you are using incorrect slab loss numbers

jp_2

after reading the post about the alaskan house.   and the house on a slab not getting up to temp and the responses, I start wondering more about slab insulation.



according to rpa web site,  inside temp - outside temp   times    .125 equals(about) what the R value should be for perimeter insulation.



this is incorrect for a heated slab and the numbers you get will have big errors in them.  I'd agree if it were FHA or BB heat.



with radiant slab, the ground does not 'see' or care about inside air temperature, but  'sees' the temperature of the bottom of the slab, most likely where the tubing is(at the bottom) .



in the house on a slab post, the house tstat wasn't at 70F, but the supply apparently was at 120F, some suggested 130F for supply.     so if  the tubing is at the bottom of the slab, the slab at ground interface would be more around 110-120F, not 70F as inside air temp.



now your temperature numbers are off by 40-50F. and your expected heat load is way off too.  



this may not matter much in a large building, but a small house....???  



Food for thought....







    

Mark Eatherton

Room loss versus panel loss...

jp,



There are two losses from a RFH system. The theoretical conductive losses from the room (70) to the ground (45). This is the true conductive loss thru the floor and is coverd by the .125 factor.



The "back loss", or "downward loss" from the panel is usually covered by the RFH load calculations and is taken into consideration in the overall picture and grand scheme of things. The same would apply to any radiant surface, and can actually be worse than a RFH system. Consider a radiant ceiling. The outside temperature potential is much lower than the ground surface temperature. Hence the recommendation of Siggy to DOUBLE the minimum required R value behind a radiant ceiling panel heating system...



Good observation in any case. Here's a link to a manufacturers load calc and you can see where the back loss of the panel is taken into consideration. http://www.heatinghelp.com/files/posts/7383/HeatLossDetail%20-%20ttttttt.pdf



ME

jp_2

no conductive room loss.....

but mark, if the rooms heat source is the floor, the floor will be warmer than the room, therefore there is actually no room losses through the floor, but the floor itself will loose to the ground, being that it is the only heat source.  if the room is 70F, won't the floor be more like 76F? (slab)  so areas very close to the tubing will be more in the high 80's.



but this too brings me back to that post where the guys slab supply was running at 120F.    and I wonder,  that little project I did is running in the mid 9o's(supply) and the room is a little bit too warm.  I realize its 1.5" gyp and not 4" concrete.



seems generally people here recommend R19 under a heated floor, crawl space or basement.



I will read more.   as I told brad, up here the code calls for 3 in foam on foundations, or you could say R15.   from what you said, I will reread more, but sounds like a radiant slab should have 2x insulation based on siggy?  or R30? 



you are throwing me a bit by the use of "panel" I keep thinking of a flat panel wall radiator.



again, I'm mainly thinking about perimeter insulation, and maybe 4ft in?







Added: so yes, you could say I'm calling into questions the .125 factor or the wrong inside temp being used.

Gordy

Conductive losses

 Mark I agree with what you are saying until you got to the radiant ceiling panels being "worse", and Siggy requiring doubling up on insulation.



 I'm not challenging you or Siggy only questioning. Would a ceiling panel not be more of a "convective back loss" verses a radiant floor panel on ground being a conductive loss. There for the conductive loss needing actually more of an R value than a convective loss under the same temperature differential.



  The only reason I say this is given the attic insulation  quantity when the house was built 5" of vermiculite back in 52, and extra insulation was never added until the early 80's so say the attic had roughly an R15 if you count a couple for the celotex plaster board originally.  I know gas was cheap then, but it says a lot for the power of the ceiling radiant, and what back losses there actually is.  When I have checked temps in the attic between the back of ceiling, and vermiculite, and between the top of vermiculite, and fiberglass. And once again on top of the fiberglass.  The largest delta t was actually between the panel itself and the back side of it.



 So an instance would be interior side of panel 85*. back of ceiling 55* top of vermiculite layer 45* top of fiberglass 35*. attic temp. 25* outside temp 15*. I was kind of expecting to see the temp of the back side of the ceiling to be more like 70-75. The celotex plaster board is 3/4 inch it must be a very good insulator because that is the thermal break between the pipes, and the attic itself.



 Thoughts??





 Gordy

Mark Eatherton

The term radiant panel...

is a universal term coined by the RPA to include any surface that is intentionally designed/configured or engineered to emit beneficial heating and cooling.



And I agree with you regarding heat flow. If the floor is hotter than the room, then the floor technically has no conductive downward heat loss, other than from the active radiant panel. It's kind of like radiant windows. If the window is warmer than the living space, then there is NO conductive heat loss from the room through the window, other than the active windows back loss. But, as engineers are wont to do, they include it in their heat loss factors. Kind of a CYA factor...



