Whom do you trust?

Gordan

Uponor CDAM radiant ceiling output chart for a 12" o.c. ceiling with plates suggests that the slope of the output vs. supply temperature curve is approximately 1.6. In other words, the same slope as the output vs. average ceiling surface temperature curve. That would mean that they would expect the ceiling surface temperature to increase by one degree for every degree increase in the supply water temperature, which would make sense if 1/2" sheetrock had a R-value of 0.



Siegenthaler's book puts that slope at 0.71, <strong>for an 8" o.c. ceiling with plates</strong>.



CDAM would also have you design to an "ideal load" of 40 BTU/sqft and room temperature of 70 deg. F, so any discrepancies between their approximation and what actually happens would be maximized. An example: according to them, this output to a 70 deg. F room is attained with a 117 deg. F supply temperature at 10 deg. F delta-T. Siegenthaler's approximation would specify that, to attain this output at 10 deg. F delta-T, you would need the supply water at 129 deg. F - and that's for 8" o.c. tube spacing. CDAM tells you that this is in excess of the maximal allowed supply temperature under sheetrock.



So, um...



Me, I find the CDAM unconvincing.

CMadatMe

Question

Is 12 degrees water temp that critical. What's the worse case you have here. Might have to turn the water temp down?  One thing I have learned in the 20 yrs of doing radiant, It's a better feeling to have the ability turn it down than no way to  turn it up to provide comfort. Nothing is an exact science in radiant every customer has a different comfort level. As long as you have the ability to overcome the loss without maxing out on the max surface temp you win.

Gordan

I think we're in agreement here, Chris...

12 degrees is not that important if the error is on the safe side, but CDAM not only errs on the optimistic side of output but also specifies wider tube spacing, so add at least a few degrees there. The worst case is that, if you follow the CDAM, you might have to turn the water temp UP - but there's no margin for error on that side because their design parameters leave none. So, then, the worst case is that the system simply won't perform... or you'll be going against the recommendations of the drywall manufacturers by bumping up the supply temp.



Here's my point: there's no shortage of Wallies who complain about the happy-go-lucky corner-cutting competition who underbid and underdesign, leaving homeowners stuck with an underperforming system. It's easy to criticize dilletantes and charlatans, but do we have here a case of a major manufacturer encouraging, in their design manuals and software, what appears to be corner-cutting?



Another example in the CDAM has me puzzled, as well: in their radiant ceiling tutorial they work out that .56 GPM is needed for the radiant ceiling, which translates to .8 ft/sec with 1/2" tubing in their own chart - the chart that says that you should maintain minimal velocity of 1.5 ft/sec. Yet they never even sweat this detail. (As an aside - how critical is flow velocity? You need at least .9 GPM per loop in 1/2" tubing to exceed 1.5 ft/sec - that seems like a lot! It would be near impossible to get a 20 F delta-T without exceeding recommended loop lengths.)



Personally, I'm inclined to err on the side of caution and design for 30 BTU/sqft. That way, there's some headroom left in my radiant ceiling design - pun intended - in case the estimates were too optimistic.

Jean-David Beyer

Delta T in heating.

"how critical is flow velocity?"



John Siegenthaler says somewhere that  it is important to get the flow rate above 2

feet/sec so that the bubbles can move down to the air extractor, and

below 4 feet per second so as to minimize noise and wear. But other than

that, who cares what the flow rate is in an existing system?



"It would be near impossible to get a 20 F delta-T without exceeding recommended loop lengths"



Since I am not a professional, I have some book knowledge, but the only experience I have is with the system in my own house. I now have two heating zones: upstairs with two 14 foot long Slant/Fine baseboard  heaters, and downstairs with at grade radiant heat using 1/2 inch copper tubing. I do not know the spacing of the tubes, but I do know that there is a separate loop for each room downstairs, with a ball valve to regulate the flows. I actually have three of those valves full open, one a little closed, and one nearly closed.



