I need some help figuring ways to run a relatively simple radiant floor system with 20 - 30 percent passive solar gain. 1750 sq.ft. slab on grade open floor plan, 6" interior drop north half to south half, about 13:1 primary mass surface to gain-window sq.ft. (mass walls & floor, 5/8" rock ceilings). Two zones, split by 6" drop (approx. split of gain space and non-gain space). Remote manifolds, 4 circuits each, circuits equalized each manifold (213' north & 238' south). 37.5k design load @ -15F... 6100ft. elevation WNW Lander, Wyo. 82520. Years ago I got advice from John Siegenthaler to isolate gain space from non-gain space and to run constant circulation due to gain space overheating potential. I (and we) thought the amount of primary mass surface along with constant circulation would avoid over heating as there is a fairly large amount of shaded floor area in and part of the gain space zone. I've read a lot about this over the years. What I truly understand is questionable. It seems mod-con boiler, maybe EMC circulator, a couple of 3-way diverter valves, a couple of small manifold circulators is the sum total of it... knowing that for sure is the question. One side issue is working with a plumber (decent residential/commercial experience) who has misgivings about valves (zone, diverter, etc.) and a long history of valve end-switch failure. He tends to primary/secondary and on-off with circulators. Anyhow, if you have thoughts, ideas, information and want to help, I'd appreciate any of it. Mike

Over controlled, overpriced, over powered but typical of today's design, if there is any.

The proper outdoor reset strategy following accurate heat loads will give the best results. Full time pumps, gain/non-gain design strategies are best left to the guys with all glass skyscrapers to deal with.

I use efficient low draw actuators (heat motors) and old fashioned Honeywell zone valves with end switches. I just replace a 35 year old model - end sw. still working.

isolate gain space from non-gain space and to run constant circulation due to gain space overheating potential.

I think that the first point is valid and the second should be "circulate everywhere when you have solar gain, preferably including a storage tank". I'd go with whatever MA says as to how practical this is.

Well, I read both of your replies and I don't know if I get either one. If you don't mind, would you run it by me again in a different way... maybe I can cross reference into getting it. Mike

I guess I need to start over. I've read a number of threads here to try and understand what gets help and what doesn't. The system was "designed" years ago for a house that is taking forever to finish (time, money, health). I talked with John Siegenthaler about 12 years ago and based on the open plan, passive gain in south half, he told me to split the house into 2 zones, one each for gain and non-gain space, and to use constant circulation. I then sent WIRSBO drawings and description of construction detail and they returned a Design Summary (room by room) and pipe plan. The piping was put in nearly as planned with the exception of equalizing circuit lengths as directed by plumbing-business friend I mentioned. There was little change in gain space piping circuit length and more change (not extensive) in non-gain. WIRSBO shows 37,460 Btu/hr under "Min. Heat Load"/"Total RPH Load" (spit 21,296 in gain, 16,164 in non-gain). I contacted John Siegenthaler recently. John told me he was too busy right now to take on the control system design. I need someone to do this but I also need to understand the "what and why" because: #1 - I'm going to have to work with my plumbing-business friend, #2 - I'm a person who HAS to understand to be able to deal with things. I'm not locked into any specifics on this system. I DO feel there is a real potential for over heating from passive gain. I feel, through reading John Siegenthaler's articles/books and reading ASHRAE papers by Chuck McClure, that constant circulation is really the way a high mass radiant floor panel works best. If I can get some more thoughts and ideas, it will help a lot. Mike

While the temperature is just right everywhere, why circulate? You might also do some reading on zones and why they are useful where there is solar gain.

