You might consider putting small electric water heaters at the point of use – under a bathroom sink or kitchen sink for example. That way for short demands of hot water you are not wasting ½ gal or more of water filling the feed lines with hot water.

You might consider putting small electric water heaters at the point of use – under a bathroom sink or kitchen sink for example. That way for short demands of hot water you are not wasting ½ gal or more of water filling the feed lines with hot water

Posted By strategery on 04 Jun 2012 06:56 PM

The standby loss of a condensing tank is quite small- about the same as with electric tanks. This is in contrast to the standby losses of atmospheric-drafted  type gas fired tanks, where the center-flue heat exchanger constantly convects heat out of the tank 24/365.

I wonder why this is? I think I understand how the Vertex condensing unit works. It has a really long coiled heat run that distributes the heat more evenly? But since it USES more of the heat it gets by burning natural gas, wouldn't it also only need to run for a very short period of time causing some lowered efficiency? I'm just thinking about the unit keeping the water at the desired temperature for the 10 hours or more where no hot water is being drawn, but SOME standby loss is occurring? I could imagine that this would need to kick on once or twice a day to keep the water at the set temperature.

BTW, we have programmable thermostats for our hvac systems because it saves money not to heat or cool an unoccupied space. Why don't we have these for water heaters? It seems like a no-brainer that we should have more control over the arguably the most expensive appliance in the home. Just curious.

Interesting about the Navien CR with its buffer tank and its real performance. They recalled this unit due to a bunch of problems and it's been replaced with a similar one called the NR. I wonder if it wouldn't have been better to test a few different condensing tankless units (such as the eternal, or one of rinnai's) just to be sure that they're not testing a particularly bad unit that spoils the entire category?

Dana, do you think a condensing tankless with a well-insulated 1 or 2 gallon electric water heater in front of it to help buffer and eliminate cold water sandwiching would work to resolve the issues with tankless?

The center-flue heat exchanger of an atmospheric-drafted gas fire tank is open a the top and bottom, and the hot water heating the air in the heat exchanger causes it to convect heat out into the room air.   Non-condensing tanks burn at about 80% efficiency, so 20% of the heat goes up the stack right off the top, but another 20% or so is from the standby losses, more than half of which is lost due to this convection.  Between 10-15% of the total source-fuel energy is from this convection loss, and maybe 10% of the energy is from other standby losses.  Thats how starting at 100% source fuel, 20% up the flue leaves 80%, then convection eats up another 15% leaving 65%, and another 10% is conducted through the sides of the tank and the connected plumbing, leaving it at ~55%, or an EF of 0.55.

In a condensing tank heater is forced-draft and has flow-prevention dampering on the vent, and the convection losses that plague it's atmospheric drafted cousin drop to near-zero.  The steady state combustion efficiency during a burn is typically 96-98%, and since they're typically better insulated than non-condensing tanks they lose 6-8% to conducted losses, leaving them at about 0.88-0.92 in an EF test, but that's also about how well they perform in real-world applications. 

Most tankless applications get short-cycled into the low to mid-80s for real-world average efficiency, unless their primary function is filling big tubs or long, endless showers.  At typical 4 person family hot water draw  profiles, take 0.10 off the EF test number for any tankless and you'll be pretty close.  So the better condensing tankless units can approach condensing-tank performance, but won't usually beat (if ever), despite scoring better in an EF test.  Only at unusually low hot water volume use does the lower standby of a (non-buffered) tankless pull ahead.

The cycling losses are miniscule compared to a tankless. With 50 gallons of thermal mass to work against the minimum burn time is always quite long relative to their flue-purge & ignition cycles, and are always within 1% of the steady-state continuous-burn efficiency.  And this it how they beat tankless efficiencies to pulp in real-world applications that have many short-draws.  With a tankless (even a mini-buffered tankless) there is an associated ignition cycle & flue purge that throws away a fixed amount of heat. With a tank there can be dozens of smaller draws before the tank drops sufficiently in temp to trigger a burn, then the burn is heating at least 50 gallons of water by 5-7 degrees (~2000-3000 BTUs of burn) rather than heating 1/4 gallons of water by 75 degrees (about 150 BTU) with  roughly the same fixed losses from ignition & flue purge cycles.

