Sizing Radiators

Wayco Wayne

where they want cast iron radiators. The Burnham book says to size using 240 btuh per square foot. My handy EDR chart on my mouse pad says that is for 215 degree water. For 180 degree water the btuh per sq. ft. is 170. Do you size for the 240 btu and turn up the water temp if necesary or do I price in more sections to match my design temp of 180 degrees. These cast iropn sections are not inexpensive.

<A HREF="http://www.heatinghelp.com/getListed.cfm?id=255&Step=30">To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Contractor"</A>

David Efflandt

240 is for steam

240 btu/sq ft is typical steam output which uses latent heat of vaporization to transfer heat when it condenses, with little drop in temperature.

If you are using water, it would have to be hotter than 215 degrees at the boiler to give you the same output. This would likely adversely affect the efficiency of a hydronic system.

Mike T., Swampeast MO

Sizing iron

The 240 btu/hr is DEFINITELY for steam!

Also remember it's the AVERAGE water temperature in the radiator that you size for--generally assumed to be about 10° less than the supply temp (20° delta-t).

The best way to look at iron radiator output is in btu/hr/sqft PER DEGREE OF TEMPERATURE DIFFERENCE BETWEEN IRON AND THE AIR SURROUNDING.

If you want to skip the below either use 170 btu/hr/sqft @ 180° AVERAGE water (in a 70° room) or use 1.5° per square foot EDR per degree of temperature difference between iron and air. (This does result in 165 btu/hr/sqft output at above conditions however.)

There used to be LOTS of study of the output of cast iron radiators and not surprisingly they found that radiators of different designs put out different amounts of heat per square foot in otherwise controlled conditions. They also found that as the temperature difference between the air and the iron decreased that btu/hr/sqft/degree decreased as well. This dropoff is NOT linear--being greatest at temperatures just below that of steam--typically 215°.

At "typical" water temperature use of 180° average they found an output of about 1.515 btu/hr/sqft/degree of difference or 166.65 btu/hr/sqft in a 70° room. Unfortunately I've yet to find old tables that go down to the REALLY low temperatures encountered when dead men water systems are put under proportional control.

This was for their "standard" radiators--two COLUMN and about 38" high. Compared to this standard, additional columns reduced output but the 3rd reduced it more than the 4th. Low, multi-column "window radiators" however were found to produce greater output than the standard--up to 20% greater depending on the types being compared.

Most of their testing was conducted with steam. Steam provided the most practical means of determining output because merely by knowing the pressure of steam delivered and the temperature and weight of condensate produced you can determine output simply, quite accurately and without having to truly measure the output of the radiator.

ALL iron radiators seem to have been tested this way and in later years and most notably with thin-fin-tube radiation EDR became arbitrary in terms of TRUE surface area. Everything was compared to the "standard" STEAM radiation through the steam and condensate measurement method.

Even when testing output at water temperatures, I believe steam was still typically used. While they may have produced steam under vacuum at lower temperatures they probably instead increased the temperature of the air surrounding the radiation.

As we know, laboratory testing chambers aren't a particularly good reflection of reality. Reality presents WAY too many variables ALL of which influence one another in incredibly complex ways. Radiator mrgrs (I presume the ones paying for testing) had one major goal--get the greatest amount of output from a given amount of iron. Typical testing chambers would find the radiator under a single-pane window (probably a well-fitting double-hung) with contrivances to increase infiltration through the window in an attempt to simulate wind.

Such was the "accepted" practice for radiator placement in real-world structures even before the testing chambers were created, I believe. But, this type of condition DID produce a very good situation for enhancing convective output from the radiator. The convection was further enhanced by the high temperature of steam on which all of the radiators were STILL compared.

They had convection "cooking" the point that "typical" air temp stratification in a room 9' high would be about 23°! And that's NOT the maximum with air temp at ceiling measured directly above the radiation.

Right now it is 27° outside, very bright sun through S window, wind rather gusty.

Air temp of room below my office is 55°--bare joists down there. Air temp as close as I could get to the floor without touching (about 1/2") is 62°. Air temp at the standing "breathing" level is 69°. Air temp VERY close to ceiling (9') is 72° in the center of the room, 71° in exposed (SW) corner opposite the radiator and 77° DIRECTLY above the radiator. Ceiling fan is NOT running and hasn't been running for a couple of days. Temp of TRVd radiator is 92°.

