Oversized Boiler, maybe?

Background - Cincinnati Ohio - 1.5 story cape cod, built in 1951. Brick on block construction, minimal (rockwool) insulation in the knee-wall space. Newer replacement windows. 2450 sqft. Radiant heat in the ceilings of the first floor and slanted ceilings of the 2nd floor. Single zone, though there is a valved manifold in the basement. Did some math based on the amount of water in the system and it appears that I have around 4K feet of 3/8 copper throughout the house. Boiler temperature when running never gets above 120F.

Current boiler is a Dunkirk XEB-4 (21 years old) 90K BTU output. At last checkup it was running @ ~80% efficiency. It's rated at 112K BTU.

I did a heat load calculation ( https://www.greenbuildingadvisor.com/article/out-with-the-old-in-with-the-new) and came up with 65K BTU @ 65 degrees (at 1.4x oversize). Am I wildly off base to think that yes, my current boiler is nearly half again bigger than it needs to be? And that would explain our rather high (275CCF) gas usage when we keep the house at a balmy 67 degrees in the winter?

If the boiler is oversized and I decide to replace it - mod-con boiler worth it, or do I go with another cast iron guy? We're not planning on moving anytime soon.

Thanks,
John

 

What water temp are u running? What type of heat emitters are u using? Fin tube, radiators or fan coil?

 

Boiler runs at 120. There's copper tubes in the embedded in the ceilings of the house - essentially making the ceilings radiant panels.

 

That would be a load of (65K/1.4x=) 46.5K at +12F (Cincinatti's 99% outside design temp), not at 65F. I presume you were (correctly) using 65F as the base temperature (= the balance point temp at which there is generally no heating or cooling load) ?

So a 112K-out boiler would be 112K/46.5K = 2.4x oversized. AFUE has a presumptive 1.7x oversize factor, and is usually very close to the steady-state combustion efficiency, even with the jacket & distribution losses supporting the heat load (as when the boiler & plumbing are inside the thermal envelope of your house.) At 2.4x you're on the knee of the curve, but haven't fallen completely off the efficiency cliff yet. The as-used AFUE is probably at least 75-78% if the raw combustion efficiency is measuring 80%.

With low temp radiation and low temp water it is going to favor a (more right-sized) mod-con for both comfort & efficiency. Running a cast iron boiler at 120F usually shortens it's life considerably, so it's likely the current system is being mixed down prior to the radiant, but a mod-con will hit the mid-90s efficiency at that temp, and will use only ~78%/95% = 82% as much gas, an 18% net savings.

If the boiler room is in an uninsulated basement it may be cheaper to air seal & insulate the basement walls to the current IRC code minimums (= R10 continuous insulation in IECC zone 4, which includes Cincinnatti), possibly for similar or less money than a mod-con. A 2" shot of HFO blown closed cell polyurethane foam + fire retardent paint runs about R13 for $2.50-$3 per square foot.

 

Surprised with system running a low water temp for a ci boiler you haven't had flue and burners rusting out. With being oversized = short cycling ,using inside air for combustion efficiency 70%. Condensing boiler running return water temps under 130* is where they get the 94% + efficiency and with modulation can track water temp closer and not short cycle. How many different loops does the system have brand and model of pump.

 

Dana - thanks for taking the time to respond. And yes, I was using 65F as the base temp. HDT was 12, as you indicated.

Basement is semi-finished, though the one external wall in the boiler room is bare cinderblock, so maybe it would be worth it to insulate that side?
The basement is heated, both by floor mounted radiators and the 30ft long 2 inch copper manifold that feeds the rooms (18 valves!). It doesn't appear that the system is being down-mixed prior to the radiant. There's a make-up/one way valve from the domestic side, but that's on the return. In the knee wall there is another manifold and expansion tank, but nothing appears to be connected to domestic water.

And that's where my concern has been - at 120 water temp (both the aquastat and the temp/pressure gauge on the boiler show this) I am worried about the life expectancy of the boiler. I mean, it's been running for 21 years (I assume) this way.
My other issue is finding a company that doesn't scratch their head every time they see this system. They generally say, 'well, if the house is staying warm at 120, it should be fine.'

