While this is not your usual greening question, it does have to do with finding a way to reduce the carbon footprint in our home. Unfortunately, we're not in a position to go solar, use underground water, etc. (although we have taken many steps to keep our energy consumption down). Anyway, here goes: Hi, everybody, I'm a newbie here so please forgive any gaffes. I'm in the process of trying to find out how well the old heat distribution system in my old historic home works. I already have a professionally done assessment of the efficiency of our boiler and measure the % fill of my propane tanks daily (adjusting to 60 degrees). In brief, there are four zones whose heat comes from a propane boiler. The lowest floor and the fourth floor have baseboards. The second and third floors have old-fashioned radiators. The pipe scheme is primary/secondary. I have a datalogger with 2 k–type thermocouples attached to the supply and return pipes from and to the boiler. As per a recommendation from a trusted source, they are located at least 10 pipe diameters from the respective inlets and outlets of the boiler and are attached on either side of the primary/secondary tees. I realize that these will not give the true temperature of the water inside the pipes, but my reasoning is that the delta between those two will give a decent estimate of how much of a temperature drop there is between the supply and the return. Is that a reasonable assumption? If not, could you explain? So, that's the first point. I'm also using the datalogger, which is set for 10 second intervals, to figure out the actual flow rate in the pipes: when the supply temperature has its first spike after a call from a thermostat, I then look at the corresponding spike on the return side and time the difference. The interval between the two tells me how long it has taken for the ""pulse" of newly-heated water to arrive back home. Since I also know the length of pipe and the internal diameter, I can then figure out the actual gallons per minute , real-world, being pushed through. And, from this, the actual BTUs being delivered to the space. Any problems with this reasoning? Constructive criticism is more than welcome. Later on, and I haven't gotten there yet, I am planning to measure the delta T's at the supply and return points for individual sections of baseboards and radiators to see if they are functioning anywhere near their design capacity. Anyway I'm hoping for some knowledgeable folks to weigh in. Thanks for any responses.

Lots of interesting questions here. I'll nibble on a few now and might come back for some more.

To get a more accurate reading of the water temperature, I'd first stay away from a T because it has extra thermal mass that will slow the response, and I'd insulate the pipe really well where the thermocouple is and at least a foot in either direction. The assumption that the drop is the same even if the probe isn't accurate is shaky--you can see that by assuming the worse case scenario that the reading is halfway between the room temperature and the water temperature.

Next I'd consider your objective a little more. Suppose one of your radiators has crud on the inside or outside and its heat transfer isn't as good as it it was when new. It's still going to be 100% efficient. All the heat that comes out of the water goes into the room. It's just that not as much comes out and the water going back is warmer than it would have been otherwise.

So I'd instead think in terms of the efficiency of the plumbing--are you using more pumping power than you need to be? And you you happy with the uniformity of the heating of the different rooms? If you are returning water from some zones without much change in temperature, you are pumping faster than you need to be pumping. In any case, you can probably dramatically reduce
pumping power by swapping out the pump for a modern high-efficiency "ECM" pump such as Grundfos Alpha or Taco Viridian. Are your zones individual pumps or zone valves?

You could also consider whether any of the pipes go through unconditioned spaces or exterior walls, and if so, make sure you don't have excessive heat losses there.

Timing the pulses to find out flow rate is a cool idea, but I don't see how that would work with radiators involved. Maybe if you put both probes on the supply pipe or both on the return?

Let's look at an exaggerated example. A hypothetical temperature sensor reads 1/2 way between the ambient (air) temp and the temp inside the pipe. So we could get:

ambient = 0F
pipe 1 = 100F
pipe 2 = 200F

reading from sensor 1 = 50F
reading from sensor 2 = 100F

difference between sensors = 50F.
actual difference between water in the pipes = 100F

On the other hand, put some insulation over the sensors and block all air flow and they will read close to the temp inside the pipe. Check that the two sensors are calibrated the same.

Timing the flow thru the system might not be a strictly accurate representation of flow rate/volume if you have parallel pathways. If you have a shorter less restrictive paralel loop in the system, it might return the warm water to your sensor faster than the average system flow...

You might be overthinking this. Chrs is right, even with crud filled radiators, the internal radiation system is basically 100% efficient. IE: all the heat you are putting into the system IS making it into the house ultimately. If you have a good evaluation of your boiler efficiency, then that percentage of THE BTU's burnt in propane is going into the envelope. Now temp measurements of the different zones might show you that heat is going places you don't really need it, but it IS going into the house...

Have you done any other energy audits in the home such as insulation coverage/quality and a blower door test and evaluation? In an older house, air infiltration can(and probably does) have the loudest voice in your heating load. Your insulation quality/quantity in an older house is probably right up there also, along with window size and efficiency. Controlling the air infiltration is probably the most inexpensive and cost effective place to start.

Your suggestion about the insulation makes sense to me and goes along with advice I've received elsewhere.

I get it and agree about the insulation. As to the calibration when I purchased the (Fluke) K-2s I had the same idea and tested them on 32º F water/ice as per instructions I found on the 'net and they were very close to each other; besides this I also used them to measure pipe sections fairly close to each other and again the results were close. So I'm reasonably sure they're good.

"To get a more accurate reading of the water temperature, I'd first stay away from a T because it has extra thermal mass that will slow the response, and I'd insulate the pipe really well where the thermocouple is and at least a foot in either direction." That sounds like a good suggestion to me and I'll try to work with it as much as possible.(I have issues such as proximity to the flue and convection from the boiler that limit where I can place TCs.) As I understand your second idea, the effect of "the crud in the radiators" is simply that the circulators have to work more to get heat to each zone. So this drives your electric bill up. And that's the ONLY effect on efficiency. (Did I get this right?) I have to think about this. As to the pipes going through unconditioned spaces and exterior walls: you're absolutely right. It's a 200-year-old building and there are definitely areas where this is happening. But outside of the budget. Everything we can get at inside the shell we've insulated. The zones each have a dedicated pump: (Lower-an oldie Anderson E6312 LR37479; all the others: two Taco 007-F5s and one Taco 007-BF3-IW). On the timing with the radiators, I don't understand. A radiator-only zone calls for heat (all other zones have been turned off for an hour or two so as to isolate the effects of the call). The supply temp near the boiler shoots up. X amount of time later the return to the boiler spikes. From this we get the amount of time for one circuit. We know the ID of all the piping and total number of feet of pipe in the circuit and know that water is not going anywhere else since we've allowed a long time after the last call from another zone. We calculate the number of cubic inches in the pipe and the multiply by the conversion factor to get gallons of water. From this we get GPM. Using the standard formula (500 x GPM x the ∂-temp) we calculate BTUs. How is this different from baseboard? Guess I don't understand. Anyway, very much appreciate your responses. Thanks. It's always good to hear from people who know their stuff.

On the timing, I was imagining a loop with cast iron radiators, but perhaps you are talking about a loop with just baseboard radiators? In that case, I think it works, as long as it's a simple loop, without branching anywhere. But that would only tell you the gpm for that one loop.

Thinking through the first part of what you said above about the heat making it into the house. Going to give this some thought. As to the energy audit part, you are right on the money. When we had one done an astonishing 43% of our heat loss was due to air leakage. We had the blower door test done as you suggested. We had subsequent air sealing done and brought heat loss due to air leakage down by 43%. Confirmed by another blower door test. So we came to the same conclusion you did, that controlling air infiltration was the most inexpensive and cost effective place to start.

Chris, On your point about the efficiency of the radiating elements (radiator, baseboard) vs the plumbing efficiency, Im trying to educate myself reading some of what Siegenthaler has out there. Reconsidering what I thought I knew but, honestly, not quite there yet on your conclusions. If I'm getting Siegenthaler right the effect of GPM on BTUs is relatively small until you get to really low flows, on the order of 1-2 GPM, mainly because you start to get laminar flows as opposed to turbulence. My initial readings on my lowest zone, which I'm going to revisit with the improved insulation around the TCs when it again gets cooler, indicated a flow of 0.99 to about 1.2 GPM. Which is why I'm a bit suspicious about that zone in particular. Anyway, I'll be back when I get better data. BTW, again, thanks for the comments. Helpful.

Chris, and Ronmar,

Still doing my homework (Siegenthaler, Holohan and others) but some points of clarification.

The loop I've done most of my work on so far is a single loop consisting of the pipe and the baseboards. This zone, one of four, has its own circulator that sucks water back into the boiler (that's how it's set up as opposed to pushing it out into the zones). There is negligible vertical push since boiler and zone are on the same level. When I say single loop I mean it has no direct interaction with the other zones. Water exits the boiler and comes back to it in one circuit. The only condition, I think, that could have some bearing on all this is that there is primary/secondary piping.

On the crud: in one way the crud acts like insulation in that it affects the rate of heat transfer into the living space and can make it so that at the end of one circuit of the loop not enough heat has dissipated per unit time, making it necessary to run the boiler/circulators more. The whole thing here is throughput into the space. And that varies not only according to efficiency of the burn in the boiler but also according to the condition of the pipe loop. If the piping is 100% efficient as all the heat gets into the space anyway, why bother with fins? Or vacuuming them, or straightening them out, to improve radiation and/or convection? Or having larger or smaller diameters of pipe? So, I'm frankly puzzled when I see this: "Suppose one of your radiators has crud on the inside or outside and its heat transfer isn't as good as it it was when new. It's still going to be 100% efficient. All the heat that comes out of the water goes into the room. It's just that not as much comes out and the water going back is warmer than it would have been otherwise." or this : "Chrs is right, even with crud filled radiators, the internal radiation system is basically 100% efficient. IE: all the heat you are putting into the system IS making it into the house ultimately. If you have a good evaluation of your boiler efficiency, then that percentage of THE BTU's burnt in propane is going into the envelope." If fewer BTUs are being dissipated into the space on each cycle through the loop, won't it ultimately take more cycles to satisfy the call in that zone? And mean that my non-modcon boiler will have to burn longer? And circulators suck in water longer? Maybe I'm missing something here. It seems to me that there is an efficiency of the distribution system apart from boiler efficiency. So, no, if my boiler is 84% efficient that doesn't mean that the whole hydronic system is 84% efficient. Anyway, that's the way it seems to me at this point. I'm looking forward to comments.

The only way I've been able to make sense of the 100% efficiency point is that if it takes longer to satisfy the call, the time between the first call and the second is reduced compared to the more efficient distribution system. So, crud-filled system A fires on at 2:00, doesn't satisfy the call till 2:20 during which time there have been, say, 4 burns. Cleaner system B would also fire on at 2:00, would stop at 2:10, with 3 burns. The second call for A would be an hour later at 3:20 and for B at 3:10. So, over time, it would "even out". Maybe, but it seems a unlikely that it would just so happen to be the same energy consumption.

On a different subject I've taken your suggestions about the use of insulation on the TCs. Will re-doing my figures and waiting for the next cold snap.

Thanks for lending an ear. Hydronics is definitely a complicated subject.

The boiler dosn't care about flow or BTU output. It's little pea brain only cares about temperature. You are correct, if the piping/radiators are fouled, heat will radiate to the space at a slower rate, and warmer water will be returned to the boiler causing the burner to shutoff sooner(or in some cases modulate it's output lower) to try and maintain it's set temperature. This will result in the burner short cycling, but the system still running/circulating as the thermostat is still calling for heat. But you still have 100% radiator efficiency as 100% of the BTU put into the water IS making it into the house. Adding or improving surface area just means heat will transfer faster and burner will run more to make up for the additional heat loss from the water. Room will heat faster and thermostat will cancel call for heat sooner. Still basically equates to the same thing with the same ammount of BTU being delivered in a shorter time.

Now wether the ammount of heat making it into the room is enough to meet demand and maintain comfort level is another story. If the radiator cannot meet demands then the thermostat will never be satisfied and system will run constantly but room temp will never satisfy the thermostat.

I think perhaps we got down a side road by talking about efficiency Your delivery system is basically 100% efficient I am guessing that you are more interested in comfort level VS fuel burnt? You already know what your boiler efficiency is, and unless it needs maintenance, I don't think you are going to improve on that significantly short of a more efficient replacement heat source.

Now where exactly that heat is being applied can have a real effect on comfort vs consumption. IF the pipes are shedding heat in the piping around the boiler, under the floor or in the walls and not the rooms you occupy, you are loosing any benefit of the insulation in the walls of the rooms by delivering heat outside the rooms.

Larger radiators might reduce cycle times and make the room feel more comfortable, but the added dissipation of heat from the radiators is going to increase burner requirement. It might improve overall comfort vs consumption by a very small margine as the water returning to the boiler is at a little lower temp and sheds less energy in an undesired location. Well insulated pipes will also do the same thing

Remember heat transfer is about surface area, R value, temperature difference and time. In a room/house infiltration also plays a role. So all things remaining the same, if you dump more heat into a room, you increase the temp difference and the heat loss from the room. The added heat will also increase stack effect(warm air rises) which will increase infiltration losses by pumping more warm air outside faster.

SO if the burner/boiler efficiency is at it's peak, and your boiler and circulation plumbing is well insulated, your fuel consumption at a given room comfort level is directly proportional to the outside temp, overall room Rvalue/Ufactor(BTU per sq/ft per degree F temp difference per hour) and infiltration rate. Difficult to overcome in an old house short of building in more wall thickness and Rvalue, increasing floor and ceiling Rvalue and updating the windows and doors and reducing infiltration rate...

If the rads are not releasing heat efficiently, the water will return through the boilers more times then necessary. Each time the water returns through the boilers, more heat is lost through the exhaust, the older the boiler the more that is lost.
The more times the boiler start/stops to deliver the same btu's to the house, the less efficient it burns.

If the radiators are full of crud, then the likelihood of the boiler being scaled or cruder up is also high. If the boiler is scaled up, a higher percentage of the heat goes up the exhaust then if the boiler (exchanger) is clean.