That's a 15° maximum floor-ceiling temp difference with 10° likely much more representative.

Temp of wall (interior partition) BEHIND the radiator is 77°. (Interesting that it's the same as the air temp at ceiling directly above the rad--while I've rounded these measurements both of these were 77.0°) Air temp at TRV remote sensor location (thermometer shielded from the radiation) is 64°.

Temp of wall (exposed West) OPPOSITE the radiator is 72.5°!!! (It's still 29° outside and the sun is NOT shining on that wall!) This is about 6' off the floor as big desk blocks most of that wall. Air temp at same height is 71.4°!!! I'm quite certain the little remote-sensing digital thermometer I'm using right now is quite accurate. It reads 32.1° in glass of ice water and 98.3° is my mouth. Am "shielding" the probe with a decent sized chunk of styrafoam to try to block readings from the temp of my fingers.

Temp of exposed South wall (sun shining on outside BRIGHTLY) at same height as W wall is 73.9°. This wall is also in "oblique view" of the radiator. Temp of S wall where it has a "dead-end view" of the rad is 72.4°. Eight inches straight down from this the wall is completely blocked from the radiator because of a filing cabinet. Wall temp there is 70.7° INTERIOR partition (E) completely blocked from radiator as well ALSO measures EXACTLY 70.7°!

Temperature of inside surface of west window glass (double-hung with well-fitting aluminum storms) is 58.1°. Temp on inside surface of storm window glass is 41.2°. (Taped the sensor with styrofoam around to the glass for these measurements.)

I've made similar sets of measurements at/around outdoor design temp. Temperature stratification rises a bit, but sincerely nothing like the dead men used to assume. Actually the radiator temp I just measured seems a bit high as my prediction at this outdoor temp is 88°. I had though left the door of a closet connecting to another room (58-60° in there) over night and TRV had been set back a bit as normal. Rad probably wasn't at "full maintenance" state. The outside temp is now rising quite rapidly (it only rose about 2° during all of those measurements).

Most interesting (to me) in those measurements is the temps of the walls. It certainly seems that radiation is playing a SIGNIFICANT role in heating this space!!! Yes, I know you can say that the warmer wall temperatures result in greater heat loss to the outdoors but remember that colder wall temperatures result in increased air temperature for the same degree of comfort!

These do not seem to be the sort of interior conditions that the dead men expected to produce with their iron radiators. A nice illustration in Mechanical Equipment of Buildings (1929) regarding air movement and temperatures in a room bears some interesting similarities AND differences. Despite a nearly identical air temp below the floor, the air temp just above the floor was significantly warmer in this case.

I rather suspect the dead men had GREAT difficulty creating "maintenance-type" conditions in a testing space. While proportional control WAS available, it was exceptionally expensive.

Studies really are under way (or have been conducted) using proportional control of heating devices at relatively low temperatures. All seem to be finding that the radiative portion of the output of such a device is more significant that once (even currently) believed.

---------------------------------------

Off on a tangent again. Sorry.

Back to sizing the iron radiation. I would sincerely use that 1.5 btu/hr/sqft/degree of temp difference for sizing. It agrees quite closely with "standards" both old and new. You could try to start splitting hairs based on the size, style and location of the radiation but systems with modern control (particularly proportional) aren't going to behave in a way that corresponds with the old testing methods.

If this is a home with reasonable insulation and reasonably well-controlled infiltration I most sincerely suggest tall iron radiators on INSIDE walls wherever possible and most particularly if you are using TRVs or a modulating boiler. The old objections to this placement (they said it ALWAYS had the advantage of loosing less heat through the outside wall) sincerely seem to have been eliminated. Formal and informal studies really are finding that this placement results in less temperature stratification, increased radiation, higher comfort, and possibly less fuel consumption--that is as long as you let the increased temp of the exposed walls do its job and keep the air temp a bit lower...

Mike T., Swampeast MO

Sun down now

Closet door closed all day, TRV untouched since this morning, radiator surface down to 87.3°. Outside temp now 30.0°, room temp up a touch at 70.5°.

My "predicted" radiator temperature at 30° outside and 70° inside is 88° for this room. I know my monitor, my body, and the cat are putting out some heat, but the doors have been opened and closed rather regularly and the spaces outside this room are at about 60° (except the radiantly heated bath that shares 5' of wall with my office currently at 69°).

