[an error occurred while processing this directive]
Posted by dana on May 12, 2008 at 18:21:31:
In Reply to: Re: Is historical usage data better than Manual-J for sizing boilers? posted by HeatPro on May 06, 2008 at 17:45:08:
: : In the current configuration the basement stays so warm (~64F) that the floor is ALREADY 62F or higher,
: ++ Through insulating the full basement, it will still not require much heat down there.
: : Even prior to heat-loss modeling it's pretty clear that the structure won't need more than 50,000BTU/hour on design day
: ++ With a 50,000 btuh load, a high-efficiency water heater is affordable or a T-K3. The difference between 85% and 92% isn't enough to pay some thousands for the extra $130 a year savings.
: ++ Water heaters like the above have settings to go no higher than 150F.
: : I haven't measured the return water temp directly
: ++ Do measure the supply and return water temperatures in each loop so you know.
: : When the smaller radiant floor zone kicks on, the burners cycle 90-120 seconds on, gets up to ~180F, then 20+ minutes off, pretty much as-before.
: ++ Very fast cycling.
The basement isn't intentionally heated- it's the losses to the basement from the too-large boiler loop to the radiant floor plus the short-cycling that keeps the basement so toasty (I'd rather it were 5-10F cooler, eh?)
With the aquastat on the high side of the AH coil set to 140F and the high-limit on the boiler set to 180F, when that zone calls for heat the boiler fires, heats up to ~140F, the AH starts and return water cools a few degrees. Within 30 seconds the return water recovers to 140F (delta T is about 15F), and over 6-7 minutes the output ramps to 180F with increasing delta-T (return water reaches 150-152F, delta-T is about 30F). The high-limit turns the flame off, then in under 3 minutes the temps drop to a return water temp in the low 130s with a delta of ~14-15F, the fire re-strikes and the heat ramp begins anew.
The average return water temp is in the low 140s, the average output water temp is around 165-170, so I doubt it's condensing (much) at least not in the heat exchangers.
It's not clear to me if letting it cycle like this is more or less efficient than letting it come up to 180F+ and letting it blast away until the zone thermostat is satisfied. Are there more losses from the short-cycle (6 minute burns in rapid succession) than gained in the lowered return water temps?
Also, in-re on-demand HW heaters, are they really as efficient as 87% when operated in this mode? Seems like they're designed to raise 55F water to 115F at highest efficiency, not take 80-90F return water up to 130-140F. Are hot-water heater efficiency rating truly equivalent to 140F output AFUE tests or do they mean something different? (I suspect different, for good or ill.)
Also at lowest fire I would assume lower turbulence of the flue gases on the heat exchangers and the stack temps start go up rather than down (could be wrong, often am.) Since it's nearly impossible to perfectly balance the heat load to the firing rate (and the firing rate would have to be almost arbitrarily low in my case), wouldn't it be more efficient to run the low-temp radiant floor zones off multi-speed pumps tied to a buffer tank and let the boiler (or HW heater) heat serve only the tank, with hi/low limits on the tank set such that the burner runs it it's most-efficient mode? There would be small standby losses, but with 15-25F deltas on the high/low limit and low operating temp there's sufficient mass in an 80 gallon tank to ensure a reasonably long firings in an efficient mode rather than the ultra-short and inefficient bursts my current system sees, or very long burns at diminished efficiency from a modulating boiler. (Better a single 15 minute burn at or near highest efficiency than 15 bursts of 2 minutes with random timing for sure!)
Just trying to think ahead, planning the system that drives the upcoming zoned radiant floors...