Does Anyone Know?

Mike T., Swampeast MO

Of any study, table, etc. for estimating output of standing iron at <I>very</I> low temperatures--like 85°?

Fred Harwood

A guess

The rate of heat transfer from, say, an iron or steel pipe to the space it's in, would be a function of the temperature difference. The greater the difference, the greater the rate of transfer per square foot of pipe. Perhaps the rate might be found in Mark's Standard Handbook for MEs.

Canuck

Burnham Heating Helper

Handbook has a table for heat loss in bare steel pipe,(Btu/ft/hr), but it starts at 100 deg F with an ambient of 70 deg F.
If it's of any value, here are some samples at 100 deg.

1/2 = 13

3/4 = 15

1 = 19

1-1/4 = 23

1-1/2 = 27

2 = 33

2-1/2 = 39

3 = 46

4 = 59

jerry scharf_2

maybe convective = 0

Mike,

I know of no such table, but that doesn't mean anything. What if you treat the convective output as 0, and look at the purely radiative output of the iron at that temperature. You should be able to make some pretty simple assumptions abot geometry without impacting the accurcy too much.

If this is chasing after your low flow TRV controlled radiators, I would measure surface temperature at multiple places across the radiator. None of the functions are linear with temperature.

After saying this, I realize that with a wide delta T across the radiator, I would not trust any tables. Too many specifics to a given radiator vs. flow to make it reasonable. I think of the water as assymtotically approaching ambient temp as it loses heat (that's the first order function.) The lower the flow, the steeper the first part of the curve is. The exact geometry of what spots are hottest and what the radiative geometry is gets very messy. The steeper the curve is, the more this impacts the effective output.

I hope this is what you were thinking of.

jerry

Mike T., Swampeast MO

Understand that, but the geometry of a standing iron radiator is MUCH different than a simple round pipe...

Mike T., Swampeast MO

Have that (or similar).

Mike T., Swampeast MO

Definitely what I'm thinking of...

I agree that it's hard to find much driving force for convection at such low temps. Sure seems that with outdoor temps in the 30s or higher that I'm heating a house almost totally via the radiation of standing iron!

The delta-ts across the rads are all over the place and I'm not even sure what number to use for the supply temp... Some of the rads have a very high delta-t (35-40°) if I measure from the TRV body to the return union nut--others much less. Some have a quite high delta-t as measured from one end of the rad to the other (20-25°) others VERY little--(2-3°)!!! In general the higher the delta-t across the radiator the lower the loss in the room, but not always. Very bizarre. Return temps are frequently at/very close to ambient--but not always and not always where you would expect...

Have tried to estimate the proportion of radiant output before but it gets terribly hairy and I keep having to make assumptions that may or may not be valid and/or accurate. One of the worst is trying to determine the effective area of radiator "tubes" that are in actuality made up of four bell curves.

jerry scharf_2

can you post a picture of your rads

Mike,

I know the pain of trying to define a solvable radiant emission geometry for complex things. You can often use the fact that facing surfaces of a common temperature cancel.

What would be great is to take one of those wide delta T radiators and measure top middle and bottom of each column. The ones in the middle (if it's a three risers per column) do very little for radiative, instead being designed to drive convection. As long as things are relatively steady state, they can be measured sequentially.

jerry

Larry Weingarten

An idea...

How about building an insulated, five-sided box out of rigid foam and putting a 100 watt light in it? Record the temperature in and out of the box. Do it again with 200 watts worth of bulb in order to plot energy input vs temperature and time. Once you know how the box performs, put it over your 85 degree radiator. It seems you could chart output without having to worry about the geometry of the iron or how the output was occurring. Does that approach sound productive?

Yours, Larry

Mike T., Swampeast MO

Here's a photo looking down on one of the rads. They're all U.S. Capitol thin tube, 3-6 tube (most are five). The shadow on the floor gives the best idea of the true shape.

Once your point of view shifts a few degrees from "straight on" you see only iron. Room walls must be riddled with horizontal "stripes" of radiant focus.

Mike T., Swampeast MO

Not Sure Larry

I'll loose the radiant "connection" to the room and with very little mass in the rigid foam, I think the radiator will just raise in temperature and start convecting. TRVs use remote-mount sensors on the walls so they'll sense a drop in room temp and the rad will get even hotter. If I put the sensor in the box, it would just turn down the flow as the air in the box heats rapidly.

jerry scharf_2

wouldn't expect \"radiant stripes\"

Mike,

this is about what I expected. There is lots of internal absorbtion/reradiation in a radiator like this. I've seen some FEA rradiation analysis work and it turns out to be fairly close to a simple box. Next you assume that the floor and rear wall are pretty much steady state and cmopute the R value loss for the wall behind that. The energy that isn't lost to this has to go into the room.

Now comes the fun part. You have a top, two ends and a face. You need to take a set of measurements at intervals across each of these, and create small square surfaces and for each face compute the radiant output. The more faces, the more accurate the calculation. We'll treat the faces as flat, since we are not interested in detailed energy distribution, just aggregate output. Once you have the radiative output, you calculate the output from the MRT of the room, and the difference is the radiator net output. Sounds like a job for a spreadsheet.

Make sure the measurements are made with thermal paste but no insulation. The insulation would cause false readings.

I hope this all made some sense.

jerry

Canuck

Standing Iron

Sorry Mike - I missed what you really needed to know. I guess we use different terminology up here. A standing iron is what we use to get wrinkles out of our clothing. The picture that you sent - we call that a rad.

Mike T., Swampeast MO

No problem--I usually say "standing iron rad".

Mike T., Swampeast MO

Makes good sense--did similar one cold, dreary winter morning and posted numbers here. Didn't adjust for variance in the rad surface temp however--just used an average. Was trying to compute the percentage of radiant to total output and many said that the numbers were "impossible" because after all, engineers insist that floors still have about 50% convective output and everyone knows that it takes a big temp difference to get much radiation...

Amazing to me how certain engineers can be when every good source I've found on heat transfer conceeds that radiation is the least understood.

At least now I have a modulating heat source so can make a very accurate estimate of total loss at any given point in time.

jp_2

close the door!

close the door and measure the temperature of the room!

I assume you know the heatloss of the room, so you'll get the output of the rad at 85F.

jp_2

stefan boltzmann


try 302 watts/sq meter.

I used iron emmissivity .64

jp_2

estimate /round pipe

you should get close enough numbers by modelling radiator with round pipe.

Mike T., Swampeast MO

Yes, I know the calculated heat loss and even with "correction" for overstatement, the numbers just don't seem right.

Prior to installing a Vitodens, I could quite accurately predict radiator surface temperature. (Old boiler operated under constant circulation with nearly identical reset curve.) The rads have NEVER been hot--but now they're consistently cooler than my predictions.

Actual fuel consumption agrees reasonably with predicted loss, but the freaking rads are so cool that their output per square foot per degree of temp difference between rad and air has to be higher than the dead men said was possible with steam in a 70° room!

It almost seems that part of the energy from that RADIANT burner is passing through the system to appear as RADIANT energy from the rads--without appearing as measurable temperature.

John Cockerill

A couple of things to consider. If the one end of the radiator is cooler than the other this may be a function of the circulation rate. The thermostat can satisfy quickly with high temps and may in fact be satisfied with the help of the anticipator before the radiator is full of hot water. If you decrease the aquastat encouraging longer circulation, you may find both ends of the radiator are the same temperature.

John Cockerill Exquisite Heat
www.Exqheat.com

Mike T., Swampeast MO

One Round Pipe is Easy

But 60 on average? All with a view of each other and the space that changes with perspective? The column rads are a cakewalk by comparison. The dead men knew that 2-column rads were very good at radiating with the interior sections of 3+ column rads only convecting.

If only I could fill the rads with vapor instead of water!

jp_2

simple rectangle

as jerry was saying, step back, cross your eyes and consider the radiator a blurry rectangle.

I'm thinking there may be more convection going on than you think? I'm assuming with the 85F rad the room is about 68ish?

also I think the big problem is that you are trying to take a snap shot in time and analyzing it. lots of variables to consider:
1.) solar effects
2.) adjacent room effects
3.) heat storage of the room and adjacent rooms.

so are these variables adding or subtracting within your snap shot in time that you are looking at?

if its partially sunny that day, you need to measure the temp of that outside room wall to do an accurrate heatloss at that moment in time. plus window gain, of course.

on a 30F sunny day my cabin will raise to 70F, all solar gain, woodstove dead cold! not much heat storage, walls & ceiling R15, floor R10, open crawl space.

I'd be interested in seeing all your data here.

"""" It almost seems that part of the energy from that RADIANT burner is passing through the system to appear as RADIANT energy from the rads--without appearing as measurable temperature.""""

I still think that radiation is measureable energy by way of temperature, not a ghosting energy. otherwise boltzmann, planck,wein and others would not have been able to derive those radiant energy equations basic on temperaure??.

jp_2

re calculate

the 302 waatts per sq meter was for a radiator with no reflections.

I assumed 65F room temps and got 40 watts per sq meter.sounds more reasonable. I think your rooms may be coasting?

if radiation is doing 'work' it will show up as temperature change in your system.

Mike T., Swampeast MO

Thanks JP. I try to do my measurements either at night or on extremely overcast day--always when the outside temperature has been quite steady. No TRV changes within days of the measurements. I've mainly studied one radiator in one room (my office). Attached document shows the first time I measured the temp of all of the radiation in the house within a reasonable time. Tried an IR thermometer (a cheapie without an emissivity adjustment) but found it inaccurate and quite influenced by ambient air temp.

The one radiator in the house that might be considered "hot" is in the NW corner directly above the basement two-car garage. The mains running through there are intentionally uninsulated, but being almost completely above grade with an uninsulated brick foundation, it's still the coldest part of the basement.

Everything is "coasting" as much as possible. MRT is steady. Boiler output is steady. Room temp is steady. Flow is steady [presumably--I have no way of measuring flow].

Overall heat loss prediction agrees quite closely with the input of the Vitodens. The Vitodens was the final step in confirming that my heat loss predictions were reasonably accurate. Impossible with the digital and greatly oversized cast iron Weil-McLain.

Problem I'm encountering is that the radiators seem even cooler than before under very similar circumstances and I've had to "up" my estimate of output per sf EDR per degree of temp difference--nearly to the level of low-temp radiant panels. That might not sound too bad until you remember that a significant portion of a cast iron radiator's surface area is merely re-radiating to itself!

From what I can glean from research, the equations for radiation have generally come from conditions with high temperature differential using "perfect" constructions. The best translations to reality seem to come from commercial food processing where you're still dealing with high differentials and simple constructions. I've yet to slog through the multi-page equations in the one text I have regarding radiant transfer in reasonably "real" living spaces--not sure I'm even capable. Even then, they're based on very simple surfaces (big, flat things like ceilings/floors/walls) providing the radiation.

"I still think that radiation is measureable energy by way of temperature, not a ghosting energy"

You're probably 100% correct. My problem--and probably everyone's problem--is that it's extremely difficult to get truly accurate temperature measurements--particularly at low temperatures and low differentials. With radiation increasing at such a high order, tiny errors become big errors when the temps involved are close.

Mike T., Swampeast MO

Am rapidly running out of time for tests. We may have a cold spell when I have time or it might have to wait until next year...

Larry Weingarten

In your spare time...

line the inside of that insulated box with copper tube. Hook up a well insulated tank filled with room temperature water and see how quickly BTUs are pumped into the tank. I'm loaded with "good" ideas. ;~) It just seems that if a physical test can be designed and made, it is bound to be more accurate than a mathmatical model based on lots of "close" assumptions.

Yours, Larry