Formula for predicting surface temperature of radiation (U.S. Capitol rads) uses the 1.5°/btuhr/sqft/degree of difference.

This is my most "studied" radiator even before my office was in this room. When left undisturbed and with no occupancy load and actual conditions in the rest of the home VERY close to those used for the heat loss calculation my surface temp predictions have sincerely been so close to that measured that it scares me.

It IS however quite possible that I have not sufficiently reduced the heat loss provided by HVAC-Calc. I used a 20% reduction factor.

Wayco Wayne

Hey Mike

I seem to have stepped into your wheelhouse. Give me an example of your sizing method. One of my rooms has a 6500 btuh heat loss. How many square feet of radiation would I need at what EDR? Show me the numbers so I can follow it and understand. Thanks for your expertese. WW

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Contractor"

Mike T., Swampeast MO

nm

Mike T., Swampeast MO

Sizing Iron

I suggest the 1.5 btu/hr/sqft/degree of temp difference method:

SIMPLE formula: AVG supply temp - air temp surrounding * 1.5 = btu/hr/sqft

Results in this table for ONE SQUARE FOOT of STANDING iron radiation.

190° = 180 btu/hr

180° = 165 btu/hr

170° = 150 btu/hr

160° = 135 btu/hr

150° = 120 btu/hr

140° = 105 btu/hr

130° = 90 btu/hr

120° = 75 btu/hr

110° = 60 btu/hr

100° = 45 btu/hr

90° = 30 btu/hr

-------------------------------------------------------

Now compute total square feet needed at desired AVERAGE water temperature.

Formula: TOTAL btu/hr loss / btu/hr output of one square foot @ x temp = total square feet required

For your sample space with 6,500 btu/hr loss:

6,500 / 180 = 36 sqft @ 190° AVG

6,500 / 165 = 39 sqft @ 180° AVG

6,500 / 150 = 43 sqft @ 170° AVG

6,500 / 135 = 48 sqft @ 160° AVG

6,500 / 120 = 54 sqft @ 150° AVG

6,500 / 105 = 62 sqft @ 140° AVG

6,500 / 90 = 72 sqft @ 130° AVG

6,500 / 75 = 87 sqft @ 120° AVG

---------------------------------------------

If a simple system under digital control, and using "standard" average temp of 180° (20° delta t = 190° supply temp) STOP!

Your radiation will be sufficient in size for any reasonable outside circumstance. Of course this assumes your heat loss calculation is reasonably thorough and that you have used the proper outdoor design temp for your area. As long as you follow good piping practice your system will be quite well-balanced as well.

----------------------------------------------

TO BE CONTINUED...

Sizing for proportional flow control, constant circulation, reset, boiler modulation, etc.

That's when it gets FUN!!!!!!!!!!!

Mike T., Swampeast MO

Dynamic Sizing of Standing Iron

When you put standing iron under proportional control you need to get a good idea of the ACTUAL surface temperature of the radiators in real-world operation.

Why? New standing iron is EXPENSIVE. Used radiators may not be of the exact size required. Existing systems may not be particularly well-balanced. A calculated reset curve will be more accurate and useful than a "rule of thumb" from the package instructions. Lower supply temperature means less transmission loss and allows you to get the most from condensing boilers. Lower supply temperatures generally equate with higher radiation, lower convection and higher comfort AT LOWER AIR TEMPERATURES. While proportional flow devices like TRVs are fully capable of "removing" sins so serious as to make a system essentially useless without them, such is NOT the intention and the life of the TRVs will be adversely affected. I could go on and on with this list...

------------------------------------

IMPORTANT ASSUMPTIONS:

Yes, I know the old saying that assumptions make an "****" of "u" and "me" but sometimes there is no choice. While there can certainly be debate regarding these assumptions, I believe that debate would hinge more on "why" than anything else.

Assumption 1) Proportional control is all about balance. The idea is to put in the same number of BTUs that are going out and to MAINTAIN this balance over time.

Assumption 2) Manual J calculations intentionally overstate heat loss. This is not necessarily a "bad" thing as it attempts to give a simple way to state heat loss in a world that is anything but simple. Manual J calculations are strictly linear for a given structure. Change the outside design temperature by one degree and it will ALWAYS result in the SAME total number of BTUs added or subtracted--if you don't believe this, just try it with your Manual J-based software! SO, it implies that the OVERSTATEMENT is linear as well.