The RPA's old general rule of thumb is 2 X the R value below the floor as you have above it between the heat source and the emitting panel..



ME

Mark Eatherton

Just reflecting what I've heard the Siggmeister say...

Made sense to me, and I'm sure HE could tell you how he determined that number.



I intend to put as much insulation into my attic cavity as I can, but for the mean time, its running just fine on 3-1/2" of high density FG insulation. I've looked at the roof at night with the IR imager, and didn't see any indications of excessive heat coming off the roof.



I saw an old DIY radiant ceiling that a plumber had put together. Kind of like staple up, except it was bare 1/2" copper pipe, laid against the sheet rock in the attic, 8" O.C., with heavy duty aluminum foil pressed tightly over the tube, then 6" of rock wool insulation placed over that. It required a higher fluid temperature (170 to 180 @ design) to overcome all the conductive deficiencies, but it still worked like a champ in delivering good comfort.



The theoretics we see in the lab are usually counter to what we see in reality in the field.



ME

jp_2

clarification

thanks for the clarification, now i understand "panel".



more later......

jp_2

common problem ME

its easy to assume that the physics change when you get away from the lab cause things appear different.  but thats not really true, I tried to explain this to jim davis once.



take a VW rabbit, attach a 5th wheel trailer, you could drive it around

the neighborhood, but try to get it up to highway speeds???   you've

pushed the limits of the VW.   this also can happen with theories.



whats really happening is that you've pushed through the boundaries of the theory.  generally in the field you have added so many variable and conditions that you end up having to make wild assumptions, which end up being incorrect  and the theory apparently falls apart.   when really other conditions have larger effects than assumed.



the hard part is realizing when other conditions are impacting.  you've grown up in a different field, seminars teach you exactly how to do or set up things.  universities teaches just the opposite.   you learn generalities and hopefully teach yourself how to applied these in specific areas, or on the job learning.



I had a 2 yr electronics degree when i started work.    I was given a electrical wiring diagram for machines.   I couldn't make heads or tails of it??????   though I knew Boolean logic I had no idea what relay logic was.   after a 1/2hr talk with the plant electrician I was good to go.   I needed to know what the symbols meant, after that  everything else made sense.   that's the usefulness of general theories.



if you ever run across a research scientist that has experience in developing equations, take them to dinner and have them explain the oddities of developing theories and equations. 



I learned theories and were tested on them in chemistry only to find out the next year they were not true, " but it is a great learning tool".   lot of times theories are there 'just to get your feet wet'.  the real theory would be impossible to understand otherwise.

jp_2

back to slab loss

if the house is 30x40, room temp 70F, slab bottom more close to 80F, 0F outside.



rpa says R10, at 6 inches down and 1 foot in on the perimeter, sqft comes out to 225 sq ft. at 8 btu's/sq ft. or 43,200 btu/day at design.  why is this accepted standard?



R15 comes out to 28,800 btu's/day at design.  adding 1 more inch or 4 inches of faom only decreases by about 8,000 btu/day.



adding 1 more inch  of foam makes sense doesn't it?  odd that local code calls for more than the rpa?



these are just quick penciled figures mine you.

Gordy

Whats readily available JP

 Its pretty easy to get 2"  XPS R10 foam at any big box, or lumber yard JP.  I also think if you jump up to 3" R15 of foam there is a bigger jump in price per inch. You could layer 1" over 2" to get you there, But then thats twice the labor. If you are going to lay foam do it once.



  There is a point of diminishing returns on insulation when you reach a certain R Value. The insulation ROI ends up being more than its really saving in fuel. How much more will a floor really call going  from an R10 to a R15 detail. Of course that is a depends based on soil temps. In reality at todays prices you may only be saving 200 bucks or even quite less a season depending on design conditions, and area of slab.



 Keep in mind I'm not condoning the use of less insulation, only speculating why more is not specified. God knows what fuel prices will be tomorrow, and when concrete is poured its a done deal.

jp_2

ah come on gordy.......

are we really going to limit ourselves to materials from big box stores and tools from walmart?



where would you put the point of diminishing returns on the example I shown?



and how does this relate 20 yrs down the road in energy costs?

jp_2

hey ME

when you looked at your roof with the IR gun, did you take into consideration the venting?    another words, was the venting working good at venting the heat going through 3 1/2 inches of insulation?



usually venting help reduce ice dams of course, or lowering roof temps even though too much heat is being lost.



added:  if the heat loss increases, and then the air velocity in the roof vents increase to compensate, you would never see an increase in roof surface temps even though your vent btu loss has increase?  how about that??

