in slab radiant

justjoedip

I am adding in slab radiant to my current boiler radiant system. I have a few questions:



1)I understand i want to keep loops to 300'(1/2" pex O2 barrier)' or less at 250' what is a good pump size for 25X25X10 area well insulated? Spray foam insulated all around 6-8"thick.



2)I have a dunkirk QSM boiler w/outdoor reset heating via old cast iron(lower Delta T) It will be heating the house via this method. The Pex is for garage & basement slab and second floor of garage, I expect to need some mixing valves, can I use the return water loop as cold water supply to mixing valve? Should I "UPSIZE" the mixing valve?



3) I am planning on 3 loops(roughly 250' each loop) in the 25' X 25' slab spaced roughly 10" apart.does this seem reasonable? (Enough btu's to keep garage above 55 Deg F?) I know some of this answer is in question #1 flow rate and "head" of the pump



4) I am going to use "warmboard" insulated underneath for the floors that have an open area(heated) under them. Any comments?

5) I currently have a supply & return header supplying for 4 zones with an air scoop. do i need seperate air scoops for the pex zones?



Thanks,for your time and effort,Joe

Rich_49

HeatLoss Calcs

been performed for this home ?  Does not matter that system is existing , if you're thinking radiant that would entail that you want it comfortable and right .



1 .   After having heat loss done you will know how many loops you require and how many feet they should be . Then you can determine what pump to employ . Your Delta T should be 10* in a floor usually . Some may disagree .



2.    You should certainly use the return water for cold side of mixing valve .  Is your existing cast iron baseboard or rads ?  Do not upsize the mixing valve , the samller the better as long as it will flow your required GPM . The less fluid the faster they can pinpoint your desired temp . Kinda like hunting in a narrow field with lotsa game as opposed to a wide one with the same amount .



3.  12" spacing or even greater will probably suffice . Again the heat loss will give this information .



4.   Warmboard is a good product . This should also be run at a 10* Delta T .  Is your air scoop an air scoop of old ?  If so I would get rid of it and use a more modern air eliminator , configure it so your make up water enters there and your expansion tank also . No need for more than one elimination point usually , if piped properly , however without seeing actual system I have seen some that could use a bit more in certain areas .



     Do The Heat Loss .

Jean-David Beyer

Your Delta T should be 10* in a floor usually . Some may disagree .

I do not know if I should disagree or not. And it is not because I think the delta-T should be 15F or 5F or 20F. It is because I do not think one can reasonably assign such numbers.



For example, when it is 0F outside, my boiler puts 120F water into the floor, and the delta T would probably be around 10F. Since it never gets that cold around here, I do not have an actual measurement.



When it is 14F around here, the boiler puts 112F water into the floor, and I would guess the delta T would be, perhaps 5F.



When it is over 50F around here, the boiler puts 76F water into the floor and the delta T is between 0F and 1F.



These supply temperatures are what are required to keep the house at 69F inside. So in some sense, those supply temperatures are correct. And the delta T values are, per force, correct, since that is what I get. If I observed greater delta T values, it would mean the system was delivering more heat than needed. In that case, I would lower the supply temperatures and the delta T values would go down again.



It seems to me that once the system is installed (and except for the boiler, it was installed over 60 years ago), I will not be changing the amount of tubing in the downstairs slab, nor will I be changing the amount of baseboard upstairs. So the delta T I get may be interesting to observe, but it is not one of the things I can control.



I know I could change the delta T by changing the circulators, but to increase the delta T by slowing down the circulators would just make the heat distribution more uneven, and I see little benefit in doing that.

Zman

The question

Joe,

 1.  Shorter is better on the loops. I my mind 200' should be the target 250' the max. Keep the loops within 10% of each other. This will keep your head loss within the range of all the affordable circulators.This also keeps the delta t low and provides a nice evenly heated floor. Everyone will have a favorite, the grundfos 15-58 works well for this application.



2. The mixing is a tricky question. You could use "dumb" mixing valves and just let them "lag" the ODR setting your boiler is firing. It would be better to give them individual "smart" control. You could use either an injection mixing setup or a mixing valve with a controller. They both have pros and cons and I would let the tie in with the existing system dictate. A smart mixing setup would also allow you to prevent cold water from returning to the boiler when the garage zone is running. If this happens it will cause the boiler to condense, this will shorten it's life considerably.