Assumption 3) For this purpose heat loss from a structure is linear. We are essentially trying to determine loss at a state when it WILL be as close to linear as possible: at night, little wind, nearly constant temperature, no snow on the roof, and with interior temperature being MAINTAINED as closely as possible to "steady-state".

Assumption 4) For this purpose--and when under proportional control--the output of a given standing iron radiator is linear in the range of temperature operation with which we are most concerned. I was given some propietary information regarding this a while back and can only say that such can be VERY reasonably assumed. The things we "threw out" in assumption 3 have EVER greater effect.

Assumption 5) The "maintenance state" we are striving to achieve is SPECIAL and when we get close to it certain old rules do not apply. At maintenance one object heating another will be supplying the MAXIMUM possible level of radiation because it "WANTS" to do this. Radiation is striving to create stability out of chaos. You can literally view it as an "object" that obeys the laws of inertia.

Assumption 6) Radiation can be viewed as a PHYSICAL SHARING OF MATTER between objects separated by space. Viewed this way its complexities become easier to understand; not the least of which is why radiation seems to be an "object" that obeys the laws of inertia. The major difficulty is understanding that radiation is occuring BOTH WAYS between objects. A single object has an utterly equal desire to both maintain its current state and to change to become identical to other objects. That object does neither and both at the same time in "connection" with another object doing both and neither. As a result we get "waves" of varying magnitude moving at the same speed over distance but traveling different distances without changing speed...


Assumption 7) When under proportional control radiators are best placed where they enhance their radiative potential and are least influenced by OTHER convective currents in the space lest they magnify these currents and produce less radiation than their potential.
---------------------------------

TO BE CONTINUED:

"Using these assumptions to predict the temperature of your radiators and sizing them as a balance between size/cost and temperature."

Mike T., Swampeast MO

One more \"assumption\"

And this is the BIG one that hinges on all of the above!!!!

When under proportional control and at maintenance state the average temperature of the water can change without changing the average temperature of the surface of the iron!

This is the power of radiation!

If the surface temperature of the radiator has not changed convection cannot change even though more or less BTUs are being delivered by the radiator in response to the demands imposed by the changing environment outside the space!!!!!!!

The outside environment can ONLY change this if it changes the convection INSIDE the space. Of course this happens but this is PRECISELY the reason why we want to ISOLATE the radiation from EXTERNAL convection as much as possible!!!!

Mike T., Swampeast MO

Friday Humor

Said the engineer to the philosopher, "You know, the earth is mainly iron because that's what the sun would be if it were here."

Said the philosopher to the engineer, "Yes, but then I would be you."

Mike T., Swampeast MO

Viewpoint

The view of the radiator in this space.

US Capitol.

Tubes of this radiator in the form of four joined bell curves.

------------------------------------------------------

This five-tube radiator VERY closely approaches the radiant output of the "ideal" two-column iron radiator in a slightly deeper but significantly shorter space.

------------------------------------------------

Wayne: You have a choice between columnar and tubular radiators. Use columnar ONLY for two-column radiators. If the required length is greater than possible use a tubular radiator.

Mike T., Swampeast MO

Adjusting Heat Loss

This entails a guess--either that or software geared to radiant panel heating.

For reasons explained below, for OLD construction that has been thermally improved used 85% of a VERY THOROUGH Manual-J based calculation. Thermal improvement to include ceiling/roof insulation (preferrably cellulose); restored windows with good storms (or double-pane replacements, well installed); wall insulation; well constructed eaves in very good condition; some form of infiltration "control" (mine is tar paper).

The rest in this table are just my guess based on the above:

Reduction factors (Manual J)

10% -- OLD, thermally improved similar to above but no form of infiltration "control".

10% -- Older construction with original insulation; windows in decent shape either double-pane or storms; lots of recessed lighting into UNCONDITIONED space.

15% -- As first described.

15% -- Older construction, original insulation in walls, ceiling insulation at "modern" levels, few or no recessed lights into unconditioned space.

15% -- New construction, "average" quality workmanship, modern insulation levels, some infiltration control, lots of recessed lights into unconditioned space.