Gordy

JP

 As I noted in my post I'm not against insulation. You are wondering why the RPA guidelines are not requiring more.  I'm just throwing possible reasoning out there materials easily attainable is one. But do also note those are minimum guidelines not an absolute this is all you need.





 Walmart??





 Point of diminishing returns to most homeowners are how long they plan to stay, and let the next person deal with it.  average is 8-12 years. Not saying that is the right frame of mind, but some people want radiant cheap, and easy, and do not fully understand the consequenses of their corner cutting. Its up to the codes to make the commitment.





 My point on diminishing returns for insulation is quite simply this . Just because 2 or 3 inches is sufficient. Adding 6" is not going to be 2 or 3 times better. Or reflect 2 or 3 times more fuel savings as a result of the extra added insulation.



 Sorry for the confusion JP I like the debate.

jp_2

diminishing returns

Gordy,



I see the diminishing value of insulation to heat loss, not cost.



graph insulation thickness vs btu loss, its all real clear. its a curve that goes flat.





added:



so I can still ask why the slab loss R10 is so low?  so who decided this minimum  is acceptable?  lot o people just accept minimum standard as "this is how much you need to use",   and not "this will work but you should really use more".

Gordy

I agree

 With your concern on wording of minimal standards.

Mark Eatherton

Yes...

My roof is ventilated via strips of screening under the overhanging soffits, and a full length ridge vent on top, so yes, I could see EVERYTHING. Not emperical, but definitely indicative.



It was below zero outside, and the heating system was running at the time of my viewing.



ME

jp_2

air cooling/ air temps

but did you understand my possible point?



couldn't the ridge vent temp be low with low air flow = low btu loss



then, vent temp low, higher air low = higher btu loss?



suggesting that heatloss is being masked by increase air flow through the vent?



thats what its suppose to do anyway.



you post suggested that maybe 3 1/2 inches of insulation is sufficient based on IR view of roof? 



just wondering.........



added: is that a steel roof or shingle?

Mark Eatherton

I need more insulation...

and intend to HEAP it into the attic area as time and money allow. 3.5" HD FG is what was there to begin with (summer use only cabin). Now that it is a year around proposition, I obviously need a LOT more insulation, that plus the fact that the ceiling is the primary heat source dictates so.



I absolutely KNOW there is heat loss occurring in the attic due to the RCH system, but am not sure to the degree it is occurring, or if it is to the degree one would think it is.



As for air cooling, I would agree that it is probably a major factor, but I do not have the instrumentation necessary (flux sensors, recorders etc) to prove it one way or another.



However, I do have one flux sensor, and in testing walls, determined through the use of the flux sensor, that the wall was NOT losing as much energy per square foot as one would think based on the theoretical U values of the wall assembly, but again, somewhat annectdotal information here...



Lastly, it is a tan colored standing seam metal roof, FWIW.



ME

jp_2

So if a person wants efficiency, what do you

recommend?



if they are building with 6-8in walls, want an efficient system, and ask you to recommend perimeter insulation amounts, what do you say?  R10?



I have read lots of posts on cold radiant slabs.........................



I agree infiltration can be a big problem. 

Mark Eatherton

I recommend...

the stated minimum code requirements, and recommend that they consider doing as much as their budget will allow.



You can pay for insulation once, but you will pay for energy again, and again and forever.



Prior to my and an associate getting involved 6 or 7 years ago, there was NO insulation required below radiant slabs or floors.



So, it is a HECK of a lot better than it USE to be...



ME

jp_2

flux sensors

I see those are pretty small devices.    so, do you do 20 experiments on 20 different  locations on the wall and do an average?



what is the expected error.   remember to, those heatloss equations are simpler forms of pretty complicated stuff.   did you ever read up on "significant figures" as i suggested long ago?  that will help you understand error in measurements.  fancy tool can give you fancy errors.

jp_2

The Chicken OR the Egg ?

sounds like the which came first, chicken or the egg debate.



on one hand, you need to know the temperature at the bottom(supply temp) of the slab to understand your true losses.



on the other hand, you need to know the slab loss and the structure loss to determine what the supply needs to be to

satify the structure.



as far as perimeter insulation, at least, it should be treated as a 'wall' and use the minimum code R value for walls.



on the other hand, a better approach is use the same BTU loss as required by the walls, but use the true slab temp

which is higher than the 70F used for inside temp.   therefore the R value will be higher than required for the walls.



so, if you keep using 70F to determine your losses, well, you'll be lost.

Mark Eatherton

From our friend Robert Bean...

http://www.healthyheating.com/Radiant_heating_designs/insulating-underslabs.htm



While you are there, check out all the other GREAT stuff the Bean has assembled on this wonderfully educating web site...



ME

jp_2

Nice stuff!

lot a data in those pictures, after I study them for a couple of hours!