3. Approx 10" spacing would be plenty for a garage.Again shorter loops.



4. Warmboard is a good idea. You will not have much flexibility on tubing layout so it would be good to have good control over the water temp in order to "tweak" the output.



5.You do not need multiple air scoops. Be sure your existing one is in the optimal position in your system. If the expansion is attached to it, the circulators should be "pumping away".





Carl

Rich_49

So almost

no heat is being transferred to the room and you are moving 20 GPM per 10000 BTUh when it is above 50 ?  You must have a big buffer tank . Most folks only require 2 GPm to actually heat the space . I would surely suggest changing that circulator to an ECM if for nothing else but the electrical savings , does your home sound like a model train is running through the floor ?  I wonder .

Zman

Delta T

In real operating conditions, delta t will vary tremendously. Tubing in thick slabs will have higher variations. Low mass applications will be more stable, as Jean-David said they will be tighter as the water temp approaches the indoor air temp.

Delta t circulators help stabilize this somewhat but nothing will completely stabilize it.

I guess I don't see a problem with these fluctuations. If your slab starts getting too wide a range it will create hot and cold spots this can be avoided with proper circ sizing and shorter tubing loops.

Why not design the radiant for a cold day design of 10 degree delta t and figure it will usually be a bit tighter. You have to pick a number  as a basis for your design ,10 for floors and 20 for convectors seems to work pretty well.

Carl

Jean-David Beyer

almost no heat is being transferred

It cannot be very much. When it is 0F outside, and it never is because design day is 14F, I need at most 24,000 BTU/hour. So when it is 50F or more outside I need very little. So I can supply 76F water and get enough heat. I calculated the flow, based on lots of unverifiable assumptions, and conclude that my 007-IFC circulator pumps about 3 gpm or a little less through that slab. Unverifiable because I do not know the spacing and lengths of the tubing. I also do not know if the slab is insulated, but I very much doubt it. Built in 1950 when heating oil was probably around $0.20/gallon, so nobody cared.



No buffer tank. The boiler holds about 3 quarts of water. You might consider the mass of the water and the mass of the concrete slab to be a buffer. I mean if I change the thermostat setting, it takes over 4 hours no notice any temperature change, and it takes about 24 hours for the system to stabilize at the new temperature.



There is no sound whatever when the slab zone is in operation. I have to go out in the garage, where the boiler is, to tell if it is running downstairs or not.



In the upstairs baseboard zone, I can tell if it is running because the baseboards get hot enough to tell. They run between 110F and 135F depending on outside temperature. (I have a lot of baseboard for a zone that needs hardly any heat). Also, if I put my ear near a baseboard, I can barely hear it. I am not sure if what I hear is the sound of the circulator being conducted to the rooms, or the sound of the water in the pipes. But if I just stand in the room normally, I do not hear it in the baseboard either.



The house is only 1250 square foot Cape Cod, with new gas filled Marvin Windows and a lot of insulation. It does not need much heat, so I put in low temperatures to the emitters, and since they are close to room temperature, they do not emit much, so they do not have a large temperature drop.



I agree that ECM circulators would save some electricity. I have not calculated it out, but for me, three circulators at $500 each (installed price) would probably take decades to pay off, and I will be dead in less than 30 years, I expect. If a circulator fails, I might replace it with an ECM, but it would not be a delta-T or a delta-P; I just do not need anything other than a plain circulator.



I could rip out both heating zone circulators and use one delta-P circulator and two zone valves. But that starts running up the cost still more. The present circulators take less than my computer takes, so what's the point?

Rich_49

How many watts

does your computer use ? a Taco 007F5IFC consumes right around 80 watts while operating .  You state your design temp is 14*F and in another thread you stated that you were effected by Sandy so I will guess that you are somewhere in my geographical area . The average size garage with the best garage door one can purchase with the best insulation you can install in a 2x4 wall with no windows has a requirement of 4,736 BTUh on a 14*F day and requires 82* AWT flowing at 1 GPM to maintain a surface temp of 71.2 and a room temp of 69* .   Let's hypothesize that you are flowing 3 GPM at design , your Delta T would be 3*F , on a design day . This sir is not a designed system at all .   When you design a system you must Choose a design Delta T , most manufacturers of Boilers , tubing , radiant heating products , heat emitters of all sorts agree on an industry standard 10* , 20* , there are a few product specific variants .  Warmboard which the OP inquired about in fact recommends a 10*Delta T be run through their product to perform at stated outputs . Most manufacturers of tubing used in radiant heat applications concur with the 10* DeltaT through a slab a radiant floor with a few 20* product specific recommendations . These are facts , not assumptions , not opinions .  They are all based on a formule known to heating system designers , unfortunately , whomever designed the boiler , DHW and controls of your system was not aware of this formula or just did not care enough to employ it .