If I went mod-con, what would you recommend?

Thanks,
John

 

It's a taco 007-F5. As for loops...each room/hallway/area has it own loop. I think. The manifold in the basement has 18 valves that correspond to the various areas of the house. Closing a valve causes the loop for that room to go cold.

John

 

With high mass radiators and a low load it probably won't really short cycle, but the duty cycle will be extremely low. If the boiler output temp is 120F it's raw combustion efficiency will be over 90% until it's so corroded that the turbulence inducing structures on the fire side of the plates have corroded completely away. I suspect the real temp on the boiler is much much higher than 120F if the last combustion analyzer test showed 80% steady-state.

Absolutely yes! The boiler room is probably the warmest room in the house during cold weather, and has a high heat loss at any temperature. Cinder block walls are notoriously air-leaky, and have an R value of about 2.

Spray foam contractors would have to charge quite a lot- the set up and break down time is the same whether they're installing 200 board feet (10' x 10' x 2") or 2000 board feet.

If you want to DIY it on the cheap it's possible to get there with 1.5" foil faced polyisocyanurate board (R9-R10) strapped to the wall with 1x4 furring through-screwed to the cinder block with TapCons (be sure to use washers so the furring doesn't split), hanging 1/2" wallboard on the furring. That adds a bit less than 3" of thickness to the exterior wall assembly. Be sure to observe the manufacturer's clearances to walls/combustibles, and if the vent stack is on that exterior wall, the code required clearances from the vent. (It's sometimes OK to insulate right up to vent pipe using rock wool, but never fiberglass or foam.)

Start by putting the foam board on the wall temporarily with blobs of foam board construction adhesive (available at box stores), keeping the bottom edge of the foam a half inch off the slab (where it could wick moisture if the slab were damp). Tape the seams with a high quality foil HVAC tap (Nashua 324A is available at most box stores). Seal the top edge of the foam to the foundation with polyurethane caulk or foam board construction adhesive. It's also important to air seal the foundation sill to the foundation, and the band joist to the foundation sill & subfloor, etc. Sometimes that's more easily done with combinations of cut'n'cobbled foam board & small DIY spray foam kits like FrothPak. Polyurethane caulk is the better material for sealing wood to wood, or wood to concrete/cinderblock for gaps under 1/2".

The final choice really depends on your local manufacturer & installer support there is, but...

Small fire-tube type mod cons are pretty easy to retrofit into cast-iron replacements, since they can usually handle the higher flows and can be pumped direct. The lower the minimum-fire input the better, independently of the max firing rate. There are several fire tube mod-cons with ~80KBTU/hr max firing rates that can throttle back to 8K. The value leader in my neighborhood is HTP's UFT-080W- it's cheap, reliable, and comes pre-plumbed with a secondary port for running an indirect water heater. (The Westinghouse WBRNG080W is the identical equipment under the paint, sometimes distributed through the big orange box store.) Lochinvar's KHB085N or WHB085N are pretty similar in modulation range, and have a lot more bells & whistles, capable of programming different temperatures for different zones, etc, but it usually quite a bit more expensive. (The KHB055N would cover your likely load, but has about the same minimum firing rate as the KHB085N, and would be slower for heating an indirect.) There are others.

Navien's NHB-80 water-tube boiler is usually competitively priced and can work here too, but MUST be plumbed primary/secondary (with an extra pump) due to the high pumping head of the water tube heat exchanger. It's somewhat more difficult to install, and more sensitive to the competence of the system designer/installer, but done right they have a good track record, with pretty good distributor & installer support in my area. There are other water tube boilers with big turn down ratios that would work too.

A Taco 007-F5 uses nearly 10x the amount of the drop-in replacement Taco 007e that uses a high efficiency ECM drive DC motor, which will pay for itself many times over the lifecycle of a pump- in high electricity price areas it pays for itself every YEAR! They run about $140-150 at internet pricing. There are also smarter pumps with programmable ranges out there that would work too, offering a bit more flexibility to how the system is operated, starting at ~$200 that might be considered. Even if using primary/secondary configuration with a water-tube boiler, replacing the -007 with a Taco 007e or Taco VR1816 (smarter pump) would be "worth it". Alternative similar pumps like the Grunfos Alpha2 15-55F or the AquaMotion Einstein AM55FVL would work too.