Turning off your hot water heater at night has very little effect on the standby loss. A good condensing tankless or electric tank loses less than half a degree per hours overnight. It's the difference in tank temp and room temp that determines the rate of standby loss, so it really doesn't change much. In a lossy low-mass house the temperature inside drops, reducing the delta-T between indoor/outdoor temps by much bigger fraction. (But the savings are much less in well insulated homes with high thermal mass that might only see a 2F drop over 8 hours on a cold winter night.)   With a cheap atmospheric drafted tank savings from programmable setbacks might occur, but the cost of the controls are probably better applied to other things (like condensing heat exchangers, and dampered forced-draft burners) to boost combustion efficiency or reduce standby loss.  While hot water heating is usually the second biggest energy use (after space heating/cooling) in a typical home, it's usually a distant second.  Saving 5-7% of the space heating bill with a smart thermostat is usually much bigger deal than saving 1-2% on the water heating bill would be, and 1% is more you'd get out of turning off your electric tank heater overnight:

Most of the energy in an 0.90EF water heater is still going into heating the water, with the 10% lost being distributed over 24 hours of standby, or 0.4% per hour.  In an 8 hour "off" period the standby loss for that period is still substantial because the water is still hot. Even if it lost over 1F/hr, for 10F over night going from 125F to 115F,  the average temp is about 120F. In a 65F basement that means the average delta-T changed from 60F to 55F, or about a 10% reduction in the standby loss for that period, taking it from 0.4% per hour to about 0.36% per hour.  Even over a 24hour period that 10% reduction in standby would only adds up to 1% of the total energy use, and in reality an overnight setback strategy is saving you less than 0.33% total. 

Spend the money on R4 pipe insulation first- it'll buy back several percent of the total energy in less tepid-water waste and lowered standby loss at a very favorable R.O.I. (simple payback under 2 years in most cases.) Drainwater heat recovery buys you a double-digit savings, but with a longer financial horizon.  BTW:  R3 min. pipe insulation will be required by code under IRC2012 in an array of situations. (See pp 17 & 18 )  The 3/8" wall goods found in box stores is only R2, so you're looking at 1/2" min wall thickness to meet code, but 5/8" goods (~R4) are readily available at reasonable prices through plumbing supply chains.

Buffering the tankless with a circulation loop to a 2 gallon tank with a 10F hysteresis on it's aquastat raises is still only (8.34lb/gallon x 2 gallons x 10F=) 167 BTU, but it inserts a standby loss of the heat abandoned in the supply and return lines to the mini-tank. (50' of 3/4" plumbing, or 25' in each direction is about a gallon of hot water.   At 5 gallons or more of buffer you might see an efficiency boost if circulation loop is insulated to at least R4 (in both directions), but a lot depends on just how much heat is abandoned in the recirculation loop and lost to the surrounding spaces.

Posted By Dana1 on 05 Jun 2012 02:54 PM

Posted By strategery on 04 Jun 2012 06:56 PM

The standby loss of a condensing tank is quite small- about the same as with electric tanks. This is in contrast to the standby losses of atmospheric-drafted  type gas fired tanks, where the center-flue heat exchanger constantly convects heat out of the tank 24/365.

I wonder why this is? I think I understand how the Vertex condensing unit works. It has a really long coiled heat run that distributes the heat more evenly? But since it USES more of the heat it gets by burning natural gas, wouldn't it also only need to run for a very short period of time causing some lowered efficiency? I'm just thinking about the unit keeping the water at the desired temperature for the 10 hours or more where no hot water is being drawn, but SOME standby loss is occurring? I could imagine that this would need to kick on once or twice a day to keep the water at the set temperature.

BTW, we have programmable thermostats for our hvac systems because it saves money not to heat or cool an unoccupied space. Why don't we have these for water heaters? It seems like a no-brainer that we should have more control over the arguably the most expensive appliance in the home. Just curious.

Interesting about the Navien CR with its buffer tank and its real performance. They recalled this unit due to a bunch of problems and it's been replaced with a similar one called the NR. I wonder if it wouldn't have been better to test a few different condensing tankless units (such as the eternal, or one of rinnai's) just to be sure that they're not testing a particularly bad unit that spoils the entire category?

Dana, do you think a condensing tankless with a well-insulated 1 or 2 gallon electric water heater in front of it to help buffer and eliminate cold water sandwiching would work to resolve the issues with tankless?

The center-flue heat exchanger of an atmospheric-drafted gas fire tank is open a the top and bottom, and the hot water heating the air in the heat exchanger causes it to convect heat out into the room air.   Non-condensing tanks burn at about 80% efficiency, so 20% of the heat goes up the stack right off the top, but another 20% or so is from the standby losses, more than half of which is lost due to this convection.  Between 10-15% of the total source-fuel energy is from this convection loss, and maybe 10% of the energy is from other standby losses.  Thats how starting at 100% source fuel, 20% up the flue leaves 80%, then convection eats up another 15% leaving 65%, and another 10% is conducted through the sides of the tank and the connected plumbing, leaving it at ~55%, or an EF of 0.55.

In a condensing tank heater is forced-draft and has flow-prevention dampering on the vent, and the convection losses that plague it's atmospheric drafted cousin drop to near-zero.  The steady state combustion efficiency during a burn is typically 96-98%, and since they're typically better insulated than non-condensing tanks they lose 6-8% to conducted losses, leaving them at about 0.88-0.92 in an EF test, but that's also about how well they perform in real-world applications. 

Most tankless applications get short-cycled into the low to mid-80s for real-world average efficiency, unless their primary function is filling big tubs or long, endless showers.  At typical 4 person family hot water draw  profiles, take 0.10 off the EF test number for any tankless and you'll be pretty close.  So the better condensing tankless units can approach condensing-tank performance, but won't usually beat (if ever), despite scoring better in an EF test.  Only at unusually low hot water volume use does the lower standby of a (non-buffered) tankless pull ahead.

The cycling losses are miniscule compared to a tankless. With 50 gallons of thermal mass to work against the minimum burn time is always quite long relative to their flue-purge & ignition cycles, and are always within 1% of the steady-state continuous-burn efficiency.  And this it how they beat tankless efficiencies to pulp in real-world applications that have many short-draws.  With a tankless (even a mini-buffered tankless) there is an associated ignition cycle & flue purge that throws away a fixed amount of heat. With a tank there can be dozens of smaller draws before the tank drops sufficiently in temp to trigger a burn, then the burn is heating at least 50 gallons of water by 5-7 degrees (~2000-3000 BTUs of burn) rather than heating 1/4 gallons of water by 75 degrees (about 150 BTU) with  roughly the same fixed losses from ignition & flue purge cycles.

Turning off your hot water heater at night has very little effect on the standby loss. A good condensing tankless or electric tank loses less than half a degree per hours overnight. It's the difference in tank temp and room temp that determines the rate of standby loss, so it really doesn't change much. In a lossy low-mass house the temperature inside drops, reducing the delta-T between indoor/outdoor temps by much bigger fraction. (But the savings are much less in well insulated homes with high thermal mass that might only see a 2F drop over 8 hours on a cold winter night.)   With a cheap atmospheric drafted tank savings from programmable setbacks might occur, but the cost of the controls are probably better applied to other things (like condensing heat exchangers, and dampered forced-draft burners) to boost combustion efficiency or reduce standby loss.  While hot water heating is usually the second biggest energy use (after space heating/cooling) in a typical home, it's usually a distant second.  Saving 5-7% of the space heating bill with a smart thermostat is usually much bigger deal than saving 1-2% on the water heating bill would be, and 1% is more you'd get out of turning off your electric tank heater overnight:

Most of the energy in an 0.90EF water heater is still going into heating the water, with the 10% lost being distributed over 24 hours of standby, or 0.4% per hour.  In an 8 hour "off" period the standby loss for that period is still substantial because the water is still hot. Even if it lost over 1F/hr, for 10F over night going from 125F to 115F,  the average temp is about 120F. In a 65F basement that means the average delta-T changed from 60F to 55F, or about a 10% reduction in the standby loss for that period, taking it from 0.4% per hour to about 0.36% per hour.  Even over a 24hour period that 10% reduction in standby would only adds up to 1% of the total energy use, and in reality an overnight setback strategy is saving you less than 0.33% total. 

Spend the money on R4 pipe insulation first- it'll buy back several percent of the total energy in less tepid-water waste and lowered standby loss at a very favorable R.O.I. (simple payback under 2 years in most cases.) Drainwater heat recovery buys you a double-digit savings, but with a longer financial horizon.  BTW:  R3 min. pipe insulation will be required by code under IRC2012 in an array of situations. (See pp 17 & 18 )  The 3/8" wall goods found in box stores is only R2, so you're looking at 1/2" min wall thickness to meet code, but 5/8" goods (~R4) are readily available at reasonable prices through plumbing supply chains.

Buffering the tankless with a circulation loop to a 2 gallon tank with a 10F hysteresis on it's aquastat raises is still only (8.34lb/gallon x 2 gallons x 10F=) 167 BTU, but it inserts a standby loss of the heat abandoned in the supply and return lines to the mini-tank. (50' of 3/4" plumbing, or 25' in each direction is about a gallon of hot water.   At 5 gallons or more of buffer you might see an efficiency boost if circulation loop is insulated to at least R4 (in both directions), but a lot depends on just how much heat is abandoned in the recirculation loop and lost to the surrounding spaces.

Well, you've convinced me. Time to start looking at the Vertex. The polaris is too pricey.

I think there's another version of the Vertex under the State brand. Bradford White might have their own as well, but I'd have to verify that.

There's another type of water heater out by Eternal the gu100 which was just released at the tail end of last year. I can't find a single review on it from someone who has actually USED it. It has a 2 gallon tank and the water flows from the bottom to the top which allegedly helps prevent sediment buildup from the constant flow and washing motion. This was designed to be a retrofit unit that can be used with a 1/2" gas line and it vents with pvc (concentric venting works too, I think). This unit intrigues me. I hope that it doesn't suffer from some of the problems that the Navien has had. GBA had an article about tankless units not long ago and found the Navien to be an energy hog (because of the buffer tank I presume) that didn't perform at the stated EF.

I've been awaiting more real-world beta on the Eternal too. With a 2-gallon buffer and some real insulation it might not be terrible, but reliability is still TBD.

Navien has a pretty spotty record on reliability too, but about half that may be the quality of the installation. I've read some pretty bizarre web accounts of mis-installed Naviens by less-than-factory-certified "pros" over the past few years. The quality of the support network for any of them counts.

Posted By Dana1 on 11 Jun 2012 06:39 PM
I've been awaiting more real-world beta on the Eternal too. With a 2-gallon buffer and some real insulation it might not be terrible, but reliability is still TBD.

Navien has a pretty spotty record on reliability too, but about half that may be the quality of the installation. I've read some pretty bizarre web accounts of mis-installed Naviens by less-than-factory-certified "pros" over the past few years. The quality of the support network for any of them counts.

I've put in lots of Naviens but never have seen or heard of "factory certified" pros in our neck of the woods. All the problems we had were, admittedly by the Navien dudes on the phone, design issues which have, so far in my experience, been fixed when the newer versions came out.

There has been a lot of over selling of the performance of tankless heaters in general so this adds to the issue. 

I have a Vertex 100 (50 gal, propane - we can't get natural gas) providing DHW and heat for my radiant floor (I heat a 2800 ft2 house with it). The Vertex is set up to easily connect to both your domestic hot water, and your radiant floor through a heat exchanger. You have a pump on the hot water tank side and one on the radiant side. It works perfectly, and I also have very cold water from a spring fed well. The recovery rate is phenomenal, and the Vertex is very efficient. I highly recommend it.

If you would like to see a pic of how I have it set up go to this link and scroll down about three quarters of the way -
http://www.facebook.com/media/set/?set=a.394434177917.171985.665372917&type=3&l=35a6857cce

-Rosalinda

I recently noticed water in the basement around the base of the water heater (Kenmore, standard gas, vented to the chimney, ten or more years old), so it looks as though I'll soon be replacing it. I'd already been leaning toward a condensing tank unit (like the Vertex), but have a couple of questions.

(BTW, my hat is off to the great folks on here who have taken the time to share their evident expertise on this subject.)

The Vertex is not cheap and the installation will have costs well beyond a same/same replacement, but I'm wondering whether it is worth the investment if the Vertex' glass lining -- the apparent Achilles heel of tank heaters -- is going to fail in the same 10-15 year time frame as my current unit. The steel lined Polaris and Phoenix tanks are that much more costly -- or so I have the impression -- and the Polaris has had reliability issues as mentioned in this thread.

I also have no drain in the vicinity, nor any acceptable way to pipe to a drain. The unit is below grade, but I definitely don't want to mess with a pump. When we did an extension 22 years ago, we had a condensing hot air furnace installed along with central air with the air handler co-located with the furnace. The condensation drain for the air conditioning unit -- and for the furnace, presumably -- consists of a simple hole drilled through the slab -- which surprisingly has not overflowed yet this year onto the garage floor. Would such a hole drilled through the basement floor work as a condensate drain for a condensing tank water heater like the Vertex?

Posted By feeneyjj on 05 Jul 2012 09:04 PM
I recently noticed water in the basement around the base of the water heater (Kenmore, standard gas, vented to the chimney, ten or more years old), so it looks as though I'll soon be replacing it. I'd already been leaning toward a condensing tank unit (like the Vertex), but have a couple of questions.

(BTW, my hat is off to the great folks on here who have taken the time to share their evident expertise on this subject.)

