Tankless vs Tank for Heating & Domestic Use

[Canada_Dave]

I need a system to do both hydronic heating as well as handle domestic water needs in my cabin. I plan on being there occasionally -- more in the summer but in the winter only once every couple of months. It is in a four-season area where temperature ranges from -40 deg. F to 86 deg. F.

Some of the tankless systems say they can handle this. Other option is to go to a boiler system and/or tank system.

What is my best option for my situation. Please include makes & models.

[Gary Swart]

This question is one of the most controversial topics on this forum. There are some who swear the tankless are the greatest thing going, but many of us cite numerous drawbacks to them. I will not attempt to present the arguments, the are found in this discussion area. I will say a prime consideration for you is the winter temperatures of the incoming water. This is assuming you're located in Canada. There are other major concerns as well that you can read aboutl

[Canada_Dave]

I have read some of the other threads arguing both sides of tankless vs tank but none (at least none that I read) cited their uses/conditions. Most posts seem to support one vs the other. I was wondering if under my specific conditions/uses, if there was a certain route I should be going. In all actuality, I would like to get a tankless supporter listing me the best make and model for my situation and a tank supporter giving me the best make and model for my situation. I can then compare the 2 options (cost, efficiency, maintenance, longevity of testing, etc). I am completely open to either option and I would just like to get professional opinions on what to compare.

[SewerRatz]

I have read some of the other threads arguing both sides of tankless vs tank but none (at least none that I read) cited their uses/conditions. Most posts seem to support one vs the other. I was wondering if under my specific conditions/uses, if there was a certain route I should be going. In all actuality, I would like to get a tankless supporter listing me the best make and model for my situation and a tank supporter giving me the best make and model for my situation. I can then compare the 2 options (cost, efficiency, maintenance, longevity of testing, etc). I am completely open to either option and I would just like to get professional opinions on what to compare.

For combination heating and hot potable water, your best bet is a tankless style boiler. The warranty will be much better and the units are designed to handle that kind of demand.

[chris8796]

I couple of questions I would want to know:

1. What's the estimated heating load (btu/hr) and DHW loads?

2. What's the fuel source and any btu/hr limitations?

3. How will you maintain the cabin in the winter when not in use, specifically the freeze protection of the plumbing?

[Dana]

Unless it's super-insulated most tank HW heaters won't be up to the heating load. There are a number of tankless units that can crank out ~150-170btu/hr if set up correctly, but it's important that the combi-system be designed to be able to handle the actual load, which takes some design skill. In practice, unless you can get sub-90F return water from the radiation it's hard to get more than about half the full rating out of them unless you're willing to exceed the manufacturer's recommended continuous flow rates (not that THAT's stopped many internet-purveyors of generic tankless-based radiant systems. :-) )

I'm comfortable designing-in a Takagi tankless combi for my ~30KBTU/H heat load, but would need to go to slab-radiation &/or an air handler coil to get 3x that out of it and still stay within the 2gpm max continuous flow recommendation on the Takagi. At 2gpm or 1000lbs/hr x 60 we get 60kbtu/hr. The max output temp on my Takagi KD20 is 176F, so I'd need 116F return water to double it's output- doable without great pains, but to get 90KBTU/hr out of it we'd need either return water from radiation of 86F (hard to do on peak-load day without superinsulation & or/slab). The alternative is boosting it to higher than recommended primary-loop flow, and suffering higher internal wear on the heater (not to mention higher pumping power costs- them suckers have a fairly high head load at 4-5-6gpm & higher.)

But first things first- do a manual-J type heat load calc on the place, figure out how much space-heating load you have on the coldest day of the year and get back. If it takes a boiler, it takes a boiler, I'm sure there's some optimal solution out there. (The hot-water side is easier to figure...) Unless your calc comes in under 75KBTU/H you're probably better off going with a boiler solution. If it comes in under 25K you can probably do it comfortably with a tank HW heater with it's thermostat cranked up a bit. In-between, there are tankless (and some larger tank burner) options...

[jadnashua]

Many tank type WH and tankless WH are not designed for extended, continuous operation, as may be required for space heating. Then, since you wouldn't want to be circulating potable water to heat the house, that means a heat exchanger, and the losses you get from that. Most modern boilers don't maintain a set temperature like the old ones did. This means that if there's no load, the water in the loop cools off. To get decent response rates, they have very little water in them so when it does need heat, it can heat it fairly quickly. A boiler with an indiretly fired tank for potable use (basically, think of this like a dedicated separate zone that heats the water in the tank) can supply all the hot water you need, if sized properly.

Boilers these days can range in efficiency from about 70% up to nearly 100%. The better ones have logic to do things like freeze protection...regardless of where you set the thermostat, if the water in the boiler gets lower than a set value, it will turn on to prevent things from freezing (the boiler, not the rest of the house). You could probably add that if it isn't built in. Some also adjust the output to match it to the load (the most expensive, but the most efficient, too - may not be worth the cost for your application).

[Canada_Dave]

For combination heating and hot potable water, your best bet is a tankless style boiler. The warranty will be much better and the units are designed to handle that kind of demand.

