Water Heating Problems

Users who are viewing this thread

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
We did major renovations about 7 years ago at which time we installed a Paloma tankless (PTG-74PVN-2) and a 50 gallon GE tank (sg50t12avgh00). The tankless feeds hot water into the tank giving us 'unlimited' hot water.

I am very dissatisfied with the tankless. Annual cleaning due to error code 11 is a pain in the neck- takes me half a day and involves taking the unit apart and cleaning the fuel rods, etc. I just cleaned it and it is showing the error code 11 again!

I am looking for a better solution to provide a large supply of hot water without annual maintenance. Is there a better tankless that's an easy replacement? Should I just get a larger and quicker to recover tank?
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
All tankless units have higher maintenance than tanks. I didn't know Paloma was a nuclear water heater though (fuel rods?) :)

Need to know more about your hot water use to know what to recommend. Filling a 250 gallon spa is a different task from needing to run 5 showers at a time, or 5 showers in rapid succession.
 

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
All tankless units have higher maintenance than tanks. I didn't know Paloma was a nuclear water heater though (fuel rods?) :)

Need to know more about your hot water use to know what to recommend. Filling a 250 gallon spa is a different task from needing to run 5 showers at a time, or 5 showers in rapid succession.
 

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
It's a two family house with 3 showers, so a bath followed by 5-6 showers is possible. We never run out with the current setup unless the tankless fails.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
Got a ZIP code?

In most US locations a regular 50 gallon gas-fired tank and a drainwater heat exchanger would deliver the performance you need.

If it's slab-0n-grade house making the drainwater heat exchanger a non-starter, the smallest condensing HTP Phoenix Light Duty has enough burner to keep up with a 24/365 continuous 2 gpm shower almost anywhere, with more than 2x the burner of a typical 50 gallon cheapo. Being a stainless steel tank it'll be good for at least 20 years, unlike the cheaper glass-lined condensing Versa (or a standard gas-burner.)

Couldn't find the specs on your GE- is that just a standard 50 gallon electric (and thus, glass lined), not wired up? How are you controlling the temp, and what is the storage temp?

If it's in good shape you could just swap in a $500 Takagi TK Jr, which are pretty robust units.

A key to using a tankless in this fashion is to not over-pump the tankless, and set it up for a BIG delta-T. So, if you're keeping the tank at 140F, set the tankless to at least 160F, 175F would be even better, but only pump it at 2-4 gpm. When the temp in the tank has dropped to 105F (a good showering temp), with the tankless cranking out 175F water it will deliver about 70,000 BTU/hr if pumped at 2 gpm, or 140,000 BTU/hr if pumped at 140,000 BTU/hr. Even in cold water country a 2 gpm shower only needs 70K behind it to keep up. With the tank set up at 140F, with 175F output and 2gpm it would modulate back to about 35,000 BTU/hr as the tank temp finishes. Odds are you'll never need more than 2 gpm if set up that way.

Instead of using a 125 watt circulator like a Taco 007 series, a smart circulator like the Bumblee that allows you to program the flow rate gives you a great deal of flexibility on how the system behaves. The lower the flow rate, the longer and fewer the burns will be which should enhance longevity & efficiency. Most tankless units need at least 1 gpm to fire up. With effectively zero pumping head other than the tankless itself, standard circulators will over-pump the tankless at 4-6 gpm, and there's really no need to abuse it that way, since you have 50 gallons of buffering thermal mass.
 

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
My house is an old NYC row house.

Are DWHEs an easy retrofit? I've just started reading about them.

Aren't condensing units expensive and relatively new and untested? Any maintenance?

My GE tank is a year old cheapo HD unit. Storage temp is not exact, I keep the dial between A and B.

Will the Takagi require a difficult annual maintenance? I am sick of fiddling with a noisy tankless.

I am confused by your calculations. If my tank is approximately 130 (between A and B) what should my tankless be set to? Tankless is currently set to 120 and has a max of 140. Could raising the temp reduce maintenance?

I had a recirculating pump to give us instant hot water, but it burned out because it was misconfigured and running non-stop. I have no other pump.

BTW, another major problem we have is waiting forever for hot water. It takes about 2-3 minutes for the showers to get hot.
 
