1/8" normal subfloor for the other areas. But cost wise, I have no idea where it lands, I saw some estimates online from old threads mentioning $10+sqft.
I've never used Warmboard, every time I've priced it out it just seemed really expensive for what it was. Everyone who uses it says it's nice, though.
For what you're thinking aluminum plates stapled up may work too.
I would contact Warmboard. I sent them my plans from Chief Architect and they gave me a couple of quotes. The carpenters installed the warmboard and ran the PEX. Most HVAC companies do PEX hookups. There are pre manufactured kits that have a lot of hydronic components to simplify the BOM.
If you're using joist hangers, sometimes you can just drop the joist hangers for a section of floor and use the same size joists. Depends on how the framing runs.
Sounds like a good setup. What system do you use for cooling?
If you are only partially heating with radiant, why bother with anything hydronic. As soon as you say hydronic you are adding a minimum of about $10k to your BOM, that buys you a lot of electricity for resistance heat especially when this resistance heat is only a small fraction of your heat loss.
Simplify your life and install resistance mats where you want floor heat.
Once you get space heat under control one of you biggest energy uses is hot water. If you are looking for a way to store and offset PV power, the simplest is an oversized resistance tank. You can get controllers that will drive the tank heater as needed to use up excess PV this way all your hot water needs can be met by your array. If you really must have hydronic, you can also do a small amount of floor heat off this tank.
I'm more thinking along the lines of DC's setup. Lower load home, where most of the heating is handled by warmly heating a portion of the floors. In the range of 6000-12000btu/hr max heatloss handled by this setup. My idea behind hydronic is that it is more easily upgradable or modifiable from a heat source perspective. There were some posts here on how to build one relatively inexpensively, between copper manifolds and simple control. Potentially still a bit more than resistance, but it's not that far off from what I can see. Maybe labor costs is the bigger factor?
Warmwire is about $5 sqft + $100/thermostat. Say 2 bathrooms and a living room at 300 sqft of heated floor. So $1800 total.
Hot water tank ($400) + Copper Manifold ($100) + Pump ($260) + Controller ($400) + Pex ($320) + Misc Plumbing, expansion tank, etc ($400). That comes out to $1880, but it's possible I'm missing lots.
Both those above would be very similar on wattage.
Hydro solar sells kits now with their heat pump. $4500 gets you most of what you need to get going, minus some of the distribution. That gets you a split heat pump for heating and domestic hot water, and it should qualify for the IRA rebate. I'm not sure if that applies to all components or just the heat pump.
That is very optimistic price for hydronic. I have done the ultra budget ones and was many multiples of that in terms of BOM cost. All the bits and pieces add up quickly. Residential tanks aren't rated for space heat, some are combi rated but that means heat exchanger and extra pump plus controls, you can use a resistance boiler though.
Resistance floor heat will probably be around $8 to $10/sqft including all the bits.
Are they not rated because traditional hydronic setups require 180* water, or because they aren't up to the duty cycle of being on all the time I wonder?
Where on your BOM is most of the cost coming in? I think if you did a full house, added zone valves, etc, it would quickly start to add up.
The other thing to think about is cooling. If you're in a heating-dominant climate, install heat pumps and air handlers, but size them for your cooling load. Then install floor heat to make up the difference.
If that gives you less floor heat than you want, there's no harm in putting in more.
It's looking like the hydronic fan coil units are becoming pretty affordable as well, and I like that I can add one to each bedroom, and one for the main living space.
I'm using hydronic fan coil units. I like them a lot but I'm not sure they're for everyone. I've gone with air-to-water heat pumps for both the floors and the air handlers. The parts are a lot more expensive is the main drawback.
What I like:
Putting an air handler in each bedroom is no problem at all, the exact opposite of mini-splits. First, the smallest unit in the line I'm using is 3,000 BTU/hr, about half the size of the smallest minisplit heads available. There is no problem with oversizing, they can run all day at very low output. They have an ultralow fan speed where the relay clicking on and off literally makes more noise than the fan. I can control all of the units from one thermostat with a zone valve, if you shut off the flow of water they turn off.