Which tankless unit do you recommend?

Units that I have looked into are the Navien cc210 and Noritz 93 which should both adequately handle my needs.

[Canada_Dave]

I couple of questions I would want to know:

1. What's the estimated heating load (btu/hr) and DHW loads?

2. What's the fuel source and any btu/hr limitations?

3. How will you maintain the cabin in the winter when not in use, specifically the freeze protection of the plumbing?

I am not sure of the heating load not the DHW load. I plan on heating a 1150sq ft basement slab and a 600 sq ft garage with it. DHW needs will be a single bathroom off the start but with a second bathroom somewhere in the next decade.
The fuel source is natural gas and there aren't any limitations.
The cabin will be maintained at a minimal temperature in the winter when not in use by using the combination of heated slab and forced air furnace. Due to not being around much in the winter, I am going to go with glycol in the slabs.

[Canada_Dave]

Unless it's super-insulated most tank HW heaters won't be up to the heating load. There are a number of tankless units that can crank out ~150-170btu/hr if set up correctly, but it's important that the combi-system be designed to be able to handle the actual load, which takes some design skill. In practice, unless you can get sub-90F return water from the radiation it's hard to get more than about half the full rating out of them unless you're willing to exceed the manufacturer's recommended continuous flow rates (not that THAT's stopped many internet-purveyors of generic tankless-based radiant systems. :-) )

I'm comfortable designing-in a Takagi tankless combi for my ~30KBTU/H heat load, but would need to go to slab-radiation &/or an air handler coil to get 3x that out of it and still stay within the 2gpm max continuous flow recommendation on the Takagi. At 2gpm or 1000lbs/hr x 60 we get 60kbtu/hr. The max output temp on my Takagi KD20 is 176F, so I'd need 116F return water to double it's output- doable without great pains, but to get 90KBTU/hr out of it we'd need either return water from radiation of 86F (hard to do on peak-load day without superinsulation & or/slab). The alternative is boosting it to higher than recommended primary-loop flow, and suffering higher internal wear on the heater (not to mention higher pumping power costs- them suckers have a fairly high head load at 4-5-6gpm & higher.)

But first things first- do a manual-J type heat load calc on the place, figure out how much space-heating load you have on the coldest day of the year and get back. If it takes a boiler, it takes a boiler, I'm sure there's some optimal solution out there. (The hot-water side is easier to figure...) Unless your calc comes in under 75KBTU/H you're probably better off going with a boiler solution. If it comes in under 25K you can probably do it comfortably with a tank HW heater with it's thermostat cranked up a bit. In-between, there are tankless (and some larger tank burner) options...

I am not familar with doing my own J type heat load calc so I took my plans to a system designer and they recommended a couple of tankless units that would do my job but I didn't ask for the calculations used to determine the units they specified. Is this easy to calculate by myself?

[Dana]

I am not familar with doing my own J type heat load calc so I took my plans to a system designer and they recommended a couple of tankless units that would do my job but I didn't ask for the calculations used to determine the units they specified. Is this easy to calculate by myself?

There are various heat-loss calculator freewares out there, but this $49 package seems to get good reviews:

http://www.hvaccomputer.com/

(I haven't used it myself- can't comment on how good it really is.)

From a system design point of view, having some buffering as part of the potable water heat exchanger saves the tankless from wear & tear by minimizing the number of on/off/flue-purge cycles, and will improve the overall operating efficiency. This architecture using a "reverse-indirect" hot water heat-exchanger basically works and has some forgiveness to the design, and reduces corrosion & scale issues in the tankless:

http://www.heatpro.us/designtree/doc...anklesssys.htm

A limitation of that architecture is that the return water isn't less than ~10-15F below the storage tank temperature, which has to be ~125 or above to get reasonably hot water out of the potable water side. This, A: limits the amount of condensing a condensing tankless can achieve (it's still pretty good though, if you can design the radiation to be able to deliver the heat at 130F.) and B: With only 115-120F return water the delta-T you can get out of the tankless, so you may need to run higher flows. (I'm not sure what the recommended max continuous flow is on the bigger Naviens & Noritz, but it's probably more than the smaller/older Takagi I've designed with.)

Another advantage to this approach is that it eliminates DHW flow and tankless "cold water sandwich" issues, since the flow is limited by the heat exchanger's plumbing, not the high-head of the tankless, and since the DHW is drawing heat from the buffer water, it never "sees" the slug of cooled water in the heat-exchanger that is created by flue purges at the end of a burn cycle.

The same guy who sketched out the tankless combi above has a hyronics primer worth absorbing, if you're committed to designing the heating system yourself:

http://www.heatpro.us/designtree/

He also has a heat-loss calcualtor freebie (and a pro version) using IBR methods (good enough for your purposes.)

http://www.heatpro.us/downprog/

(Again, haven't used it, can't really comment on it.)

There are cheaper combis out there using air handler coils, no buffering. I like the one they did with a Rinnai here:

http://dsp-psd.pwgsc.gc.ca/collectio...2-106-108E.pdf

which is similar to (but perhaps better-designed than) the one they did with a Takagi here:

http://www.toolbase.org/pdf/fieldeva...luationSWD.pdf


Using potable water in the heating system (as in the Florida combi) is in general a bad idea (legionalla growth, more corrosion, wear & scaling issues on the tankless heat exchanger, etc.) Even if unbuffered, use a heat exchanger, and keep the tankless/heating loop at at least 12lbs pressure to A: keep efficiency-robbing sizzle out of the tankless heat exchanger, and B: keep oxygen out of the heat-loop water to limit corrosion on pumps, radiation, & burners.

Think about what you want/can-afford for radiation (radiant floors/walls/ceilings are the most comfortable and provide the lowest return-water temps, but are also more expensive. Fin-tube baseboard or an air-handler coil might be the cheapest. But low temp panel-radiators are also quite cushy, and somewhere in-between cost-wise. )

Navien sells a handy space-efficient pre-packed heat exchanger kit for doing combi-systems called the "Heating box" (http://www.aqenergy.com/ ) but I've recently read about a few longevity issues with it. They will probably fix those issues and stand behind the product, but for the money you might be able to do better designing your own (if more bulky) system.

Whatever you do for a tankless combi system plumbing it as a primary/secondary allows you to set the tankless & radiation flows independently of one another. Trying to run it all on one pump limits the amount of modulation (and efficiency) you get out of the tankless while in heating mode, and takes a much larger pump.

In the buffered reverse-indirect model promoted by the HeatPro guy, the tank behaves as the hydraulic separator in the primary/secondary, and allows you to run the tankless at low flow/big-delta-T with tankless output temps much higher than the radiation temps. This buys you a lot more BTUs/hour out of the tankless without resorting to a 300watt pump, and without overstressing the tankless. The radiation temp is set by the aquastat on the indirect- the output temp of the tankless is whatever you program it to be, based on flow and anticipated max BTUs/hr you need. You move 1000lbs/hour for every 2gallons-per-minute you pump through the tankless. Multiply the lbs/hr times the delta-T on the tankless to get the BTUs/hour output. No matter how big the tankless, if you're pushing more than 4gpm through it, it's gonna take over 100 watts of pump. (over 200W on some tankless units.)

Even though specs for various tankless will give you "100 F rise at 6gpm" or something, don't assume that the unit will last very long if run continuously at those flow rates. It's one thing to run high flow rates for 20minutes/day for showering, quite another to run 10 HOURS per day running a heating system. Figure out from the documentation what the MINIMUM flow rate is to guarantee ignition, and if you can get the BTUs out of it that you need with flow rates merely 2-3 times that you'll be much better off in the long run.

If you're hell-bent on designing this yourself rather than hiring a pro, read up on it- a LOT. There's a huge amount of info on the web. PM Engineer mag (http://www.pmmag.com ) has a lot of info and articles, many by an academic John Siegenthaler are VERY good, eg:

http://www.pmmag.com/Articles/Column...00f932a8c0____

The "inside out tank" example is the same topology as the HeatPro combi system (substitute a tankless for the boiler.) But I'd recommend adjusting & simplyfying that schematic slightly, with the boiler loop pumping toward the tankless or boiler, not the tank, since pumping away from the boiler lowers the pressure in the heat exchanger making it more likely to form micro-bubbles that sizzle and reduce the heat-exchange efficiency. The pumps for the radiation should also pump toward the radiation, not the tank, as shown. The ErgoMax versions of the indirect have separate heating system sources/returns, distinct from the boiler loop, which simplifies the external plumbing considerably. If you can run your radiation under 140F, you won't need the complexity of mixing valves, etc either- just put a tempering valve on the DHW out to keep the scald-risk under control.

Also, you may be able to further simplify by slaving the tankless loop to the buffer-tank's aquastat only rather than firing whenever there's a thermostat call, letting the thermostats run the secondary pumps only. The primary will start up once the tank's drops below the setpoint, by then a slug of cooler return water will be in the bottom of the tank, and the modulation of the tankless will be able to adjust the flame according to the return-water temp. During simultaneous heating & DHW load will super-chill the bottom of the buffer (32-50F water entering the heat exchanger makes for the largest delta-T and greatest heat transver at the bottom of the tank), resulting in even colder boiler-loop return temps, without affecting the temp of the water entering the radiation, since the top of the tank where radiation water is source is being filled with water at the tankless output temp (which is constant). But the tankless will be firing higher to achieve that output temp with the cooler return water it is getting from the tank due to the DHW load. With the buffer setpoint and tankless setpoint temperatures optimized this can be a quite efficient system that never short-cycles, and runs near the tankless' max thermal efficiency. (Short cycles KILL the efficiency of tankless heaters & boilers- design it keep all burns longer than 10 minutes if you can, or at the very least greater than 10 gallons pumped through the tankless. The EF rating number on the tankless is based on multiple 10.2 gallon draws, not 1-2 gallon draws, so it's often overstated for "real world" situations where startup and purge cycle losses totally trash the operational efficiency of even condensing units to something less thatn 50% efficiency).

OK, too much information, I know... :-) (Sorry, I just designed one of these for my own place- I can't help myself. <LOL>)