Thanks, Jon... "While the temperature is just right everywhere, why circulate?" I felt the temperature was never really "just right" everywhere. That there was a constant flux and continuous circulation would even that. That under constant circulation the Delta T of supply/return was the true measure of the system and accounted for input factors like outdoor/indoor temp increase/decrease due to the instantaneous nature of the heat exchange at mass-floor/room-air interface. Jon, you said this: "You might also do some reading on zones and why they are useful where there is solar gain." I'm not finding references on the net or in Siegenthaler's 3rd. Ed. Can you point me to one or two comprehensive ones? Barring that, I have to fall back on my basic understanding of the situation... that there is a sudden influx of heat into a fully charged emitter system (seems it would include all mass... not just floor). Also seems amount of "overheating" (internal gains?) must be measured by ability of directly connected mass to take the gain. Disregarding the emitter for a moment... if all directly connected mass surface is not fully charged, it will take some amount of the internal gain. I understand, as the emitter sees temp difference, it is equalized instantaneously. What I don't see (only guess at) is a temp gain in emitter surface with a lower tube-water temp. My guess is heat will move to the cooler water... and if there is continuous circulation... be then moved to areas where the emitter surface is cooler than the water. As I look at the specifics of this house and run it across the standardized elegance of outdoor reset control, I wonder if indoor reset (pg. 358, Siegenthaler, MHH, 3rd.Ed.) doesn't have a higher potential for smooth control. I'm not saying that is right... simply don't know enough to do so, but it seems like the nature of sun coming in a window is going to be seen sooner (more accurately?) than through outdoor reset. I suppose the answer relates to the near-infinite aspect of a very-high-mass emitter but I don't know how to put the puzzle pieces together. Mike

I think you have it down. While I can't calculate how much, but generally you will be able to reduce the overheating of a solar heated room some by circulating to cooler areas, but not eliminate it. Outdoor reset isn't smart enough to say that it's 7am, let's let the house (and slabs) temp drop because there will soon be a lot of solar gain.

Jon...

And I don't know if there is anything that will help with the passive gain coming into a high-mass emitter space, really. If I understood, it is what Morgan said about the situation... "Full time pumps, gain/non-gain design strategies are best left to the guys with all glass skyscrapers to deal with." I just don't see a way dealing with it other than constant circulation. Or is it a matter of it can't be dealt with? If that is so, then what does the system look like?

As far as ECM circulators, I don't really know if two zones (relatively equal in most aspects) is a differential pressure well suited to the circulator-type.

Mike

I think you are asking for a simpler system that does not require sensors or computer logic. Using two zones and a relay/timer to always turn on circulation during the day would fit that. This doesn't preclude a outdoor reset thermostat to control the boiler.

These people are adapting finite element analysis to radiant design: http://www.healthyheating.com/Radiant_heating_designs/radiant_floor_heating_designs.htm. They also tend to be more receptive of questions that don't have an off-the-shelf commercial answer.

Love the Wind Rivers. Have you read John McPhee's Rising from the Plains? Fascinating part of the world.

Jon,

Simple that works well is good, but works well is really the driver for me. I'm still trying to get a working understanding of the solar gain effect... especially how it affects low end of modulation and how ODR deals with that. Seems like the internal gain will create more time in the year when the load is less than the bottom end of modulation ability... especially in the direct gain zone. So maybe a person sees a 0F day in the non-gain side and a below-modulation-day in the gain half at the same time. What's a boiler to do? =]

Todd,

Yup, this is a really special part of the world all right. I don't know anywhere else would suit... fun to see, maybe, but I think I'd want to go home. Thank you for putting up the link. I'll look there. I've been to both NRT.Rob's and Morgan's "stores" a couple of times. I know I need to get help with this... just hoping to get through the understanding part first.

Mike

Out door reset and zoning are the key. Full time pumps are possible but the loads and the whole system have to be designed at once. There is no cookie-cutter design for custom hydronics, let alone passive solar applications.

Regarding modulation range - water tanks help with that. Or it short cycles occasionally.

Maybe they already exist, but perhaps there is a market for a simple control system that can be programmed remotely - ie, you buy a few hundred dollars of hardware and a couple hours of custom offsite programming (through the Internet) by someone who understands HVAC to get an optimal system that includes temperatures, times, indoor and outdoor reset, zones, etc.