20% -- New construction, as above but either very few recessed lights into unconditioned space or nicely made, insulating "boxes" around them.

25% -- New construction, high-quality workmanship, recessed lights properly installed for convective control.

30% -- "Super" construction. Peferrably active HRV/ERV ventilation and good passive ventilation to ensure drying of cavities when inevitable water leaks occur

Wayne: am going to assume new construction; few (or properly-installed) recessed lights into unconditioned space.

The 6,500 btu/hr heat loss is reduced by 20% (1,300 btu/hr) to 5,200 btu/hr.

Now back to total square footage at various average iron temperatures:

5,200 / 180 = 29 sqft @ 190° AVG*

5,200 / 165 = 32 sqft @ 180° AVG*

5,200 / 150 = 35 sqft @ 170° AVG*

5,200 / 135 = 39 sqft @ 160° AVG**

5,200 / 120 = 43 sqft @ 150° AVG

5,200 / 105 = 50 sqft @ 140° AVG

5,200 / 90 = 58 sqft @ 130° AVG

5,200 / 75 = 70 sqft @ 120° AVG

*not recommended

**highest recommended average iron temp @ outdoor design.

----------------------------------------

Reason for that 15% reduction: I sort of "reverse-engineered" a heat loss for that type of construction. However, this WAS BASED on the 1.5°/btu/hr/sqftEDR/degree of temp difference. Therefore, there could certainly be an error, but such error will be on the "safe" side. By using this method I can predict the temperature of radiators in my home to a surprising degree of accuracy regardless of outdoor conditions.

If this 1.5° number is too low and the iron is actually putting out more it only means that the heat loss in the space is less than calculated. If the number is too high, it would seem to contradict both "dead men" and "new school" research and teaching. What really matters however, is that it appears HIGHLY linear when under proportional control at relatively low temperatures.

If you have access to proprietary radiant panel heating heat loss calculations (I've never seen or used), I would compute and compare to the above reduction factors.

Mike T., Swampeast MO

Final Sizing Selection

The last step is selecting the temperature of your supply water at design conditions.

This is where you have to juggle cost and physical space requirements of the radiation with "comfort". If your temperature is low enough (and the iron large enough) you will have comfort levels approaching that of radiant panel heat. While this will likely be the case in EXISTING systems using standing iron, the cost of radiators so large may not be feasible.


Were this space in my home with 5,200 btu loss, there would be from 75 - 120 square feet of radiation and I would need no more than 120° average iron temp @ outdoor design temp (8°)! I sincerely do not believe however that you have to go THAT low and that big.

What I would suggest is coming up with some sort of "average" surface temperature on a "typical" day and using such as a minimum size.

(Below is off-the-cuff. I have no idea how it is going to turn out, and how it is going to relate to the "highest recommended average iron temp @ outdoor design" mentioned in the previous post.)

Take a simple average between when you need the most heat (outdoor design) and when you need none (say 60°). For here in Southeast Missouri I will use 30° outdoor temp. By the way this is pretty close to the average low temperature from Dec-Jan in my area.

Recompute your heat loss at this outside temperature.

Wayne: I think you're in D.C. and I'm going to guess that the heat loss in that space at your "average outside" temperature is going to be about 3,100 btu/hr. Don't forget to include the "heat loss reduction factor!"

Now use this loss in the table for output at different temperatures:

3,100 / 135 = 23 sqft EDR @ 160° AVG

3,100 / 120 = 26 sqft EDR @ 150° AVG

3,100 / 105 = 30 sqft EDR @ 140° AVG

3,100 / 90 = 34 sqft EDR @ 130° AVG

3,100 / 75 = 41 sqft EDR @ 120° AVG

3,100 / 60 = 52 sqft EDR @ 110° AVG

3,100 / 45 = 69 sqft EDR @ 100° AVG

3,100 / 30 = 103 sqft EDR @ 90° AVG

3,100 / 15 = 206 sqft EDR @ 80° AVG

Compare the SQUARE FOOTAGES above to the table under Adjusting Heat Loss.

The highest recommended temperature of 160° average uses 39 sq.ft. This compares closely with the line with 120° average above.