Last edited:

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
Condensing tank hot water heaters have been around for decades, and were a popular way to do low-temp radiant floor heating back in the 1990s and early 2000s (still are in some quarters.) The Vertex has been around for at least 10 years. The HTP Light Duty is a smaller-burner version of their well-regarded commercial condensing hot water heaters.

Condensing tankless units have been around for about a decade, but Takagi's units are only maybe 50 years old. But if the entering water at the tankless is going to be higher than ~120F or so you won't get much condensing efficiency out of them.

If you set the tank to 130F, you want the tankless to be set to a MINIMUM of 130F + 25F (=155F), but 130F + 45F (=175F) is just fine, even better. Unlike hydronic boilers, these things tolerate very high temperature differences from in-to-out. They are designed handle 35F wintertime incoming water temperatures with 120F output without a problem, which is an 85F delta-T. They aren't very well suited for 110F incoming water with 120F output (a 10F delta-T), where efficiency suffers, and higher flow rates are necessary to keep the burner from flaming out.

With the output of the tankless to only 140F and maintaining the tank at 130F (a 10F delta-T) you'll be running at lower efficiency due to the low temperature difference, and probably a lot more burn cycles too, which lowers the efficiency even further.

Ideally you'd want the controls to turn on a ~2 gpm flow whenever the tank dropped below say, 115F (still plenty hot for filling tubs), but continue until the tank got up to 135F, raising the thermal mass of the tank about 20F. With 50 gallons of water it takes a bit over 8000 BTU to deliver a 20 degree rise. At 2 gpm and an output temp of 165F the burner would be modulating down to 30,000 BTU/hr at the end when the tank hits 135F, but would be firing at 50,000 BTU/hr at the initial turn-on at 115F. During an endless shower pulling the tank temp down to 105F it would be firing at 60,000 BTU/hr, which is enough for a forever-shower if using a low-flow head. If that turns out to be not enough (unlikely), you can crank up the tankless to even higher temps without damaging it.

At an average firing rate of about 40,000 BTU/hr, with the controls set up for a 20 degree rise in the tank you have a guaranteed minimum burn time of about 0.2 hours (12 minutes) to deliver that 8000 BTUs, which translates into far fewer ignition cycles and less wear & tear overall. Error code 11 on a Paloma (and Takagi) would usually be the flame-sensor crudding up, which is partly a function of ignition cycling. Other factors include local air pollution (particularly volatile organics- never install one of these in a hair salon where aerosol hair products are used.)

Drainwater heat exchanger retrofits are dead-easy if it's a plastic drain that has free access and located reasonably close to the water heater. It's more complicated with antique cast-iron tangled up in framing or plumbing. The tallest & fattest one that fits will deliver the most performance, and is worth the marginal uptick in upfront cost.

Most old NYC row houses are heated with either a steam or hydronic boiler. If that's the case for you, running an indirect HW heater as separate zone off the boiler is probably going to be the best bang/buck.
 

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
The water goes the other way. Cold water into tankless, tankless feed hot water into tank.

I have no other hot water in the house, heat was changed to forced air.

Is my tankless condensing? I thought condensing was for tanks.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
The way you have it configured it's guaranteed to short-cycle the tankless (the way most of them are operated), since it fires on every hot water draw. The only benefit you're getting from the storage tank is avoiding the "cold water sandwich" problem of the turn-on delay.

If the cold water enters and leaves the storage tank as if it were wired up, and you used the heating element ports to hook up the plumbing to the tankless, you can then use the tankless & pump as the "heating element". I've heard (but not seen or tried it) that people sometimes use the upper temperature controls & internal heater element switching to either the lower or upper element to switch power to the pump, saving something on control costs. The power required for the pump is a fraction that of the heating element of an electric hot water heater, but with standard circulators it presents as a inductive load, not a resistor like a water heater element, but it probably works. (Don't wire up the electric tanke to 240VAC and put a 120V pump on it, eh? Most have controls that work fine at 120V.)

