I am in process of redoing the radiant heating system in my home. The house is two story 1930's bungalow in Minnesota (winter design temp -14F). With addition and finished basement, it's about 3200 sq ft.  Heat load was estimated at about 60K, but I may have been generous in estimating R-values in walls and attic, so I'm bumping it up 10%.

It will have radiators (for now) on second floor (heat load 20K); underfloor 1/2in PEX in extruded aluminum plates on main floor (heat load 36); and 1/2in PEX embedded in 4 in of cement in basement (heat load 10K). I will have three zones.

I have a conventional 120K Burnham natural gas boiler (about 84% efficient) and would like to run water through the two in-floor zones constantly during most of the heating season.  I'm currently re-reading Siegenthaler's two books ("Modern Hydronic Heating"  and  "Radiant Precision") for final design details, but I will probably use a primary circuit and inject hot water into two or three secondary circuits. I'll also add an outdoor reset and boiler (lower temp) protection that were missing on the system I inherited.

What are people real world experiences with adding a buffer tank? I haven't yet looked how much water the cast iron plates hold, nor have I calculated how much water my combined circuits hold. I am worried about my desire to always pump water through the two in floor systems will exacerbate the frequent cycling I had last year.  Right now I'm leaving the DHW supply separate, but I am thinking about using a 20 to 30 gal water heater as the buffer tank. 

Opinions? Please be kind as it's my first post.

Design temperature here is -15°F but hey you're new.

Great text books, but of little use to the layman. People show me their dog-eared copies all the time. Great for my friend Ziggy (and me), but just creates more questions for most.

If you have an accurate heat load done by a professional, a Mod Con boiler is the answer, with all the controls and options you will need. If you are stuck with over-size, low efficiency cast iron boiler, outdoor reset is your friend. My last upgrade saved the homeowner 15% with his re-conditioned 50 yr. old boiler.

Full time circulation is of little use here.

Buffer tanks are rarely used and then usually an expensive compromise.

Controls are the key, but experience required.

I agree that they are rarely used, but that doesn't mean that they are a bad idea. A good buffer tank is cheaper than a really good control system... doesn't do as much either, but it does solve cycling questions definitively. if that's all you need, that and cheaper control can make for a great system.

with only 3 zones and a boiler reset control, I would be inclined to be wishy washy about it... though I'd want to run as wide of a firing differential as possible to best leverage the mass of the boiler itself as a buffer, since your boiler is 2x the size it needs to be.

It rarely makes sense to strip out an existing boiler, but if you have tax credits or the like that might help, that might change things. certainly mod/con is where you want to be "in the end" anyway.

Posted By NRT.Rob on 26 Aug 2010 09:44 AM
I agree that they are rarely used, but that doesn't mean that they are a bad idea. A good buffer tank is cheaper than a really good control system... doesn't do as much either, but it does solve cycling questions definitively. if that's all you need, that and cheaper control can make for a great system.

with only 3 zones and a boiler reset control, I would be inclined to be wishy washy about it... though I'd want to run as wide of a firing differential as possible to best leverage the mass of the boiler itself as a buffer, since your boiler is 2x the size it needs to be.

It rarely makes sense to strip out an existing boiler, but if you have tax credits or the like that might help, that might change things. certainly mod/con is where you want to be "in the end" anyway.

I was thinking along similar lines- an Intellicon HW+ or Beckett Heat Manager type economizer with the high-limit aquastat on the boiler set to max may provide sufficient buffering to keep it from short-cycling itself into lowered efficiency and an early grave on zone calls by maximizing hysteresis on the boiler's mass when heat loads are low.  It would likely be beneficial with or without the mass of a buffer tank in the system.

With $1500 tax credit and $3-500 tax credit, I strip out old boilers all the time.

Coupling a buffer tank with a cast iron boiler is just a bad idea, a symptom of text book confusion, it is why they make instructors.

It is just not that difficult to dress up an old cast iron pig, but the ROI has to be kept in mind. You really can't change the fact that 20% of the fuel you buy will go up the flue, every time the burner fires.

Cycle time and hysteresis may save fuel but 15% fuel savings is good if you know what your doing and add ODR.

Then of course you could have the boiler de-rated if you knew the load.

with ODR, you still have to address cycling, and the mass of boilers in this size range do not provide what I consider to be adequate buffer capacity in most cases. Unless you have a particularly high water content.

there is a reason why the vitola blow away all other oil "cast iron" boilers in efficiency ratings. That reason is largely due to the 40 gallon buffer tank it built into its very well insulated construction.

cycling efficiency on cast iron is a much bigger deal than in low mass boilers.

This is a emitter/load question which can't really be answered without an accurate heat load and measured radiation. I like to slave zones when possible, but we are really not talking about a lot of money saved, certainly not enough to warrant a buffer tank and trim.

Thank you for the replies thus far.

About the professional heat load calculation, I did use HVAC-CALC, and was careful to measure and spec every space, window, and unheated surface in the house. Subject to some limitations, I think it was pretty good. Insulation settles. Infiltration happens after the best efforts. I didn't do a room by room blower test, nor did I do an infrared photo-shoot. It's a 75 year old house!

I intend to add more complex mixing controls, including ODR, but I'm trying to understand what you're saying here.
If I understand outdoor reset correctly, on my conventional boiler, with only thermostatic mixing controls, ODR would only lower the upper cutoff temp of the boiler in warmer weather. Is that right or wrong?

Even with more advanced controls, and again given my existing boiler, don't I still have too many BTUs being generated to not frequently cycle the boiler?

I'm also having a hard time understanding BadgerBoilerMN comment "Full time circulation is of little use here." The constant circulation is the design goal here, not a proposed solution to the boiler cycling. The whole point here is to reduce the temperature fluctuations (across time) of those aluminum plates and the wooden floors above them.

Even if I designed a state-of-the-art control system, including a modulating furnace, wouldn't I still want to
run fluid through the circuits at a constant rate?

Let me rephrase the question to allow a more general answer to my question:

1. Assume (stuck with) a conventional boiler,
2. Three heating zones
Z1. radiators (design heat load of 20K)
Z2. hard wood floors with PEX through extruded aluminum plates beneath (36K)
Z3. basement with PEX embedded in cement and very low heat demands (10K)

3. Assume my head load calcs were given to me by God.

With all other parameters and imaginary budget open, are there any conditions for which a buffer tank would help to reduce the cycling of the boiler?

Please, if you think that the question I've asked cannot be answered without knowing the color of the PEX
on the second floor bathroom, please explain exactly why.

ODR, resets the supply temperature to the outdoor temperature continuously.

You may need an two temperature system - which can also be done with piping and controls - but no one can tell from here.

The heat load only tells you what you need for output. You currently lack a design temperature for the three different emitters you describe. You still need to know what temperature to set the ODR (or buffer tank if you insist).

And then there is boiler protection.

All your questions I used to cover in the third day of a three-day hydronic design seminar for professionals.


All the answers you seek are in Siegenthaler's Radiant Precision.

Thank you BadgerBoilerMN,

Heh, are you located in St Paul? People keep talking about this guy around here, very knowledgeable on all things radiant (all good things).

I have left much out of the discussion for brevity--dangerous when treading in a new area. I will be using ODR and boiler protection. I have calculations for BTU requirements for each room and a reasonable estimate of flow and water temps needed to get there. I admit I still have a ton to learn.

I would almost go for new boiler, but wouldn't I just about have to start from scratch (again) on the controls when I tear out the radiators on the second floor in two years?

For now, I have to tear down about 1200 sq ft of poorly installed PEX, staple up those plates, and insulate before the cold weather returns. Plus a new mixing system. I thought I would leave the boiler and second floor alone for this year.

If you start with a rebate and tax credit $2000.00, a condensing boiler sized to the load and a near piping plan, the manifold can easily be modified to serve any residential emitter you may have now, or choose in future. Domestic hot water with an indirect of course.

You get the tax credit, comfort and start saving fuel now.

I work in St.Paul a lot.

if you can get all your emitters to one temp, going modcon would also save you several hundred dollars on a proper mixing system. another point in favor there.

with constant circ though I think I would almost always advocate for a buffer tank. Your load can vary from near zero on up. bang/bang could use the water capacity of your zones as a buffer, but not constant circ... either your flow is too low or bypassed to provide for adequate heat extraction for buffering.

you can do this cheaply with a home cheapo 10 gallon water heater, if 3/4" tappings are adequate. Otherwise there are some nice buffer tank options out there that aren't too bad cost wise.

There hasn't been a post for awhile now, so I wanted to thank everyone for the replies before the thread is closed.

I will probably go with a small buffer tank and the existing boiler for now. I did a couple of the cost-benefit analyses, including the fairly detailed spreadsheet version put out by the DOE. Given that existing boiler is less than 10 years old, the return is just too low to rationalize the purchase, even with the tax credit and other local incentives.

I may have gone for the new modcon boiler anyways, just to simplify the design; but with the old radiators still part of the mix, a mod-con boiler would still have to provide the high temp water whenever our second floor (radiators) called for it. True, with the outdoor reset the temps could be lower, but still much higher (for a given outdoor temp) than the two radiant circuits. We would be back to cycling again, albeit high-low rather than on-off.