Now predict the actual surface temperature of 39 square feet in a 70° room at this "average" outside temperature:

Heat loss = (avg iron temp - air temp) * (1.5 * EDR)

3100 = (x - 70) * (1.5 * 39)

3100 = (x - 70) * 58.5

3100 / 58.5 = x - 70

52.99 = x - 70

52.99 + 70 = x

x = 122.99

x = 123° average iron temp @ "average" outdoor temperature

-----------------------------------------------

Wayne: Please note that 39 square feet EDR is the SAME as would be "typical" sizing using an unmodified heat loss with 180° average water temperature.

Is this water temperature low enough and the radiation large enough to achieve a good level of the radiant "control" I mentioned that seems to occur when the space is close to constant state?

I don't know for certain, but I DO KNOW however that if ALL of this is just BS that the system will STILL provide sufficient heat at "normal" water temperatures!

If it's not BS (believe me, I'm being as sincere as possible) and you are using a condensing boiler, it should be condensing like MAD most of the time. Personally, if budget and space allow, I would suggest a 140° average at DESIGN conditions, or 50 sq.ft. of iron in this space. You could even go lower and bigger if you want.

Mike T., Swampeast MO

Adjusting for the Type of Radiation

Iron Radiators: For two-column and tubular radiators use the 1.5 btu/hr/sqft/degree difference with NO adjustment.

AVOID THE USE OF THREE OR FOUR COLUMN RADIATORS. If you MUST use them, still compute using the 1.5 number but add 20% to the radiation for three-column and 30% for four-column.

Why? Remember that this method is ALL ABOUT RADIATION. Those 3rd and 4th columns add almost nothing to the radiation surface exposed to the room. Even the dead men realized that 3 and 4 column radiators were not as efficient regarding total output per square foot as 2 column. Since we are WANTING radiation I am further "penalizing" these radiators as we are doing our best to slow convection.

Mike T., Swampeast MO

Achieving Radiative Control

Never forget that this is ALL about trying to match the btu output to the btu load of EACH space.

Tubes embedded in concrete floors ALREADY do this--because of their sheer mass. Even with very simple on-off btu input from the boiler, there is an ENORMOUS volume of BTUs available in the slab to allow the FINAL radiating surface to respond to changing conditions and effectively modulate their own output.

As long as the final surface temperature is low and nearly constant, convection-induce heating is GREATLY reduced and radiation is in FULL control. As temperature outside changes the temperature of the exposed surfaces TRY to change. Any attempt at change is instantly met with a POWERFUL force--radiation that increases on the fourth order of power with differences in temperature. This POWERFUL force is able to use that reservoir of BTUs to its advantage. As long as the outside conditions do not change VERY rapidly, the space stays under PROPORTIONAL RADIANT CONTROL even though the rest of the control is completely digital. Outdoor reset in this instance is there to try to keep either too many or too few BTUs available in the reservoir and thus avoid over and under heating with rapidly changing outdoor conditions.

This "reservoir" of heat DOESN'T REALLY EXIST however. It is a phantom that CANNOT BE FOUND. The very instant you think you have found it in the slab, it will appear in surfaces of the space. It is a thing that is not a thing; an "object" that defies all rational reason for existence. You will NEVER see it; you will NEVER find it but you CAN imagine it.

It is the part of the floor that is "one" with the rest of the objects at the same time that it is part of the other objects that are "one" with the floor. It is the part of your physical body that is part of the room and the part of the room that is part of your physical body. It is NOT an "object" but neither is it purely a force.

It is the portion of an iron radiator that is "in" you and a portion of you that is "in" the radiator BUT it depends on perspective. To the radiator, more of it is in you but to your body more of you is in it.

Convection-based joist bay radiant heating has a problem. The reservoir of BTUs in the panel is greatly diminished from that above. Rapidly dropping outside temp (or outdoor temp that stay very low for a protracted period of time) cause the problem. The POWERFUL force of radiation overwhelms the BTU reservoir and the surface of the floor drops because the convection cannot transfer BTUs RAPIDLY ENOUGH to overcome the control that radiation wants to exert.

Conduction-based joist bay and conduction based surface panel heating overcome this deficiency to some degree. They do this by using conduction to replenish the reservoir of BTUs. How well they do this has MUCH to do with how the system is controlled. If using a simple on-off system like in the concrete slab, they will be found somewhat deficient as the smaller reservoir due to a less massive object can become overwhelmed. So, we do our best to give it the temperature it "wants" via outdoor reset, constant circulation, etc. This helps the surface of the panel stay at a quite constant temperature similar to that big concrete slab.