Since you are pumping 3 GPM with a Taco 007 your head loss must be 9.5' - 9.75' head loss , this conclusion is arrived at by observing the pump curves as per Taco as of 09/09/09 .  Your system that requires around 24,000 BTUh as per you must short cycle quite a bit at any temp above design considering your narrow Delta T's , this also is a fact , if it does not then sir the laws of physics ceased to exist at your front door .  Many of us install new systems in buildings that are existing , some of us perform a heat loss before installing the equipment and less of us take the time to tailor a system to the needs of the home taking into account the existing emitters that will be our distribution system . 

Your boiler is not working to it's potential , your emitters are not transferring enough heat to your space and I can almost with certainty say that one of the following is true , A.  see the bottom of all my posts , B. You researched how heating systems work on the internet and heard more opinions and hypotheses than fact and this was verified by a plumber who did not really know and listened to what you said and is now spouting that off as gospel on some site .

Live and learn I suppose

Jean-David Beyer

This sir is not a designed system at all .

I suspect my heating system was not designed at all. The house and its heating system was built in 1950, so if it was designed at all, it would be before that. In particular, I do not know where the tubing is in my slab. Not the spacing, not the lengths. I do not know if the slab is insulated (I very much doubt it). The upstairs was heated by finned tube baseboard: three feet in each room. The entire system was one zone run at a single temperature. The only way to get more heat upstairs was to close a valve to the slab somewhat, but when enough heat went upstairs, downstairs was too cold.  And so on.



I had three computers in here. One took about 600 watts, one takes about 200 watts, and my newest one (more power efficient) is taking 198 watts at the moment. My Taco 007 IFC circulators, nominally 1/25 horsepower are not electrically efficient and take about 90 watts each when running. My main computer runs 24/7 and the other one runs 24/7 8 months a year and is off in the summertime. With my heating system, at most three circulators run at a time. Sometimes none, sometimes one (heating the indirect), usually two (boiler circulator and slab circulator), and sometimes three (if both heating zones are running). About 4 years ago, I replaced the 60 year old oil fired GE boiler with a gas fired mod-con, and replaced the three foot long baseboards upstairs with 14 foot long ones so I could run lower supply temperatures up there. Actually, they were about the same as before, but they put out enough heat because the emitters were 14/3 larger. I also had it set up as two zones: one upstairs (baseboard) and one downstairs (radiant slab).



Why are you talking about garage doors, garage size, etc.? My boiler is in the garage, but except for waste heat, the garage is unheated.



Likewise,I have no idea what water temperature my house was designed to run at. I.e., supply temperature. I am pretty sure they did not design it to run 180F in the slab. I do not know the actual temperature the former owner used. There was a thermometer on the boiler, and I ran it from 130F to 140F. This was a little too hot for the floor and the asphalt floor tiles would not stay down, and the overshoot was way too much. But at that temperature, the upstairs was too cold. I did set the spread to 10F and it cycled 12 times an hour most of the time. The former owner had the spread way less and it went on and off about every minute. If I made the spread wider, the relief valve would weep because the old style expansion tank was not big enough.



Similarly, I do not know if they designed it for a particular delta T. If so, that too is secret. In any case, they got the upstairs wrong. I could not get enough heat up there and keep the tiles downstairs glued down. Period. I do know that whatever delta T I get now, and it varies greatly with outdoor temperature, it is quite low. 0F to 1F when it is over 50F outside, and when it gets down to about 10F outside, the delta T in the slab goes up near 10F. But it gets that cold outside for only a few hours once or twice a year. This year, it never went that cold outside.