As long as the 18 port manifolds isn't 18 separately controlled micro-zones, each with it's own thermostat it's fine to pump a fire-tube boiler directly. If there ARE 18 thermostats even the fire-tube boilers would likely need to be plumbed primary/secondary to meet the minimum flow requirements of the boiler under all conditions.

 

Thanks again, both to Dana and Fitter30 for answering my questions.

Looks like I'll be insulating a wall this spring at the minimum, and looking into the HTP as a boiler replacement. I gather the UFT-080 with an indirect tank is the way to go, like @PC7060 did in his setup?

Interesting about the taco - I hadn't realized how inefficient it is. There's only a single thermostat in the house.

So all this talk of failing boilers due to low temp has me worried - the honeywell aquatstat (L7224U) shows 120 on the display for water temp - then again, it also shows err 10 but there's no outdoor reset connected to it.
The gauge on top of the boiler shows 105. My multimeter shows 101 just past the aquastat when placed on the copper output line, which matches up with what my thermal temp gun shows. Return water temp shows at 83F right before the circulator.

A thermal camera show of my living room ceiling shows this - the runs for the LR are closest to the boiler - not sure if that makes much of a difference...

 

It's a pretty good setup, if you have reasonable local support for HTP. (My house is less than an hour drive from the company HQ- support it quite good near me. YMMV.)

Think of ECM drive smarter pumps as the pump equivalent of LED lights vs. incandescent. Like high quality LED lighting of the era ECM drive pumps were still crazy expensive a decade or so ago, but not a huge cost adder in 2021. If the boiler & pump were installed 21 years ago suitable ECM drive pumps weren't even available. The standard -007 has been an industry workhorse for decades, is highly reliable and has a good track record, but I'll never install another one, just as I'll never install another incandecent light bulb anywhere other than in the oven.

I'm not sure how you take a temperature with a multimeter(?). Be aware that IR thermography has inherent assumptions about emissivity of the objects in the image. Bare copper is very LOW emissivity, and will always measure lower than reality. I put splotches of paint (any non-metallic color) on copper plumbing to be able to use IR thermometers reasonably when trying to analyze a system.

The most important place to measure is on the return plumbing where it enters the boiler. A properly designed radiant floor system with a cast iron boiler will have a bypass branch mixing direct boiler output with the return water just before it enters the boiler to mitigate condensation risk. Any measurement point on the system side of the return path from that branch connection will read lower than what the boiler is actually receiving.

On p.11 of the XEB series manual they specify a minimum entering water temp of 130F, and in Figure 6 show a rudimentary drawing of one type of bypass branch referred to as a "boiler bypass", which is one of the more appropriate solutions for your low-temp radiation system. Take a look at the near-boiler plumbing and sketch it out. A pump located inside the bypass loop pumping toward the branch point to the radiation (as in Figure 6 of the manual) is equivalent to being inside the bypass loop pumping toward the boiler, with the inlet to the pump pulling from the other branch point as shown here:


There are other configurations where the pump is on the radiation side of the branch, which behaves somewhat differently, yet still provides some protection.


An even safer approach is to use primary/secondary piping, with a second pump to drive the bypass branch, using closely spaced tees as the point of hydraulic separation between the low temp radiation flows and the higher temp boiler flows:



See if your boiler installation has any similar features.

 

My multimeter has a tape on thermistor probe. I that used to measure the return temp - I shot the temp gun against the black duct tape I used to adhere the thermistor to the return line. I'm reasonable confident that the return temp is not at 130f. There's no boiler bypass on my system either - just a make up line from the domestic water to the return side of the system - CH out is the copper in the photo, and the tan/white line in front of the expansion tank is the CH return.



Funny how none of the service tech who came out for service ever mentioned any of this.

Well, the multiple CO detectors in the house haven't gone off yet, and the one in the boiler room shows '00' in the display so I guess the boiler won't kill me. Yet.

With 21 years of improperly installed cast-iron boiler, should I start planning for a replacement?