Now how can we do this with something relatively small and hot like an iron radiator?

By accurately delivering the BTUs required for maintenance at any outdoor temperature. Of course it is impossible to do this with perfect accuracy, but we can get pretty close.

The most flexible way is to continuously vary the flow rate of BTUs into the water in proportion to the number leaving FOR EACH INDIVIDUAL SPACE. TRVs do this in the most reliable and inexpensive way that I know. If INDIVIDUAL control of in the spaces is too expensive or for some reason undesired, an EXTREMELY accurate heat loss "curve" should prove a reasonable, if much less flexible, method.

Unfortunately, it is impossible to develop a simple linear reset "curve" that will be "perfect" under all outdoor conditions. The btu reservoir will sometimes be too big and other times too small and the "maintenance state" of the space will rarely be approached for any length of time.

"Smart" electronic reset controls can certainly help. With such you should be able to keep water circulating nearly constantly and the btu reservoir "just about right" in size for most conditions. Since this is a system-wide (or at least zone-wide) control, the radiation in spaces served by this system or zone MUST be HIGHLY balanced.

With TRVs on ALL radiators however, the need for a "smart" reset control is diminshed as you can calculate a reasonably accurate curve and let the TRVs do the fine-tuning FOR EACH SPACE! You can also easily set a curve appropriate to systems where the radiation is poorly balanced.

A proportional HEAT SOURCE is the ultimate as far as this type of control is concerned. They "learn" and continuously attempt to keep their "curve" dead on target. Like with a "smart" reset controller however, this is a system-wide control and your radiation needs to be VERY accurately balanced. (Remember--we don't have the luxury of the extreme mass and reservoir of that concrete slab.)

Below is a ranking of general system controls suitable to this form of heating--in descending order of suitability. Number one will have nearly constant fire from the boiler and constant circulation of the water.

1) Modulating boiler and TRVs.

2) Modulating boiler and no TRVs.

3) Digital boiler, reset (mechanical or electronic) TRVs and TRUE constant circulation.

4) Digital boiler, electronic reset and NEAR constant circulation.

5) Digital boiler, mechanical reset and "sort of" nearly constant circulation.

Mike T., Swampeast MO

Placing the Radiation in the Space

Tall radiators on INSIDE partitions in oblique view of the most highly exposed exterior wall.

You don't HAVE to do this, but doing so will maximize radiation and minimize convection in reasonably insulated rooms with reasonably low air infiltration.

Advantages:

1) It's often easier to run piping in, to and near interior walls.

2) No need to worry about protecting or insulating like you must when pipes run through exterior walls.

3) Plan your layout carefully and you'll likely use significantly less pipe.

------------------------------

Do not cover radiators used in this position! You will turn them primarily into convectors and loose most of the benefits and encourge cold drafts across the floor.

Mike T., Swampeast MO

Computing Reset Curve

You can do this manually, but a simple spreadsheet will be HIGHLY useful--particularly for an existing system with poor balance in the radiation. Output of such is in the attachment. NOTHING difficult about I've give all you need to make one similar, but I kind of consider the original to be "mine."

If using mechanical reset use ONLY the type with two adjustments--one for base temperature and one for slope. A warm-weather shut-down control is recommended in addition. A setting of 55° - 60° should be good for most structures in most locations. When researching WWSD on the web a while back found that 55° was frequently used in Milwaukee.

BASE TEMPERATURE:

Add HALF of your delta-t (generally assumed to be 20°) to the temperature required at 60° (or so) outside. Merely compute average iron temperature of one space at 60°. If radiation is poorly balanced compute this for ALL spaces and use the highest temperature requirement. If using TRVs and you want enhanced recovery from setback add 10° more.


SLOPE:

Simply compute estimated radiator surface temperature (don't forget to put in your reduction factor in the heat loss) at different outdoor temperatures. You will find that a 5° change results in the same relative change in total number of BTUs with a Manual-j based calculation.

Plug your numbers into the formula for the slope of a line and the slope will be your reset ratio.

(rad temp hi - rad temp lo) / (outside temp hi - outside temp lo) = reset ratio

If radiation is well balanced you need only do this for one radiator. If radiation is not so well balanced use the one requiring the HIGHEST degree of surface temperature change for a given change in outside temperature.