Upstairs, I get greater delta T than downstairs, but it is still low. Due, in both cases, to the low supply temperatures. When the supply temperature is only a few degrees hotter than the desired room temperature, you do not get much delta T. And there is no point raising the supply temperatures to get a greater delta T, because the higher the supply temperature, the higher the return and the higher the return, the worse a mod-con works. And slowing down the circulators to get a higher delta T results in uneven heat distribution. Upstairs baseboards are in series, and downstairs in the slab, I do not know where the pipes go in much detail. But different parts of the slab in each room vary in temperature, so I figure if I slowed down the circulator to the downstairs, the unevenness would get even worse.



When I was figuring out how much baseboard to put in upstairs to ensure the return temperature water would never go above 130F (wanted condensing all the time), I did not pick a delta T. I knew the heat load, so I knew how many BTU/jhr I needed on design day. I had some idea how much baseboard I had room for. From that I knew what supply temperature I would need. And that had some delta T or other. True, I could have picked a delta of 10 on design day, and a supply temperature, and calculated how much baseboard it would take. And then tried to fit it into the room. But I did not think it made sense to design for a delta T. My requirements were the amount of heat (BTU/hr) and return temperature, and no more than 14 feet of baseboard. And take the delta T that I got. What do I care what the delta T is? I wanted enough heat up there and a return temperature of less than 130F under all conditions, preferably a lot less.



Do you not really do the same? Is delta T more important than comfort and efficiency? Seems nuts to me.



Discussing how to design a heating system, etc., is surely important when designing a house. But this house is how it is, and I do not have any idea what design considerations were considered. My guess is that the contractor who build it did not ever install hot water systems before. GE had only recently started making oil fired home heating boilers, and he got the smallest one. That was more than the house needed, as I figured out later. And as far as 10F, 20F, etc. Those numbers were decided, in part, because it made the calculations easier and not because of any particular physical law of nature that mandated 10 F rather than 9F or 11F or 18F or 20F or 22F. There is no magic to those numbers.



So maybe the contractor who build this house should have used warmboard on top of the slab instead of copper tubing inside the slab. But there was no warmboard then. Maybe he should have used two zones instead of one. Perhaps he should have used radiant upstairs too, but he did not.



Should I have gutted the house and started over? If someone else was paying the bills, I might have considered it, but lacking a rich uncle on his death bed, that was out of the question. Replacing the boiler, putting in more baseboard for lower supply and return temperatures, and two zones instead of one were all my budget could manage.

Rich_49

Comfort and Efficiency

are all that matter as the end result .  My point is the wider the Delta T the more heat  that is being transferred into the space .  The most efficient boiler is the boiler that is not running . So you are the occupant of a home that has a poorly (if at all) designed heating system , not all that uncommon , welcome to the ranks .  With any luck your radiant tubing is in a nice layer of sand and not right in the concrete like so many ,

One can assure return temps without compromising heat transferred into the space , running a system with no Delta T is absolutely nuts . You don't need a rich uncle to finance your project or little tweaks here and there . 500.00 installed for an ECM is a bit high in my book , I sell comfort and efficiency on a daily basis and know what these things cost to install and believe me 500.00 a piece is  too high to change out a circ for a local contractor .

 You would not hurt your efficiency at the boiler by raising the SWT 5* and using a circulator that more closely matches your load requirements and getting a wider Delta T , more heat is being transferred then and your boiler will actually gain efficiency along with running less . You need heat transfer period . The less heat that is being removed from the delivery side of the system the less efficient it is . You cannot nor can anyone argue this simple fact . Less cycles of the boiler whether they are short or nice efficient 10 minute ones is a good thing . 10* , 18* , 20* , 36* , 40* Delta T's are not some magic numbers and they are not just mathematically easier as a choice , they happen to be the numbers that have proven for these emitters to be the most efficient  and getting the most heat out of them at any outdoor condition .

Jean-David Beyer

getting a wider Delta T , more heat is being transferred then and your boiler will actually gain efficiency along with running less .

A month or more ago, we had two threads that seemed to run forever on this point, and I disagree with almost all of it. I do not want to go through that ever again, and I imagine those who participated in that, or those that endured reading it, do not want to rehash all that ever again.