Controlled that way the tankless only fires only when the temp in the tank drops, and the burn time is a function of the hysteresis in the HW heater's controls (unknown), the output temp setting of the tankless (which you can program) , and the flow rate of the pump.

water-heater-construction.jpg


Your seven year old Paloma is probably about 80-84% steady state efficiency, and non-condensing. (It has stainless exhuast vents, not plastic?) You can buy a condensing 95% efficiency Takagi for about a grand, but it's not clear that it's worth it in your sitution, but it might be. The only current condensing Takagi model that can be set up for higher than 140F output is the biggest, the 199,000 BTU/hr T-H3-DV. For half the money you can get a Takagi-Jr (about the same efficiency as the Paloma), which has a max firing input rate of 140,000 BTU/hr.

The first thing you might try is buying a flow-settable pump like the Taco Bumblebee (not sure if it won't crap out in a fresh water situation) , or the wimpiest lowest-flow potable water recirc or a wimpy 1/40 hp bronze hydronic recirculator pump to not over-pump the tankless , and use the height & low element ports on the tank as plumbing taps. Pump into the top element port, return to cold side of the tankless with the bottom port. Set the Paloma to 160F+, and the tank to 130-140F. That alone will reduce the number of burn cycles by about a full order of magnitude.

If you replace the Paloma with a condensing tankless you'll probably have to set the tank temp to 120F to get much condensing efficiency out of it. That storage temp is too low to kill legionella, but still high enough to prevent colonies from growing. Don't store water below that temp.
 

TerryFan

New Member
Messages
7
Reaction score
0
Points
1
Location
New York
My tankless is non-condensing and aparently not direct venting. Rheem has told me that my unit is prone to problems because the air supply comes from the inside, not through the vent. New ones only vent from the outside and are also much easier to clean.

I want something simple that will supply sufficient water but not require babysitting. I will talk to my local plumber about my options. So far I like the idea of decommissioning the tankless, replace the tank and install a drainwaterheat exchanger.

Which kind of tank should I consider? Condensing units look snazzy, but do they justify their high price tag? Also, not sure about the installation complexity: drainage, etc
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
The AO Smith Vertex works fine, but it's a glass-lined tank with a 10 year warranty.

The HTP Light Duty is more money, but is all stainless, modulating and should last 2x as long.

Assuming you heat the place with a boiler, a stainless indirect like a SuperStor Ultra operated as a heating zone off the space heating boiler would be better overall, (even if your boiler is not a condensing boiler), and cheaper than the HTP Light Duty, maybe cheaper than a Vertex, and should be good for 20+ years.
 

Jastori

Member
Messages
122
Reaction score
0
Points
16
Location
Illinois
Ideally you'd want the controls to turn on a ~2 gpm flow whenever the tank dropped below say, 115F (still plenty hot for filling tubs), but continue until the tank got up to 135F, raising the thermal mass of the tank about 20F. With 50 gallons of water it takes a bit over 8000 BTU to deliver a 20 degree rise. At 2 gpm and an output temp of 165F the burner would be modulating down to 30,000 BTU/hr at the end when the tank hits 135F, but would be firing at 50,000 BTU/hr at the initial turn-on at 115F. During an endless shower pulling the tank temp down to 105F it would be firing at 60,000 BTU/hr, which is enough for a forever-shower if using a low-flow head. If that turns out to be not enough (unlikely), you can crank up the tankless to even higher temps without damaging it.

I realize this thread is a year+ old, but I am very interested in this tankless + tank approach.

I'm wondering why you recommend only a ~2 gpm flow here. With this relatively small delta T, most tankless heaters could easily supply ~6 gpm. Wouldn't it be preferred to reheat the tank more quickly at a higher BTU/hr? In an article by Michael Chandler (see link below), he recommends a large pump (Taco 009) pushing as much water as possible through the tankless to reheat the tank.

http://www.chandlerdesignbuild.com/files/fhbDecJan08.pdf

I'd be really interested in any comments you have about the pros / cons of the large pump at ~6 gpm compared with your recommendation of a smaller pump at ~2 gpm to reheat the tank. Thanks!
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
It's more than just the BTU/hr translated into a delta-T.

A tankless will wear out the flow sensors and even erode the internal plumbing if it's always pumped at 6 gpm. Tankless heat exchangers are designed for and very tolerant of very high delta-Ts. It'll use far less pumping power per delivered BTU that way too- these heat exhangers have significant pumping-head.