If you have made a spreadsheet, start "playing"--particularly with the size of the radiation and noting the effect on the reset ratio. You'll find that as the size of the radiator increases its surface temperature decreases and the reset ratio decreases as well. Make the "radiator" the size of the floor and you'll probably notice something VERY interesting.

Remember how I said this is "all about the surface temperature of the iron"? I believe that calculating reset as a function of change in iron temp over change in outdoor temp is rather unusual. If you ask me if I really believe the numbers that come out with absurdely large radiation I can only say that the radiator in my master bedroom has NEVER been over 90° at maintenance and tends to overshoot quite a bit when recovering from setback. Temp response in there is nearly as fast as forced air!

Wayco Wayne

Good Gravy Mikey

When's the book coming out. I can see I touched on a favorite subject of yours. Thanks for your help. I am better prepared for the design on this job. We may not even go with cast iron now. The Architech is talking panel rads. I would like that because then I could give them all TRV's and constant circ. WW

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Contractor"

Mike T., Swampeast MO

Should be able to do the same thing with panels. Output graphs I've seen of those are linear. A few weeks ago I posted a link to a Canadian study. Tall panels (2 x as tall as wide I believe), operating at fairly low temperature in a maintenance-type situation. Measured proportion of radiant output was EXTREMELY high.

I'd avoid the use of the type of panels with convective passages in the middle if at all possible

Mike T., Swampeast MO

Question Wayne

"I would like that [panel rads] because then I could give them all TRV's and constant circ."

Do you say that because the lower cost of panels will allow TRVs and constant circulation in the budget? TRVs are WONDERFUL on iron radiators!

bob_25

Digital

Mike, isn't all energy transfer digital? I was told all radiant energy is transfered in discrete bundles called quanta. The quantity of energy in each quanta is the frequency times Planck's constant. Now what? bob

Mike T., Swampeast MO

Digital Energy Transfer

Meaning either happening or not happening?

Always that two-way street.

Does it happen one way because (or when) it's not happening the other for some instant of measure? But then I guess it would have to happen the other way in a "different" instant of measure? Is this happening because of differences in the objects? See the conflict? The kind of conflict that might produce a particle-like wave composed of two components at right angles to one another?

Does it happen both ways at the same time? Is this happening because of the similarities in the objects? See the handshake? The kind of "handshake" that might produce gravity?

Is it doing both at the same time that it's doing neither in yet another "different" instant of measure.

Those quanta "bundles" are yet another name for a "photon", correct? Interesting that a massless thing that only appears to behave like a thing can also be described as a "bundle"! Last time I looked "bundle" was a group viewed as a single collection.

Photons give high evidence that they travel in waves--but not simple waves--a DOUBLE wave with one at a right angle to the other...

Photons all travel at the same speed. Right?

Waves have different frequencies (distance between peaks) and different amplitudes (height of peaks). Right?

Now, this "bundle" that has no mass but has characteristics of a single particle is traveling in two paths at the same time that vary in length. The speed at which this "bundle" can move is fixed as being the "speed of light". YET, "bundles" with different wave characteristics (meaning a different wave form or wave length) all arrive at the same time despite each having traveled TWO paths and over different distances!!!!

I can buy that--but only if I imagine that part of that bundle was already there before it arrived!!!!

When you research any of the terms regarding electro-magnetism or gravity the only real "constant" I can find is the total confusion of what it is, when it is and even if it is an is.

Why is it so difficult to think of it as a force that is not an object resulting from simultaneous sharing of matter between separate bodies? Is it because if it could be "proven" by science it would give a rational basis for spirits (dare I say God?)--something that rational scientists NEVER want to encounter? Is it because if "proven" by religion it would actually explain the nature of a spirit (dare I say God?)--something that religion NEVER wants to do?

Please don't get your feathers ruffled over those last two questions. I don't claim to have the answers or even to have asked the proper questions.

I just believe that it is not particularly difficult to manipulate infrared radiation if you keep the available energy at a level appropriate to the mass of the emitter and that by doing this you can reduce fuel consumption and increase human comfort. The best part is that you don't at all have to understand "why" in the least. "Why" is just a question I have never been able to stop asking...

Mike T., Swampeast MO

Cookie Reset

nm