I insist that running a high enough flow rate so that there is almost no temperature drop along the emitter is the way to transfer the most heat. Anything less than that will have the emitter cooler at one end, so the average temperature in the emitter will be less, so the heat transferred will be less too. Now transferring the most heat may not be the most important objective, but it surely is one of them.



This is not to say that one should actually try to run it at 10 feet per second or something. I think something high enough to get a degree or two drop over the length of the emitter on a very warm day, where the supply only needs to be a little over the room temperature is just fine. Of course it needs to be high enough so as to transfer enough heat on design day too. But since, with outdoor reset, the supply temperature goes up when the outside temperature goes down, it may not be necessary to vary the pumping rate at all.



I agree that few cycles are better than many. For my radiant zone, the supply temperature is so low, even on design day or a little colder, that my cycles can last over 12 hours. I.e., on for twelve  to 18 hours, off for 12 hours, etc. It seems to me that that could be considered efficient. Only when it is very warm outside do the cycles shorten because the boiler will not modulate down far enough.



As far as tweaking being inexpensive, I disagree, but it is difficult without naming prices for parts and labor, and we are not supposed to discuss pricing here.  Since I am a homeowner, I must pay list price for circulators, and an hourly rate for the labor. I do not choose to buy the circulators on the Internet and expect to find a cheap contractor to install them. Besides, what I could do in two to three hours, an experienced contractor can do in 15 minutes and another few minutes to purge the system. I know what it cost me to have my contractor swap out a noisy 007-IFC, and the number I gave assumed an ECM circulator would cost about double that. And why should I do that? To get more delta T, I would have to slow the flow down. And if I got more delta T, the second baseboard room would be too cold. If I raised the reset curve, the efficiency of the boiler would go down and one room would be colder than the other, or I would have to add baseboard in the cold room to make up for it. That would mean running baseboard along three walls instead of just one. Just to get higher delta T? And for the slab, I would have even more expense to even out the heat. No wall space for panel radiators. For ceiling radiant panel, I would have to rip up the ceilings of all the downstairs rooms, or the floors in the upstairs rooms to run the pipes. That would be ridiculous.

Rich_49

Enjoy

your heating system Jean-David .

Eastman

summer is pretty slow for heatinghelp

I am happy to rehash anything controversial, Rich.  I keep trying to think of other issues to discuss on heatinghelp.com but Delta T arguments seem to be the best.  I suggest optimal delta T for thermal losses below minumum modulation.  Anyone interested?

SWEI

Zero ∆T

equals zero BTUs.  This is not an opinion.

Eastman

how do you create that

delta symbol?

Jean-David Beyer

Zero ∆T equals zero BTUs.

Of course this is technically correct. But my supply and return thermometers read to only one degree. And when supply and return indicate the same temperature, I call that a delta T of 0. Assuming they read off as much as possiblle in each direction, the actual delta T could be just short of 2F. And that would deliver some heat. I am sure that if it is 50F or a little more outside, for several days, and the house is 69F that there really is a positive delta T, and that heating is taking place. After all, the boiler is firing (at minimum firing rate); that heat must be going somewhere, and the house temperature is not dropping. I have not calculated it out, since I do not know how much tubing is in the slab and do not know its spacing either, so I cannot calculate the exact temperature drop there must be. It would be nice if I could buy some 4-inch thermometers to put in there, but I do not have the money and there is really no place to put them anyway (crowded where boiler is installed).



So If I put 76F water into my slab and it says it comes back at 76F, perhaps the supply is really 77F and the return is really 75F. And my room temperature is 69F according to my Honeywell digital thermostat that reads really funny (it rounds off towards the set point, so perhaps it is 68F or 70F in there) and my cheap IR thermometer says the floor is 72F in some spots and 76F in others, ...  . So sure, it is putting out some heat. But the more thermometers you have, the more readings you get.



I even have a very expensive mercury thermometer calibrated against a reference thermometer calibrated against one at the National Bureau of Standards (as it was called way back when that thermometer was made, perhaps back in the 1960s). That is probably the most accurate, but it measures the temperature of the air or water around it. And it reads differently too. I have a water mixing valve that claims to hold +|- 1/2F if the flow exceeds 1/2 gallon per minute. I put a bunch of thermometers in a large beaker and run the water for 15 minutes or so, and they all read differently. Temperature measurement, if you want to be really accurate, is very tricky.