The Vertex is not cheap and the installation will have costs well beyond a same/same replacement, but I'm wondering whether it is worth the investment if the Vertex' glass lining -- the apparent Achilles heel of tank heaters -- is going to fail in the same 10-15 year time frame as my current unit. The steel lined Polaris and Phoenix tanks are that much more costly -- or so I have the impression -- and the Polaris has had reliability issues as mentioned in this thread.

I also have no drain in the vicinity, nor any acceptable way to pipe to a drain. The unit is below grade, but I definitely don't want to mess with a pump. When we did an extension 22 years ago, we had a condensing hot air furnace installed along with central air with the air handler co-located with the furnace. The condensation drain for the air conditioning unit -- and for the furnace, presumably -- consists of a simple hole drilled through the slab -- which surprisingly has not overflowed yet this year onto the garage floor. Would such a hole drilled through the basement floor work as a condensate drain for a condensing tank water heater like the Vertex?

A condensing gas water heater is going to need either a floor drain or a pump nearby. If you don't have one, it could just drive up the already high installation costs even more.

What part of the country are you in? How many people in the household?

You may be better off with a regular powervent gas water heater or a regular gas tankless. Or depending on your climate and number of members in your household  a heat pump water heater.

Thank you, Strategery. I may well follow your advice. Any opinions on powervent brands? (This and other forums have pretty much convinced me to stay away from tankless.)

We are in Washington, DC. Normally four people in the house, but one is serving abroad for the next year.

The main (cast iron?) waste drain is about 18 feet away. Could a hole be drilled in that and a pvc condensate drain line be inserted and sealed into it? (We have never had a backup into that waste drain, and I don't think the clean out has ever been opened since the house was built in 1934). Or, would doing that risk compromising the waste drain in the long term?

Anyone have an opinion on how long a Vertex, glass-lined tank would likely last?

The Vertex and Polaris are made by the same manufacturer and are indeed quite similar. The qualitative difference is a ~12-15 year lifespan for the Vertex and 20+ for the Polaris. Swapping sacrificial anodes ever 6-7 years is usually sufficient to get 20 years out of a glass lined water heater though. I've seen several gas-fired tanks last well into their third decade this way (none of which were mine though.)

If you have good drainage under the slab you may be able to just drill a condensate drain/sump through the slab (sub-code, but not often a hazard.) But the tiny pumps used in the HVAC trades to deal with air conditioning condensate are ubiquitous, cheap and reliable.

Boring into an antique cast iron drainpipe is risky even under the best of circumstances, and in this situtation it's a complete non-starter. Without it's own drain trap and enough burner use to guarantee that the the trap never runs dry it's bound to leak sewer gases into the house at some point. It would never meet code.

feeneyjj-

I would start by looking at some good plumbing or hvac companies in your area and asking them what they install. There's a lot of good brands out there- A.O. Smith, State, Bradford white, etc but more important than that is whether the company that installs it for you is good and they stand by their work. Even a good manufacturer can occasionally produce a lemon and you want a reputable dealer/installer that cares about their reputation and willing to make it right in the event that something goes wrong.

Or as Dana said, it might be possible to just drill a hole for the condensate drain if you have decent drainage. I don't know how long you plan on living there, but this might bring up inspection issues down the road (though I can't imagine anyone would say it's very serious) if that's an issue. It's not like it's a fire hazard or anything.

If it were me, my preference would be just a simple high efficiency power vent. Simple, tested, and likely no code issues to worry about. Plus it's upfront cost is going to likely be considerably lower than a condensing tank. If you had a drain or sump pit near I might say differently.

A serious thanks to Dana1 and Stratergy.  I have contacted an AO Smith installer in the area.  I'll see what they have to say between the Vertex -- I still like the idea that one of my sons won't shower the rest of us out of hot water --  versus a power vent.

Choose world classic water heater will is very comfortable for every one visit go here http://ourgreendirectory.com/

I can confirm that the condensate from combustion is very corrosive and you don't want it to touch most metal or concrete.

Jonr -  thnks for your input. 

I take it you are referring to the option of draining the condensate through a hole drilled in the concrete slab. Right?

Yes, although I send it down a plastic drain. I had to make sure that mine didn't touch the concrete on the way to the drain. I wouldn't do it at all with metal drain pipes.

Posted By jonr on 20 Jul 2012 11:34 AM
Yes, although I send it down a plastic drain. I had to make sure that mine didn't touch the concrete on the way to the drain. I wouldn't do it at all with metal drain pipes.

Most condensing appliances have optional neutralizing kits that should be used prior to disposing of the condensate.