One thing you're going to have to work out is that the radiant floors use water at about 95F, which may be a little low to get much heat out of the air handlers. I needed two heat pumps anyway to meet my design day heating needs, so I plumbed up two separate circuits, one for the air handlers and one for the radiant panels. That works well because the heat pumps have automatic outdoor reset where the water temperature changes based on the outdoor temperature. The outdoor reset isn't going to work if you have mixing valves.
Thank you, appreciate all the insight! My thinking is similar to yours, use the sections of radiant floor to do most of the heating, maybe a small amount of supplement from fan coil units. And then in the summer, the fan coils take over for cooling.
If you look at the usage in most houses, the radiant floor only needs to be hot for 8h to 12h per day. The rest of the time, it can be off. If you are looking at say 6000BTU output (600sqft at 5F above room, which is a fair bit of heated area) for 8h per day you are looking at only 18kWh/day. At my electricity rate, that is $2.75/day, even if that runs for half a year, the ROI on slightly higher COP hydronic AWHP is never. Add a couple of extra PV panels on the roof to supply this load and you are way further ahead.
Akos, I actually made a pretty in depth spread sheet taking into account equipment costs + rebates, my solar distributed generation program, and heat loads based on average temperatures (not design load) to compare resistance+solar vs heatpump with less solar. There are some assumptions as to when heat is being called vs when the sun is shining for solar costs. It's a close call over 10-20 years if you assume the heat pump is taking on all your heat load with something like a 10,000 btu design load.
Your assumption that it turns off for half the day, is that because another heat source is supplying or because the loads are 3000btu/hr average over the 24 hour period?
Tim,
My assumption is there is a ASHP supplying the bulk of the space heat 24/7. When the floor heat clicks on, it will run a bit less, so my operating cost is actually too high as you are only really paying the cost difference between a COP 3 and COP 1 heat source for that bit of heat. This would definitely put the ROI well beyond never for hydronics.
Maybe if you really sharpen your design pencil and you can get something like a Senco2 system to run in combi mode it might work as now you get your DWH at higher COP. This still won't be cheap but won't cost as much as a full AWHP setup. You still don't want the Sanco2 to supply the bulk of your space as the COP of these in combi mode is pretty low, better than resistance but nowhere near an ASHP.
Ok, sounds good, that makes sense, I appreciate the feedback!
Being able to supply 100% of the domestic hot water would be useful. I estimate DHW being another 10-14kw/day for a household, year round... Having a split heat pump that hits 5-6 COP on that portion in the summer helps a lot with payback period.
If you're doing it the way I suggest in post #1, where you prioritize the floor heating because you prefer the feel of it, then electric in-floor no longer makes sense. In my setup roughly 95% of the heating over the season comes from the floors -- and I like it that way.
We have Warmboard installed on both 1st and 2nd floors. We set the tile areas to 72F and the rest of the floors to 69F. kind of like a 2 stage control. This keeps the bath floors very comfortable for bare feet. The rest of the house stays warm, I.e. the second stage comes on when the temperatures drop below 0F. The water temp is kept in the 80F to 100F range, and comes off the water heater with a Taco XPB heat exchanger.
So you don’t need to install Warmboard throughout the house. We did the whole house thing with Warmboard-S and ripped out the old squeaky OSB sub-floor and replaced it. Got rid of the hot water registers. Air sealed and installed new insulation. No regrets.
Glad to hear it! It does seem like a nice product, and I think their pricing includes all their design work, which can be nice, especially for whole home setups.
(Replying to Akos #19 because we can't nest any deeper).
So putting the numbers into a mortgage calculator I get the annual energy savings supporting about $6000 in additional mortgage. Of course there are a lot of things you can argue about with that but it gives you an order of magnitude. It probably only comes close to making financial sense if you're already doing hydronic and even then likely you have to be doing the work yourself (and undervaluing your labor).
This is one thing that might be the tipping point for me. The hydronics work is something I'm comfortable doing, and it interests me, so it would actually be enjoyable/satisfying work I think.
I need to check the language in the IRA bill, but it would be interesting to see where the line is drawn on components for a system like this when powering by a heat pump. It's possible a lot will qualify, so the max $2000 federal refund could be taken, putting another large dent in the costs.
(reply to post #21)
"My assumption is there is a ASHP supplying the bulk of the space heat 24/7. When the floor heat clicks on, it will run a bit less."