The contractor chose to zone with Taco 007-IFC circulators. The boiler is a mod|con with outdoor reset. It is the smallest model in that manufacturer's product line, but it is still somewhat too large. It modulates from 80,000 BTU/hr input down to 16,000BTU/hr input. I figure that the house requires about 35,000 BTU/hr at 0F outside and design day is 14F.



Since I can see the supply and return temperatures from the zones (and the domestic hot water as well), provided only one zone is running at a time, I observe much smaller delta-T than 20F. For the upstairs zone, delta T is typically only a degree or two, and downstairs it is a little more, but only if the circulator just started and the water has not made a pass through the system yet does it does it get to 10F. In spite of the low delta T I experience, I have no difficulty heating the house.



 If the flow rate is "too high", but not over 4'/sec, all that happens is that the temperature uniformity is a bit better than it would otherwise be. The cost to run it might be a little too high, and I could replace the circulator with a Taco 005-IFC or something, and that might be slightly cheaper to run. Were I to install a delta-T ECM circulator, I suppose I could dial in any delta-T (within reason) that I wanted, but I imagine the flow rate would be so low that it would cause other problems, including not getting the upstairs bubbles to come down to the air extractor, and non-uniformity of the series-connected baseboard units upstairs.



If I were putting hotter water into the zones, I would want more temperature drop through the system so as to get more condensing. But since the maximum temperature I put into the slab is 120F, I should get condensing all the time from there. And upstairs, the maximum temperature I put into the baseboards is 135F (and that at 14 degrees colder outside than on the design day), I get condensing even with no (measurable) temperature drop at all.

CMadatMe

CDAM

The design in the CDAM is based on a 10 degree delta-t not a 20 and uses 40btu sqft.

Gordan

I'm aware of that.

40 BTUhr/sqft, 10 F dT, 117 F supply temperature, 12" o.c. tube and plate.



There are two distinct questions that I was asking:



1) Leaving a paltry 3 degrees F of headroom for the supply temperature in their "ideal case" design - is that not cutting it pretty close, especially if their design estimates turn out to be overly optimistic (which there's arguably reason to believe they are - Siegenthaler's estimates are much more conservative)?



2) Why are they specifying flow velocity that's one half of what their own charts say should be the minimum?



That's all. I'm sorry if I'm confusing anyone.

Gordan

Thanks, JD.

I didn't mean to imply that there would be trouble heating the house with a smaller delta T (as long as the flow is sufficient.)

Jean-David Beyer

Delta-T again.

I did not think you were implying anything.



It is just this concern by many about having the right delta-T. And what the right delta T is.



In my case, I can control my mod-con to put out any temperature I want into any zone I want. And right now the only way I could control the flow would be to change circulators. I infer from what I read that most people would think my delta-Ts are too low. But with a mod|con boiler, what I am most concerned about is the return temperature (providing I get enough heat to heat the house): the lower the better. And if the supply temperature assures condensing, I do not think I care what the delta-T is. If I put a smaller circulator in my upstairs zone, with a 1/35 HP moter instead of a 1/25 moter, it would take quite a while to pay off the cost of the circulator and its installation.

NRT_Rob

you said your contractor zoned with circulators

so you are wildly overpumped, and wasting energy. that is why we would say "your delta-Ts are too low", not because of any heat output issue.



If you had one circulator, sized properly, installed from the beginning, you'd have used less power from day one and not paid any more for it.



also, lower return temps are better for your mod/con efficiency. it's not on/off around the condensing point, it can condense more at lower temps.



most pros routinely ignore the minimum velocity question without consequence. It's a relic of older days I think. If you have a good air separator on the system, you don't have to physically move bubbles back to the separator. They will simply reabsorb into the oxygen starved water.

NRT_Rob

I design radiant ceiling regularly

and I do think there is more to the story than the CDAM chart shows. especially at very low temps such as 100 degrees or lower.



that said, I use it, but I also stick with a 30 BTU/linear foot max typical. Just in case.

Gordan

Well, that about answers all my lingering questions!

Thanks, Rob!

Jean-David Beyer

I suspect you are right.