Also, running the burner at it's max rate all the time will shorten it's lifespan. Ideally you'd have a peak firing rate no more than 2/3 the maximum rate, and let it modulate the firing rate, rising as the tank's temp falls, and falling to something well under half the max rate when the tank's storage temperature is satisfied.

The equipment will last longer and run more efficiently if you crank up the output temp and run it at a higher delta-T rather than just over-pump the hell out of it at a low delta-T.

At 2 gpm (~1000 lbs/hr) and a 100F delta-T you can re-heat the tank at a rate of 100,000 BTU/hr, which is 3x the rate of a typical standalone tank's burner. A tank kept at 140F, and a tankless set to it's max 170-180F a the output would be burning at 30-40K at the end, but would modulate on up to over 100K by the time the tank's average temperature drops to human body temp (the low-limit of a reasonable shower.) Plumbed correctly (as in Chandlers second example) the top of the tank would be taking in the 2 gpm of super-hot water, but the tankless would be drawing from the much cooler bottom of the tank, where the 35-50F incoming water is being injected. A 2.5 gpm shower with a 70F rise (35F-in, 105F at the shower head) can go forever with 88,000 BTU/hr of input, so there's rarely a need to go more than 100K or so for tankless output in this sort of configuration, and a ~150K tankless is just fine. A 40 gallon tank with a modulating burner peaking at ~100KBTU/hr on the side can easily support two or three simultaneous showers of normal length. If you need more heat than that , you can pump it at 3 gpm without overstressing it, as long as the burner isn't maxing out every time.

A smart ECM drive pump with a programmable flow rate would be worth the upcharge in high electricity price markets, rather than Taco-00x series. It has to be a bronze or plastic pump too- the dissolved oxygen in the potable water will destroy a standard iron heating circulator pump.
 

Jastori

Member
Messages
122
Reaction score
0
Points
16
Location
Illinois
Thanks so much for the detailed reply.... your recommendations make a lot of sense!

It seems like one of the key differences between your approach and what Michael Chandler proposed is the max temperature of the tankless. In his example, he had the tankless temp set at 140 deg, and the tank at 125 deg. With that very small delta-T (at least at the end of the cycle), maybe it is necessary to pump at the higher GPM? If the tank thermostat has a 10 deg differential, it seems like it might be difficult to reach the shut-off temperature with a 140 deg supply from the tankless.

It seems that many of the tankless units are limited to 140 deg. max (especially the smaller ones). I know the Takagi-Jr can go up to 165 deg., but I think the smaller condensing units are typically limited to 140 deg max.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
"..maybe it is necessary to pump at the higher GPM"

Not "maybe" it's built into the definitions:

1 BTU is the amount of heat needed to raise 1 lb of water 1F. If you are raising the temperature 10F it takes 10BTU/lb.

A gallon of water weighs 8.34lbs, thus 8.34 lbs/gallon x 1 gpm x 60 minutes per hour = 500.4 lbs/hr. for every gpm

To hit an BTU per hour RATE you can either bump up the lbs per hour at a low delta-T (Chandler's approach) or you can increase the delta-T (which is much nicer to the equipment) and run fewer lbs per hour. As long as the lbs/hr x delta-T is the same, it's the same firing rate eg:

1 gpm (500 lbs/hr) x 50F= 25,000 BTU/hr

2 gpm(1000 lbs/hr) x 25F= 25,000 BTU/hr

At a showering temp in the tank of 105F and a tankless temp of 140F you have a delta-T of 35F. A 2 gpm (~1000 lbs per hour) with 35F incoming water with 105F is a 70F delta-T, for 70,000 BTU/hr. Since the incoming water to shower delta-T is 2x that of the tank-to-shower delta-T, the whole thing keeps up just fine pumping at 2x the shower flow, which would be 4 gpm. That's still a somewhat reasonable rate for a large tankless, but is pushing the limits on some smaller tankless units, which puts wear & tear on the flow sensor & internal plumbing.