But you can't do that. If you want the floor warm, it's going to emit as much heat as it's going to emit. You can't say, "I want the floor at 80F but not emitting any heat because I have a heat pump too."
Having resistive heated floors is kind of like having heat strips or electric baseboards to make up the difference between what the heat pump can produce on the coldest days and what is needed. The difference is that instead of asking the resistive heat to produce the last few BTU's, you're asking it to produce the first few.
Note that the financial argument for using resistive heat for the floor is similar to the argument for supplementing the heat pump with resistive rather than going to a larger heat pump -- the long-term cost doesn't justify the additional equipment cost. In a heating-dominant climate there's an additional argument that an oversized heat pump doesn't deliver comfort as well in the cooling season, so you're better off sizing properly for cooling and then supplementing with resistive for heating.
I think this is a great concept. The usual argument against radiant floors in a tight house is that you aren't going to be able to run the floors warm enough to be enjoyable because the space would overheat. The way you are doing it, you completely avoid that problem. The floor become your "baseload" heating running continuously most of the heating system and the other emitters are the "peakers".
I also hope that increasing popularity of hydronic will drive the costs of it down. I think that's very possible. Hydronic has lots of advantages, including compatibility with large tanks to store heat to allow pausing high electric consumption during times when the grid short on capacity and storing extra heat when the grid has plentiful renewable energy available.
Charlie,
That seems to make a lot of sense to me too.
Note that you only really need "peakers" if you're in a heating-dominated climate and the heat pump sized for your cooling needs isn't enough. Not that you can't install something you don't need -- HVAC is all about comfort, after all.
I didn't mean to imply that that peakers would need to be additional heat pumps or other heat sources. Just extra emitters whether fan coil or panel radiator, perhaps on the same heat pump or on a second heat pump, according to the load as you say.
Another big advantage of hydronics is that oversizing isn't a problem. You can have a 6K BTU/hr air handler and only need 100 BTU/hr and it will provide exactly that, with no sacrifice in comfort or efficiency.
The apples-to-apples comparison with a minisplit is a heat pump with one or more air handler units. If you look at 1:1 units, it's pretty lopsided.
A site like HVACDirect.com will sell you a 12K 1:1 system for under $1100. Something equivalent from Chiltrix would be a CX34 heat pump ($4,889) and a CXI120 indoor unit, $1,199 for a total of $6,088, or more than five times as much.
With more advanced setups it's a little more balanced. Let's look at 4:1 systems, HVACDirect will sell you four 6K heads and an outdoor unit for about $6,000. You could use the same CX34 heat pump ($4,889) and four CXI65 heads ($839 each) for a total of $8,245. It's still more, but not outrageously more.
The thing is, I don't see anything about the technology that makes the air-to-water fundamentally more expensive. Rather, I think it should be cheaper -- it's easier and cheaper to run water around in PEX tubing than it is to run freon around in copper tubing. You don't need an EPA license to handle water! I believe that the cost difference is entirely due to hydronics being a niche product.
Hydronics is very popular in Europe and China. If you look on AliBaba you can get hydronic equipment for less than the equivalent minisplits.
Say you have an 80sqft bathroom with a 500BTU heat loss (20F outdoor, 70F indoor). If that is supplied by a ASHP at 50btu/cfm (~120F supply temp), it will be running at 10CFM.
Tile to feel warm you need 80F, but 85F is nicer. So you set your floor heat to 85F. The steady state temperature of the bathroom will be about 83.5F. At this temp the losses will be slightly more at 635BTU.
At 83.5F the floor heat supplies:
-(85F-83.5F) * 2 * 80sqft=240BTU
ASHP supplies:
-(140F-83.5F)*1.08*10CFM=394BTU
So the heat pump sill supplies a fair bit of the overall heat need.
The problem you can run into if the thermostat for the ASHP is near a lot of floor heat, this should be avoided or at least minimized by maintaing a reasonable floor temperature.
You can re-run this math for a heated kitchen area, as long as this is a reasonable fraction of the main floor and you don't crank the floor heat, the bulk of the heat will still be delivered by the ASHP.
Depending on the design loads and the amount of floor heat, in shoulder season the floor heat can be a large percentage of the house heat. In this case, probably best to turn the floor heat off in anything but bathrooms.
What your example leaves out though, is that if you have one room at 83.5F it's going to be losing heat to the rest of the house. The Manual J model assumes that the entire house is being heated to the same temperature so there is no heat flow between rooms.
"Depending on the design loads and the amount of floor heat, in shoulder season the floor heat can be a large percentage of the house heat. In this case, probably best to turn the floor heat off in anything but bathrooms."
This is true. In my house often the only heat that is running is the bathroom floor heat, it's essentially heating the entire house.
Good point. Assuming an open door flows about 300CFM, the above example would stabilize around 75F. At that point:
-floor heat: 1600BTU
-ASHP: 450BTU
And a 75 F bathroom can be a nice luxury. You might even save energy since you might not run that hot water in the shower quite as long if you knew you were still going to be warm with it off.
I converted to radiant (from two furnaces) for a midcentury modern remodel. Walls upgraded to R34 and roof upgraded to R45. Blower door test next Friday to determine how airtight it is. Ducted system would not have worked I refused to have soffits or exposed ducting. The house is 2800sq ft on the main level with 1000 sq ft basement. We converted a ‘sunroom’ that is north facing into living space. Radiant was the best way for us to manage multiple zones appropriately. I used a 12 zone Cross manifold for simplicity. I built a detached garage and retrofitted the existing attached garage for radiant as well. One boiler does everything including DHW which is the biggest load.
Most of the house is ran in the walls, both garages in slab, and the ‘sunroom’ on top of subfloor with thin slab overtop. I believe it was the best fit for this project and running in walls saved quite a bit on tubing. Every bedroom is its own zone, the master bathroom is its own zone, basement, sunroom, and both garages are separate zones. We built a shed and even ran radiant in that slab as we had an extra zone free on the manifold. The main living room/kitchen is about 900sq fr and we did a wall mount mini split for that area which was cheaper than tubing (we had no wall area and the subfloor is 2” thick so in floor was not practical). The cooling load is so low that that single unit should be all we need. We’re a week or two away from firing the system up I’m excited.
It’s not always a comfort decision sometimes the project and expectations necessitate zoning and the additional amenities that come with radiant. That’s my opinion anyway.
Good luck
Yeah, a big factor for us was not wanting to have soffits.
"I used a 12 zone Cross manifold for simplicity." I have to admit I laughed out loud at that. Four zones for simplicity I would have bought.
I would read that as "manifold for simplicity", which can be the case. I've recently had somebody proudly show me their wall of hydronics, almost an entire side of a basement was covered in very neat beautifully routed copper work of art. I didn't want to tell them that I comfortably heat a place larger than that with an extra story using 8 zones on three thermostats done in mostly PEX that fits into a 3x4 mechanical room including boiler under the basement stairs.
Old MCM is definitely hard for ducting. On the other hand, it is pretty easy to design a new build with no bulkheads and couple of thicker walls as service cavity. You already need to fit ducting for ventilation, on a low load place HVAC ducts are maybe 1 or 2 sizes up from that.
I've recently helped with duct layout on a gut reno of a century home. There are no ducts or bulkheads anywhere in the place, even the main supply trunk is buried in a wall. Took a bit of thinking and careful planning but not impossible even in a place with lumber floor joists.
What is the square footage of glazing, square footage of your home, and wall construction? Any skylights? Do you heat your garage as well? Any other accessory buildings? If your answers are similar to mine, 4 zones would be irresponsible. I won't make the argument 12 zones was necessary. I will make an argument that to offer individual controls for each room is a luxury that I rarely see in housing. I don't know if you're familiar with the cross manifold but the operation is the same regardless of zones. The added cost is thermostats, thermostat wire, and an upgraded manifold. I have no affiliation with the company it was a decision I made to keep the mechanical room clean and simplify the install. This response is to DC.
Anyhow, the sunroom is 6" OC spacing because of the heat load. We ran Pex-AL-Pex in the walls it was difficult to work with because it kinks easily. We have a few repair couplings. The return end panels were well intended but the pex didn't snap into the groove like it does the heat transfer plates, in hindsight it wasn't worth it. Most of our window sills were already 4' in height which worked well with walls. I don't know if this helps the OP just putting it out there for anyone else who comes across the thread.
Akos, my mechanical room is 3'x4' as well. We had sheetrock hung, taped/textured, and painted before installing everything. After installing a dedicated ERV I realized we could have ducted some of the house. By then I was already deep into the radiant overhaul.
I can't throw stones, I have nine zones in my house.
I'll say that the more complex your system, the more competitive hydronics becomes.
I you just have one zone and a single 1:1 minisplit can provide that, that's hard to beat. It's when you have lots of zones and sophisticated controls that hydronics starts to have an edge.
There's a reason hydronics dominates in the commercial space.
If you have two heating systems, and each just has an on-off thermostat, they won't split the heating load. One will have a higher setpoint, and that one will do most of the heating, the other will only come on when the first one is unable to provide enough heat.
Imagine you live somewhere where the 1% design heating temperature is 20F. You have two heat sources, each rated to one half of the 1% heating load. Stage one has a thermostat set for 70F, and stage two has a thermostat set for 69F. Stage one will run any time heating is needed. Stage two will run only when stage one can't keep up, which is temperatures below 45F (The midpoint between 70F and 25F). At temperatures below 45F, both systems will run, but they won't share the load equally. Stage one will run 100% at all temperatures below 45F, but stage two will only provide what stage one can't. So if it's 44F outside, stage one will run 100%, stage two will run 4%. At the 1% temperature and below both systems will run 100% of the time.
So how much will each system run overall? If you assume that temperatures follow a normal distribution, you can use the area under the bell curve to estimate an answer. In a bell curve, the tail with 1% of the area is 2.3 standard deviations from the median. So with a 20F design temperature, 70F interior, you get one standard deviation is 50/2.3= 21.7 degrees. The point where stage two turns on is 45F, which is 1.15 SD from the median. Just 12.5% of the area under the curve is to the right of that point, so you would expect that stage two runs at all only 12.5% of the time.
But the two systems don't split the load 50-50 -- stage two only runs when stage one can't keep up. At 45F stage two runs 0%, and at 20F and below stage two runs at 100%, so if you average those two -- that section of the bell curve is approximately straight -- you get that when stage two is running, on average it runs at 50% capacity. So if it's running 12.5% of the time, and on average it runs at 50% capacity, it runs at 6.25% on average.
Stage one runs half the time, the whole heating season. At 45F and below -- 12.5% of the time -- it runs at 100%. The rest of heating season -- 37.5% of the time -- it's between 45F and 70F it runs part of the time. To save some calculus let's say it averages 50% during that time. That gives an annual average of (100%)*(12.5%)+ (37.5%)*(50%) = 31.25%.
So if stage one runs 31.25% and stage two runs 12.5%, that means stage one provides 84% of the annual heating and stage two provides 16%*. This is why it's a good idea to have resistive heat back up a heat pump, but not to have a heat pump back up resistive heat.
An additional consideration is that stage two runs only on the coldest days, when the advantage of heat pumps over resistive is the smallest. If stage one is heat pump and stage two is resistive, that's good. Vice versa, not so much.
*(This is kind of consistent with what I reported in post #1)
The way I have found to run these is to use a thermostat that does floor and air temp control for the floor heat. You set this thermostat to maintain the floor surface at some minimum temperature and the air temp setpoint bellow your ASHP setpoint. This way the floor heat only runs enough to keep the floor warm, as long as this is not too far above room temperature, it won't effect the operation of the house heat. If the house heat starts dropping out when very cold, the floor heat will automatically crank up and assist. I don't have hard numbers of how much energy this combination uses but it seems to work well enough as the ASHP is running most of the time.
That would be like a 3-stage system. Floor temperature is BTU output, so the first stage is whatever the output is at that minimum floor temperature. Second stage is the ASHP. Third stage is the floor cranking up when the ASHP can't maintain temperature.
The choice between those approaches depends a lot on what your heat sources are. If you have an air-to-air heat pump plus a fossil boiler, and you are trying to minimize the fossil fuel use, your scheme sounds good. And if you reduced the percentage of the floor that is heated, you'll use less fossil fuel to keep that area warm all the time.
But if you are doing the floor from an air-to-water heat pump, there's not much reason to use the air-to-air heat pump at all until you have maxed out the heat delivery you can do through the floors.