John

 

Looking at dunkirk xeb boiler manuals there all power vent looks like single wall flue pipe.

 

Even if it were installed properly it's always good to have a plan when the existing boiler is north of 20 years, so yes. When and how you execute on that plan has flexibility as long as the boiler isn't leaking and is still firing up reliably, etc. A reasonable expectation for service life for a cast iron boiler is about 25 years, give or take. Even though they might still be running OK decades beyond that, reliability eventually sufferes, and the combustion efficiency will eventually start to fade.

If the 80% measured combustion efficiency was tested with an entering water temp south of 120F rather than the nameplate 84% for the XEB series means it's already slipping. At temps that low the raw combustion efficiency should exceed 84% by quite a bit, even edging north of 90% until the excess fire side corrosion has started to become excessive (which it probably has.)

Low combustion efficiency due to worn/corroded heat exchanger plates doesn't affect the carbon monoxide levels at all, it just raised the net stack temperature due to less heat being transferred. As long as it's not starved for combustion air and the chimney is drafting fine the CO levels in the exhaust can still be well below 100ppm. The CO levels in the room won't go up unless the heat exchangers is sooted up or otherwise restricted creating a flame roll-out problem, or if the chimney isn't drafting well because it has become restricted from squirrels nesting in the flue or some such.

Some pictures of how that supply & return plumbing is hooked up to the boiler & manifolds might be useful.

 

It's straight shot, so to speak - supply side is 1 1/4 copper out of the boiler and the return is...well another 1 1/4 pipe that is fed from a run in the ceiling. I believe it comes down from a manifold in the knee-wall space on the 2nd floor. It's all in parallel.



I was trying to be funny with the CO comment - the XEB manual has big warning banner about improper return temp causing flue blockage - but the boiler hasn't killed anyone in 21 years. That I know of, at least.

This thread degraded into me complaining about my boiler - thanks for taking the time to hear me out.

Of course, I then went down a internet hole of trying to understand why the system is behaving the way it does so I can plan properly for a replacement. Not knowing exactly how much copper is buried in my ceilings makes trying to figure out the head and flow rate needed (or what is currently being delivered) rather...difficult.

Am I wrong is assuming that the current circulator is delivering too much flow for the current system - which appears to be 18 parallel loops of 3/8 copper of indeterminate length? All I do know is that the system when drained down hold 35 gallons of water. I did some math to figure out the total length of 3/8 copper, which came out to ~4K feet. I think I'll need roughly measure each valved loop individually at this point.

For this formula:
GPM = 0.002*BTU/(Temperature Drop, F) - temp drop in my system is 10f - the heat load I did earlier is ~45K BTU - which gives me 9GPM. I'm not sure if that high or low for a 3/8 piped system?

I'm sure I'll have more questions as this project progresses.

John

 

Without the boiler in the picture it's impossible to see how it's all hooked up.

If the entering water temp (EWT) of the boiler is too low there is a greater chance for the heat exchanger to soot up as well as close down some from corrosion. If that's happening you'll get flame rollout by the burners, which is both a CO hazard and fire hazard. Like most boilers that vintage I believe the XEB series has a rollout switch to kill the flame if that's happening, but some (soon to be brain-dead) geniuses jumper across the switch to keep it running after it trips a few times.

A delta T of 10F on each of the 3/8" piped branches to radiation would be fine, about right for a low-temp system. If <<5F (120F out, 115F or higher return) that radiation loop may be overpumped, but may have been flow adjusted to better match radiation output with the load. If you're measuring just the manifolds or supply/return pipes to the boiler you're getting the average. But the basic math is right- 9 gpm = ~4500 lbs/hr, x 10F= ~45,000 BTU/hr. With 18 loops (18 supply, 18 return) that would be an average of 0.5gpm per loop, but it's likely some are tweaked higher, others lower.

 

Thanks again Dana. I have a company that does hydronics coming out next week to give me their thoughts, though I am leaning towards doing the boiler swap myself. To that end I started looking into the HTP UFT-080W that everyone seems to love - and well, my local distributor says it has been discontinued. A call to HTP verified that - no more UFT units at all. The replacement is either the ELU or EFTU series.

John