ODR cannot account for solar gain, but a boiler is always looking system wide. so imagine you had two equal load rooms. one has sun, one is in the back away from the sun. you might have zero load in the sun room and design load in the back room. then you're just at 50% load... but you would still need the water temperature requirement for design day or the back room won't heat enough, so you don't WANT ODR to adjust for this.

so the boiler modulates water temperature based on outdoor temperature, and it modulates OUTPUT based on return water temperature difference from its desired output water temp. return water will vary with the amount of heat actually removed... otherwise known as the load. if it needs 140, in other words, and X GPM at 120 is coming back, it puts enough heat in to raise X GPM up to 140.

If both rooms open, flow becomes X times 2, output doubles. it's a lot more complicated in real life as rooms with less load will come back warmer, but basically water temp and output modulate on different variables and it works out.

Indoor feedback is even better, but that's pretty much limited to Tekmar's TekmarNet 4 system, or Honeywell's IAQ system I believe, at this time at least. Maybe Uponor at the very high end.

Let me know how you make out. I am moving air to mitigate overheating events in my passive solar home under construction. Dunno how effective hydronic would be addressing acute issues. The floor's surface will be hot but it could take literally hours, thermal lag wise, for the heat to reach the pex. Hydronic could solve chronic distribution issues.
As for automation, I passed. I'll read slab temps at dinner time, consult weather.com, decide how much thermal lag I need to add to my radiant slab, and rely on mass to keep temp amplitudes within reason.

I apologize for not getting back quicker.

I'm pretty well stuck with the zoning/piping I got from Wirsbo with the heat loss calcs. I understand having a gain space with part of the mass in it being an emitter can be a problem (not at the time, but now). What I thought was a constant circulation scheme would likely help the situation better than any other scheme. Seemed especially helpful moving "too warm" to cooler in the south zone (20% of area... most of one loop) won't see direct gain and has a fresh air replacement drop in it).

Though the planned water temps in the 2 zones are less than 1 degree apart, the south zone seems like it will need lower temp water during gain times all the way from Sept. 1 through June 1 (or abouts). I guess that is another way of saying I don't understand how ODR and return temp modulate for that, or how the boiler keeps from running below lowest modulation point a lot (5:1 turn down on Lochinvar 50 is, in theory, 10kBtu/hr. and the heat loss calcs are likely to be a little high at 37.5k), without there being some further water temperature modification.

The north to south drop of 6" (not as ideal as 12" to 18") is supposed to cause convective loop from cool to warm and there are 2-each, north and south half ceiling fans. Whether any of that helps is going to remain unknown until it's a working system.

Todd,

I don't get the following

As for automation, I passed. I'll read slab temps at dinner time, consult weather.com, decide how much thermal lag I need to add to my radiant slab, and rely on mass to keep temp amplitudes within reason.

Mike

I agree with Todd that moving air around is going to be quicker and more effective in evening out temperatures.

It's curious thing trying to picture the ceiling fans and how to use them for moving air front to back (or opposite). They, at least, will provide enough breeze in an over-heated environment to make the elevated temps feel better.

Mike

I am using a heat recovery ventilator and duct work to move heat from downstairs, in the area of my 40 feet of window walls, to the upstairs which has little solar exposure. I'll need an hrv anyway because the house is so tight. I'll start out with continuous operation at, say, 50 minutes of recirculation (moving heat) and 10 minutes of air exchange. My hrv has adjustable fan speeds so at some point of trial and error I should have the house at a consistent temp most of the time. Plan B is a heat pump water heater installed in the vertical chase that connects the floors. Somthing as simple as snap switch could start the pump in overheating situations,which would then heat water by cooling the air.

I am holding off on solar hot water to see which source of DHW in the summer makes more sense, which is to say whether I have a predominant problem of overheating or underheating.

When I say I passed on automation, I am agreeing with NRT Rob. There is no thermostat-based system that can account for future solar gain. If you have a boiler operating on thermostat to maintain your preferred temperature at all times, you will have overheating on a regular basis -- hrvs, heat pumps, or hydronics notwithstanding. Concrete is a relatively poor conductor of heat. The heated surface of your floor in late afternoon will be radiating more heat than it conducts.

In my approach, I pick a single point during the 24-hour period, and again by trial and air, determine how much heat should be in the house. It makes sense to do this at dinner time because the sun has done its work by then. I add heat then or don't add heat. Next morning is a new day.