Doing a little more research on the control side. I found it interesting that chiltrix includes their controller, which looks pretty advanced, with the purchase of their heat pump. The other A2W on the market like Arctic, HydroSolar, and MbTek require purchasing controls. It seems like for a couple zones, you are quickly looking at $1000-1500 for just electronics between zone valves, zone control, heat pump control, and thermostats. Not sure there is any way to simplify that.
You also have to be careful that the standby power of the controls doesn't outweigh their benefit. Primarily that's a problem with zone valves that consume power whenever they are on, and there are modern ones (e.g. Zone Sentry from Taco) that use much less, so it's a solvable problem.
This is where you have to sharpen your design pencil. Those fancy manifolds with a zone actuator for each branch add up very quick.
Instead of one big manifold, what I found works the best is to get everything onto a single pump (delta P or delta T ones are great for this), a couple of in-line motorized valves for each section to a small manifold. This reduces the number of thermostats you need and you can still balance the system by adjust the flow in each zone. If you want micro zoning, TRVs are your friend.
Generally an indirect adds more cost and complexity than it is worth.
Also check the flow rates and pipe sizes. There are very few things that will need anything more than a 3/4" line in a low load place. Even 1/2" might be oversized for a lot of it but it is hard to find smaller if you want to stick to standard fittings.
If your design is starting to look like this you are doing something wrong:
https://www.pmmag.com/ext/resources/Issues/2021/09-September/Hydronic-Heating_Hero1-Spacepak-1.jpg
That's good if you want people to think your basement is the control room of a nuclear submarine.
If you count the circulators there are only 8 zones plus indirect, so it is actually not doing all that much.
Probably don't want to know what that cost though.
Deleted
Newb question - are these motorized inline valves different from zone valves?
Same as zone valve like Grundfos UP-ZV or Taco zone sentry.
Direct acting actuators meant to mount to zone manifolds add up pretty quick and usually you don't need control to each zone.
It is always good to learn something new but DIY air to water might not be the cheapest place to start.
If you want a simple and effective hydronic system take a look at the build blog here:
https://www.greenbuildingadvisor.com/article/flatrock-passive-taking-a-tour
I found this page had the best description of the system:
https://flatrockpassivehouse.blogspot.com/2018/04/fire-it-up-commissioning-hydronic.html
Panel radiators, thermostatic valve on each radiator, constant pressure circulator. No zone valves or thermostats.
Thanks!
There's a full description here: https://flatrockpassivehouse.blogspot.com/2018/04/fire-it-up-commissioning-hydronic.html
Low-temperature pane radiators, thermostatic valves on every radiator, constant pressure circulator. No zone valves or thermostats. They're doing it with wood and resistive electric (which wouldn't be my first or second choice) but a setup like that would also work really well with a hydronic heat pump.
Deleted
There's a better description on his blog but GBA won't let me post the link, let me try this:
https: (slash slash) flatrockpassivehouse.blogspot.com/2018/04/fire-it-up-commissioning-hydronic (dot) html
There's a better description on his blog, but GBA won't let me post a link.
The blog is called flatrockpassivehouse on blogspot dot com. The post is from April 2018 and is titled "Fire it up!: Commissioning the hydronic heating system."
They have low-temperature flat-panel radiators with thermostatic valves on every radiator. Constant pressure circulators, water design temperature of 120F. No zone valves, no thermostats. The heating source is wood and resistance electric, which wouldn't be my first or second choice. But it would also work really well with an air-to-water heat pump.
I do think I read that one. I like the simple design.
Would it be crazy to use thermostatic valves with floor heat? I guess the risk is over/undershoot on a system that is slow. My mother-in-laws house is all thermostatic valves on radiators. Simple, but I really like the simplicity.
DC,
http://flatrockpassivehouse.blogspot.com/2018/04/fire-it-up-commissioning-hydronic.html
Thanks Malcolm. If I post that exact same thing I get "pending moderation." You obviously have powers I lack.
In the simplest configuration you can hook the Chiltrix heat pump directly up to an air handler unit (AHU) with no other controls. The AHU has a thermostat on it, which controls the fan speed and thus the heat output. The heat pump has a built-in variable-speed circulator and it adjusts the circulator speed and the compressor speed to match the demand for heat, by monitoring the temperature drop on the water that is circulated. A small amount of water is circulated at all times through all of the AHU's. If no drop in temperature is detected it means the fans aren't running, and the compressor runs just enough to keep the water temperature close to the set point.
If a room needs heat, the fan in the AHU kicks on, and the return water temperature to the heat pump drops. The heat pump then speeds up the circulator to keep the water temperature near the set point, and modulates the compressor to match the demand.
The AHU's have variable speed fans and adjust the fan speed to meet the need for heat in the room. The whole thing is actually quite slick: you set the room temperature at the AHU, and the water temperature at the heat pump, and then the fan speed in the room, the circulator speed and the compressor speed all modulate to keep the room at that temperature. I find my AHU's keep the room within 0.5F of the setpoint. You can put as many AHU's as you want on one heat pump, in parallel. If you have many and the piping runs are very different in lengths the manual says you may need to put balancing valves to keep the far away ones from being starved.
That's the basic configuration, and with 4+ heads it starts becoming competitive with minisplits, pricewise. Where it gets complicated is if you want to mix in other kinds of hydronic emitters. If you want to have radiant floors or conventional radiators you need to have thermostats and zone valves to control them. With zone valves you have to have a buffer tank so that the heat pump doesn't short-cycle when all of the valves are closed, and the buffer tank means you need an external circulator. So it gets more complicated.
The AHU's have a 24V output for a zone valve so you can also use them in a configuration with zone valves.
Seems like a "Heat Pump - Hydro Air System"
(Responding to Tim #55)
"Would it be crazy to use thermostatic valves with floor heat? "
Conceptually no, but there are two catches. First, you'd have to use a constant-pressure circulator. Traditional zone valves have a switch that closes when the valve opens to turn the circulator on and off. With a constant-pressure circulator the circulator will turn itself off when all the valves are closed.
Second you'd have to figure out a way to mount it and to get the plumbing to it. They're meant to sit on the inlet to a radiator and regulate the flow of water into the radiator, you'd have to figure out a way to mount it on the wall and have the tubing for the floor go up the wall to it. Since they're meant to mount on a radiator they are designed to compensate for the heat coming off the radiator when they're on, the compensation may not work when they're mounted away from the floor.
I'm not sure what it buys you though. Part of the appeal of hydronics is that you can have a thermostat in every room but control it all centrally.
Thank you, this all makes sense. Zone valves aren't particularly more expensive than a TRV it seems anyway.
You can get TRVs with remote sensor and dial, these have a small capillary tube that runs from the dial to the operator on the valve. The cost of these does get close to thermostat + valve actuator. It does simplify local controls though, in most cases you rarely if ever adjust the thermostat setpoint in a low load home so the accuracy of a TRV is moot.
I put together a fairly accurate BOM (I think) using HydroSolar's heat pump, thin fan coil units, and Taco circulators. 3 tanks, one being for heat/cool, one for domestic hot water preheat, and one for DHW. Total comes in at $11k, so right on Akos's estimate. I built this as a single circulator, 4 zones (1 being garage), and 4 fan coil units (again, one being in the garage). I imagine having it all built and installed would double that cost.
I'm not sure if the fan coil units need the buffer tank or not. If not, one could use the buffer tank for radiant heat only and DHW preheat, with a 3 way valve to run the fan coils, eliminating one of the three tanks.
"Taco circulators. 3 tanks"
You are already going down the expensive path. A simple inexpensive system should have a single circulator and no buffer tanks. Maybe one buffer tank if you really absolutely must have micro zoning.
To put this in perspective a budget slim ducted cold climate heat pump is about $2k.
I think a lot of that is added due to incorporating a heat pump that is doing heating/cooling/DHW.
I think when you look at a ducted unit like a Mitsubishi, you are looking closer to $3500. Add a mini split for my garage. Even at $2000 though. Warmwire cost for some electric floor. In comparison to the 3 tank zoned system, it comes in about $2000 cheaper with greater operating cost. I think because I am counting in my garage it messes with the estimate, you obviously can't just tie a duct in for that. But in hydronics, I could just add it as a zone with either a fan coil unit or fan Coil+slab.
I attached a screenshot of my spreadsheet. You can see where I break down costs. Of course, this is all material costs only. For resistance, I included 2 bathrooms, a section of living room and a section of garage.
With all that said, if using a heat pump, you can only go so simple on the mechanical side. My thinking with this post was more on the distribution side, as I think that can quickly add a lot of cost as well, and it's relatively pointless to run heat under more than a few select sections of floor. Without the heat pump, I agree, delete the buffers and everything. A pump, maybe a couple valves, and an electric boiler would be simple.
I'm not a big fan of combining DHW with the heating.
Of course my preferred alternative -- a heat pump water heater -- is pretty expensive too, I paid $2300 for my 66 gallon tank.
Why is that? Complexity of the system? Seems like efficiency, especially in the warm months is huge.
I haven't been able to answer the efficiency question.
During the warm months, the heat pump water heater provides free cooling and dehumidification. If you're running AC, you come out ahead taking the heat out of the house. If you're not running AC, presumably you have the windows open and it's the same either way.
During the heating season, I haven't seen anyone make a convincing case that it's more efficient to heat the hot water in one step from outdoor air vs two steps. I started a thread about it here a few months ago and nobody knew.
I just checked out that thread, and I thought Akos answered it very well.
https://www.greenbuildingadvisor.com/question/heating-domestic-hot-water-with-a-heat-pump-one-stage-or-two
What lingering doubts do you have?
The savings of a Heat pump to heat pump vs using the split HPWH directly are calculatable, but I can't remember the exact equation now. I was dumb and didn't save all that in my equation in excel. It's below. Ends up being the COP of both added together over 4. It's the 4, I can't remember exactly where that comes from now.
This spread sheet shows some high efficiency mini splits supplying heat to a HPWH and the ending COP at various temperatures vs a split heat pump supplying directly at 113F and 131F. I suppose if it can output 131F water, you really don't need a preheat tank.
EDIT: Based on the reply above, my math is different from Akos... His is showing better for a system efficiency than my math.
Akos' answer in the thread was: "So overall, it doesn't look like it matters much."
Tim, the thread I just linked has the right math. With the 4.07 and 4.2 combination, it comes out to 1.84.
The answer depends upon the actual COP's of the equipment, which depends upon conditions. There's more variability in actual COP than there is between the two configurations so there's no set answer.
With all that said, if efficiency is a wash, and you already have the split heat pump, why not use it for DHW?
(Replying to Tim_O, #78)
With my Chiltrix heat pump, adding DHW is about a $3000 option -- you have to use their tank, which is $2300, plus a 3-way valve ($180), plus backup element and thermostat, plus possibly a booster circulator ($300) plus shipping, etc. It's cheaper to go with a HPWH.
Plus I'm not sure I like the way it works in the summer, if there's a call for hot water and it's in cooling mode it switches to heating mode, heats until enough hot water is produced, and then switches back to cooling. If it was scavenging waste heat from cooling I might be interested but it's not doing that and it seems less efficient than having a separate HPWH inside the house.
DC, There's no set answer, but it's not like there's a big unsolved mystery. And most importantly, we know that neither configuration will be much worse or much better in the winter, so you can choose based on other criteria.
I guess the other thing that we should check, however, is the summertime performance comparison. You cite the advantage of "free" cooling inside from the HPWH. But are you giving up the opportunity to get super-high heating COP for the hot water based on the high outdoor temperature? For the Chiltrix, it looks like not--it looks like it's still about COP = 3.3 for the highest outdoor temperature it's rated at, to produce 130 F water. That's on par with the HPWH, and doesn't provide the side benefit of cooling and dehumidifying indoors.
So my summary would be that the efficiency is about the same either way in the winter, and the efficiency advantage goes to the HPWH in the summer. Other than efficiency, the HPWH has the advantage of being more standard equipment, but using the main heat pump for water heating might be cheaper overall.
Sorry, but this appears to be race to the bottom rather than the comfort of the people who are living in the house. I have Reynaud’s Syndrome so I keep the floors at 72F in my study. No flare ups, steady comfortable. The zoning with hydronics allows me to stay comfortable because I can microzone the rooms. Can’t do that with a furnace or a mini-split. Not to be rude but design the space and heating for people. Good buildings will follow.