For home heating purposes, I set the thermostats to 69F, and I do not really care what the temperatures are. When I do that, I like the temperatures, and that is what counts. I look at what the boiler says it is supplying, and what it says is returning. I subtract the return temperature the boiler indicates from the supply temperature the boiler indicates (on its digital readout), and I say the difference is the delta T.  If the errors were the other way, I could probably get an indication that the return is hotter than the supply by a degree or so. I do not, but that is just luck.

Zman

Sorry Joe

Joe,

So sorry your post turned into this ridiculous debate. Please post again if you have more questions.



Jean-David if you have a delta of anywhere near zero, your curve is set too low for your install. If you are just interested in wasting electricity only to gain a little heat, why not just leave a light bulb turned on? You could even put it on a t-stat. The end result will be the same, you just won't wearing out expensive boiler components. Highjacking threads for this kind of nonsense is rude and counterproductive .



JMO,



Carl

Gordy

Agreed

Carl



This whole delta t thing is running away.12 years here, and never seen it so picked apart ridiculous .

Jean-David Beyer

if you have a delta of anywhere near zero, your curve is set too low for your install.

Could you explain why that is so?



If I am getting the heat required with my curve is as it is, why should I raise it? Raising the curve would make the boiler run hotter (lowering its efficiency through the heat exchanger), causing more heat loss in the distribution piping, and less condensing as the water came back from the emitters at a higher temperature. Also, running it hotter would result in more rapid cycling of the boiler, and that, too, would seem to lower the efficiency.



It still seems to me that running with the lowest possible temperatures that will deliver the heat required is the way to go with a mod-con. Why is this not so?

Eastman

perhaps a new thread is best

then Joe can keep this one on his project.

SWEI

∆ symbol

On the Mac it's [option] + J



Windows depends on your locale settings (and somewhat on Windows version, whether it's Unicode or ANSI) http://en.wikipedia.org/wiki/Delta_%28letter%29

Eastman

∆∆∆∆∆∆∆∆

haha  y∆∆∆hh

justjoedip

Heat exchange debate

Well,fellas thanks for the info. As I work in the industrial heating industry and i believe if the stack (return flow)is higher then your process goal you are wasting energy.



I was just wondering how many btus I should expect the slab to transfer reasonably and not waste energy I have a condensing boiler right now and I know it is not very efficient because I have big old cast iron rads with a very low delta t compared to as I read I can expect with in floor heating. I know the flow rate and temp difference (room-water) will dictate the amout of energy I can release or supply to the room. This will be a very well insulated install 6-8" of spray foam insulation on all sides few windows.



Thanks,Joe.



Ps anyone in the Pgh,Pa area willing to provide some help for a reasonable fee??

Please MSG me justjoedip@Gmail.com

SWEI

Minimal difference betwen stack heat and process heat

implies excellent performance, just as minimal difference between room air temp and average emitter water temp delivers near-perfect steady-state comfort in a radiant distribution system..

justjoedip

Explanation

We design for stack at or below process temp. I have some industrial furnaces that actually have a lower stack temp then the process we are applying the energy into. The process does have a large heat capacity though so it is not as difficult as the physics theory says it will be.



Joe.

Zman

Condensing

Joe,

Check out this  www.pugetsoundashrae.org/PDF_files/AshraeCondensingtechnology.pp



The trick with your system is to get the return water below the dewpoint. Your cast iron rads may perform better than you think at the lower temps. The simplest piping design would have the entire system designed for the same boiler temp curve. You will get pretty good efficiency .If  the coldest design day can have return temps less than 130.

Your garage floor would be a bit overheated at that temp, the warm board system would depend on the design. A simple non-thermostatic mixer would allow you to have a zone temp that simply "lags" the boiler ODR curve.

I did not realize that you had a condensing boiler. That changes things.



Carl

justjoedip

Better heat

Yep, I bought the best boiler I could at the time but did not have (read didn't want to delay moving in) longer to install the pex under the first floor. I'm now kinds regretting that but "if " I getter comfortable with the pex on this install I may do the upgrade. I can access the floor from the basement and don't think it will be that difficult to install the aluminum plates and pex. I just have to figure out "if" the gain is worth the effort. It is a 110yr old house

Thank for all your comments.



Joe