I suspect you are right about being overpumped. I could calculate what flow rate I really need for my upstairs zone, because I know how much pipe (about 75 feet of 1/2" copper and about 55 feet of 3/4 copper (including the baseboard) all in series, plus lots of elbows. So I should be able to calculate the head. I know how many BTU/hour I need (not much). But without calculating, I can infer, since I have such a small delta-T that a Taco 007-IFC is too big for there. I could guess that an 005-IFC would be enough up there. It might be even too much.



Downstairs is a different problem. It may be overpumped as well, but it is in-slab radiant and I have no idea what the pipe spacing and lengths are. I know it is 1/2" copper tubing. Now since the delta-T there was never over 10F, even when I got abouit 6 degrees below design, I am probably overpumped there too. (I do not even know if there is insulation under and around the slab (house built around 1950).)



Originally, that system had one zone with a large three-piece circulator. And domestic hot water was electric. It was not balanced correctly because the convectors upstairs were way to small for the water temperature used in the radiant floor. I tried slowing the water through the floor to force more upstairs, but before upstairs got enough, the downstairs did not get enough. Now there are 14' of Slant/Fin in each room up there, and they require around 3000 BTU/hr each at 0F outside, and design day is 14F, so I should not need much flow.



All that was replaced at the same time, the DHW with an indirect unit. So my electricity costs dropped, my fuel costs dropped. So in one sense, I do not have to do anything. If one of those circulators burned out, I would replace it, probably with a Taco 005-IFC, but as it is, I would go from 1/25 horsepower to 1/35 horsepower, and I think it would take quite a while to pay off that circulator. But I could calculate that. New system has outdoor reset.



Another thing I could do is remove the two zoning circulators and use a delta-T ECM circulator and zone valves instead. I suspect that would take even longer to pay off. Had I known about delta-T circulators, I probably would have specified one. Anyone know how long Taco circulators last? I suspect it is quite a while.



My return temperatures are quite low (I forget exactly what they are), but they are surely no hotter than the supply temperatures. For the radiant slab, I use water between 75F and 120F, although I do not recall it going over 110F. For baseboard I use temperatures of from 110F to 135F, and it may have gone up to 135F on a very cold day. I sure wish I could log that information on my computer, but there is no USB socket on my boiler for that. I do not feel like buying thermocouples or NTC sensors, A-to-D convertors, etc.

NRT_Rob

you're right

replacement wouldn't make any sense. it only makes sense if you are not putting the equipment in, in the first place, so it's not additional cost.



if you only have two zones though, and they must be two separate water temps, then I'd retract my statement... you may be overpumped, but you aren't wasting that much electricity, as you would need 2 pumps no matter what. 005's probably would have been adequate and worth using if you were starting from scratch, but you're definitely right not to replace already existing equipment.



though working emitter sizing to get down to one pump is fun and cool, it may not always be feasible either. sure is cool though.

Jean-David Beyer

I assume my (now former) contractor just uses 007 circulators

for everything because that leads to simpler inventory for the contractor. Like one size fits all for residential boilers.



I know they did not do much of a heat loss calculation, so I did my own. I got about 35,000 BTU/hr at 0F outdoors, even though design day around here is 14F. Cape Cod house with 700 sq. ft. downstairs and 400 sq.ft. upstairs. They recommended the W-M Ultra 3 105,000 BTU/hr to provide a margin of safety. That seemed nuts to me, so I got the smallest one in the line, the 80,000 BTU/hr one that modulates down to 16,000 BTU/hr. I would be happier if they made a 40,000 BTU/hr one that went down to 8,000 BTU/hr. It would not cycle so much when only the upstairs is running.



I have three zones, each runs at a different temperature (or range). Domestic hot water indirect gets 165F. Radiant zone gets 75 to 120F, and baseboard gets 110 to 135F. This is easily done with the W-M Ultra 3. If I were to have the piping redone, I think a delta-T ECM circulator and two zone valves would do the job. I do not see why I would not set that up for 10F delta T for both zones. But if it takes 23 years to pay off, that would be at the likely end of my life, so not worth it to me/.

NRT_Rob

cut flow in half

20 DT. beautiful stuff