If the storage temp of the tank is set to 130F and the tankless is pumping 4 gpm (=2000lbs/hr), with a 140F output on the tankless the finishing firing rate would be (2000lbs/hr x 10F=) 20,000 BTU/hr, which is fine, as long as the unit modulates that low. If you raise the temp much above that many tankless units would be firing intermittently, or would never finish. If you're pumping at 8 gpm it would be firing at 40,000 BTU/hr at the end, which is still quite a bit higher than a typical 40-50 gallon standalone. T

Pumping at the highest delta-T and lowest pumping rate that still delivers the desired performance is always going to be better for the equipment. To increase the delta-T you can also lower the storage temp of the tank, but going below 120F pushes it into a higher legionella-risk zone. At 120F you'd be looking at a 20F delta, so you could back off a bit on the pumping rate and let the tank buffer the heat, if it's still delivering the needed performance.
 

Jastori

Member
Messages
122
Reaction score
0
Points
16
Location
Illinois
Thanks again for the great information!

I was hoping that there was a way to combine the efficiency of a smaller condensing tankless (such as Takahi T-H3M) with a tank, but the 140 deg max temperature seems like a significant concern. Pumping 4+ GPM through the T-H3M would require a pretty big pump (~30+ ft Head) and the excess wear & tear doesn't sound good.

The Takagi-Jr has the benefit of the 165+ deg max temp, but the lower efficiency (non-condensing) plus the cost of the stainless venting make it less attractive.

Do you know if there are any smaller condensing tankless units that can exceed 140 deg for this type of a system?

It seems like if you need to step up to the T-H3 to get the higher max temp (and higher efficiency), then the total cost of the system (tankless + tank + pump + tempering valve...) may not be much savings in comparison to something like the HTP Light Duty.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
"...may not be much savings in comparison to something like the HTP Light Duty."

There is a lot to be said for just buying an already pre-engineered system with a modulating burner rather than re-inventing the wheel with cobbled together parts such as a Takagi and a tank with a pump that may or may not be optimal for the application.
 

jamesrav

New Member
Messages
4
Reaction score
0
Points
1
Location
Tijuana, Mexico
sorry to hijack this person's thread, but I can't seem to create a new one .. but my question does fall under the Topic for sure. I had what is no doubt an inexpensive 'made in China' tankless installed outside the house, which happens to be right by the wall for a bathroom (so the unit can't be more than 5 feet from that faucet). The water coming from that faucet is quite acceptable (about 102), so the unit appears to be functioning properly. However the situation upstairs in the bathroom is another story. Shower water reaches a max temp of about 86 degrees after about 30 seconds. The distance from the unit to the upstairs bathroom (2nd floor of course) is fairly substantial, probably 80 feet. Is that causing the dramatic drop in temperature? I am using only the Hot faucet in the shower (and I get roughly the same temp, maybe 88, using the Hot in the bathroom sink), so it seemingly can't be caused by the addition of Cold water that I am unintentionally bringing into play. What would be potential causes of a 16 degree drop in temp? I live in Rosarito, Mexico, so the air temp is never much below 60 degrees, even at night. Are these somewhat cold pipes that the water travels through bringing the temp down that much over the course of 80 feet? The owner of the place who I'm renting from isn't going to change it, and the plumber just shrugs his shoulders ... any hope I can get a higher temp at the shower?
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
Shower mixers have internal (usually adjustable) stops that limit how much hot vs. cold water in the mix to limit the risk of scalding. If the hottest water you can get out of a dedicated hot water tap (not a mixer) is 102F, the mixer stops are likely to be the problem.

Also, anti-scald valves operate off changes in pressure between the cold & hot side. With a tankless there is a significant pressure drop from the input side of the tankless to the output side, which can interact with anti-scald valves on showers.

When calling hot water on a long run the initial temperature at the bathroom sink might be several degrees cooler than a sink near the water heater, but after a few minutes the pipe should be up to temperature. If it isn't already insulated, it's worth insulating the hot water supply pipe with 3/4" walled closed cell pipe insulation, especially for such a long run.

A temperature of 102F is pretty tepid for showering- most people like it a few degrees warmer than that 104-107F. If you can adjust the output temperature of the tankless up to 115-120F you may be able to fix the shower temperature limitation problems.
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks