The Tankless Paradigm w/"Virtual Storage Tank"?
Hello HeatingHelp.com community! : )
prelude:
Commercial 100% tankless domestic hot water in a motel environment with dedicated "keep-alive loop"
The manufacturer's capacity design recommendation for this motel is eight (8) Rinnai CX199iN units (networked together).
Each tankless capable of max set temp at a max flow rate of 11gpm.
Total system capacity of 88gpm at set temp.
Target set temperature at guest rooms: 125 degrees F.
My thoughts:
First, I assume that in this networked configuration, any flow demand under 11gpm, only one tankless unit is firing up (and modulating) at any given time. Then for each subsequent call for heat, a different tankless unit fires up (all in a "round robin" sequence).
Second, my intention of incorporating a "virtual storage tank" into this network is by installing an auto-mixer valve to its pipe distribution.
Now, if I were set each tankless unit to 185 degrees F set temp, and mixer set temp to 125 degrees F ……….. this in of itself creates a "virtual storage tank" greatly increasing the "effective" total hot water system capacity available to the guest rooms.
My question is:
In doing this "mixer" inclusion, wouldn't my system design now only require lets say, halve the required tankless units?
Comments
-
….. continued:
But, In this "virtual" configuration, I don't want the tankless units trying to heat the "under 11gpm flow" demand up to its tankless set temp of 185 degrees F!!
This would be a complete waste of the gas heated BTUs … just to be later mixed (cooled) down by the mixer valve.
I believe to resolve this uncalled for heat increase is to simply move the primary tankless heat output sensor to the output of the mixer valve (while keeping the tankless set temp at 185 degrees F.)
Only when the tankless units reach there max deliverable flow rates (total 88gpm) …… only then, will the tankless units begin to ramp up (modulate) to their tankless set temp of 185 degrees F. , only then now enjoining the the cold water addition thru the mixer (the virtual storage tank).
Note: this desired paradigm of control may require a software patch in the tankless units PLC (Programmable Logic Controller).
In summary:
I desire the networked eight tankless units to enjoin each other only as needed at a tankless set temp of 125 degrees F.
Only after all the tankless units are maxed out of their initial heat setting (125 degrees F) capacity …… will the tankless units then be allowed to ramp up to 185 degrees F set temp! …… only now, entering into the "virtual" mode of operation.
0 -
You must have some pretty warm well water to be able to get 185 degree F water out of a single tankless at 11GPM. Generally in my area a 199 tankless can maintain 120-125 ish at about 5GPM and that is it. Max flow rate of the unit does not mean it can hit max temp at that flow rate, that is just the maximum flowrate that can move through the unit btw
A few things.. Most tankless will max out at 140 in residential mode, 180 in commercial 185 is really pushing it, most condensing boilers will not even have an adjustable high limit up to that level. I have to use commercial equipment to reach that setpoint in potable water tanks, and i've only ever done that for sanitizing lines at a dairy (factory process work basically). You run a very high risk with water temps that high, but you know that.
Most manufacturers staging controls for their tankless and boilers are not made to be fully customizable, there are a lot of reasons for this, but the biggest one would be that they haven't tested it. Most of what you want to do is probably not built in to the units by default, you will need to alter the programming of the units to do this which might be tricky, not possible for the average DIY or even contractor. There are concerns as well about altering the units programming, as this also controls ignition, you could make the units unsafe if you are not 100% sure of what you are doing. Unless the manufacturer specifically greenlights this, they will drop all warranty on the unit if they ever catch wind of it.
0 -
Hi GGross!
(FYI : I'm just a HAVC dabbler, so bear with me!) : )
"You must have some pretty warm well water to be able to get 185 degree F water out of a single tankless at 11GPM. Generally in my area a 199 tankless can maintain 120-125 ish at about 5GPM and that is it." - GGross
Rem: I'm maintaining a 1" "keep alive return loop" + "its 2" hot water distribution feed pipe" …. that equates to 100 gallons of a preheated volume of 125 degree water 24/7! (even under a no water use demand)
So, do I understand/assume that a single 199K BTU tankless unit can really only produce maybe only 1.5 gpm at a tankless set temp of 185F ?? : (
0 -
The best way to answer this would be to get your particular units spec sheets, they will have a graph on there showing flow rates at a given temperature rise. Take your output temp, subtract the inlet water temp from that, and that gives you the temp rise, now match that temp rise to its given flowrate on the graph
Like I said in my region we use 5GPM as a general max outlet flow rate from a 199 tankless, for standard residential use. but this is based on a given temp rise, your cold water may be quite a bit warmer than ours, but I doubt its warm enough for 11GPM from each unit. Of course if you are recirculating and blending some semi warm DHW with the cold inlet this will increase the cold water temp and lower your required temp rise. The key is to know the incoming water temp, and the required outlet temp, that will give the flowrate you can achieve with that unit
0 -
So, with a street temp of 60F this 199K tankless can maintain a constant 125F (65 Delta "T") hot out temp of only 6 gpm ……. and at 185F (125 Delta "T") just 3 gpm?? : (
0 -
That appears to be correct. Do you know what the peak DHW flow rate to the units would be, if everything runs at the same time?
0 -
"That appears to be correct. Do you know what the peak DHW flow rate to the units would be, if everything runs at the same time?" - GGross
No ….. we specified a 125 degrees F POU over the guest rooms without a mixer. The manufacturer calculated an eight (8) tankless units to meet their industry standard expectation of "peak load" from number of guest rooms, Jaccuzies, washers, etc.
So, I don't have an 88 gpm total system heating capacity …. but only a 48 gpm heating capacity @ 125 degrees F ??
….. correct?
And a total system capacity of 24 gpm @ 185 degrees F ?
Note:
I do believe in some very high end (cost no object) hotels, they do design for 100% peak demand (ie: every shower,sink, washer,etc … running at full flow with water pressure booster pumps in the mix as well)
I'm guessing a typical motel is around a 65% peak demand calculation target.
0 -
this Water Design Calculator will give you some DHW loads
I’ve stayed in a number if hotels that have a bank of tankless that cannot keep up with morning and afternoon loads I imagine there is a bit of a fudge factor in hotel dhw sizing
It will be a constant deliming process also if you have hard water
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1 -
"I’ve stayed in a number if hotels that have a bank of tankless that cannot keep up with morning and afternoon loads I imagine there is a bit of a fudge factor in hotel dhw sizing" - HotRod
And I assume the reason for this is there were not enough tankless units installed?
0 -
Since you have put a lot of thought into this I think it might be worth your time to figure the total DHW flowrate for your building so that you know what would be required for full flow. you may not need full flow, and a storage tank might help cover that up though. Something to keep in mind with any system here, what happens when a unit goes down? how much redundancy do you need? one nice thing about the banks of tankless heaters is that if a problem happens with a unit it is only 1/8 or 1/10 of the system going down. and each tankless heater is relatively inexpensive compared to a big HE gas tank unit. A downside is that once the units start to break down you may have new problems relatively regularly unless you plan to change them all out at once as they all operate about the same amount with the same water they will have a similar service life. Off hand I see mostly redundant HE gas tank systems but admittedly I don't spend a ton of time at hotels, I do bid out the projects pretty regularly though.
1 -
First, I assume that in this networked configuration, any flow demand under 11gpm, only one tankless unit is firing up (and modulating) at any given time. Then for each subsequent call for heat, a different tankless unit fires up (all in a "round robin" sequence).
I don't see why they would operate the way you describe, or even how to get them to operate that way.
0 -
OK! …. lets say my 65% based peak demand is 48 gpm @ a POU temperature set of 125 degrees F. (this works out great for the entire week days … with no hot water shortage problems)
This requires all eight (8) of the tankless units firing at the same time (max system output) at times to the varying demands.
Now, its Friday night and the motel rents out every room and all six (6) jacuzzis.
Now my actual short time demand is at a 65 gpm @ 125 degrees F.
My thinking is to incorporate an auto mixer valve (hot+cold) into the outfeed pipe to guest rooms and allow the tankless units to ramp up to their commercial max set temp of 185 degrees F.
Now, with a max 25 gpm @ 185 degrees system arriving at the "hot-in" port of the mixer ……. what would be the total system flow rate of 125 degree F POU water out the mixed port? (ie: Cold water gpm addition + 25 gpm tankless addition)
0 -
Its done via a data cascading cables connecting all the units PLC's together (ie: like a typical rotating lead/lag multi pump setups)
( I've "ladder coded" quit a few multi pump PLC's in this fashion … it load shares all the pumps equally over time).
I included a "HOBBS Meter" as well as the vast run time dynamics could be 1 hour or several months of just one pump's run cycle. I coded the PLC to put the longest pump time run last on the rotation until all the other pumps eventually had the same run time on them. (load/work sharing)
0 -
In both of these scenarios you are inputting the same BTU/hr without storage so I think your at-temp flow rates will be the same. Now if you started off with a well insulated tank that is holding 140f or more water that would add capacity.
0 -
One BTU will raise the temperature of one pound of water by one Fahrenheit degree.
If you take the flow in gallons per minute, multiply by the temperature rise in degrees F, and multiply by 500*, you get BTU per hour. Regardless of the flow, you're going to be constrained by the BTU/hr that the burner is capable of producing. Increasing the water temperature doesn't increase the output of the burner. If your burner is rated at 199K BTU/hr, flow in GPM times rise in temperature can never be bigger than 400. You can get 10 GPM at a 40F rise, or 5 GPM at an 80F rise, or 20 GPM at a 20F rise, or any value in between.
If you want more BTU, you either need a bigger boiler or more of them.
*(There are 60 minutes in an hour and 8.3 pounds of water in a gallon. 60*8.3=498. 500 is close enough)
0 -
I lost you then : (
Lets take a single 199K tankless unit and set the tankless set temp to 185 degrees F. (@ a max 3 gpm flow demand)
Run its output into a mixer valve (its out temp set to 125 degrees F)
Roughly, doesn't this significantly increase the total available (mixed) hot water capacity (@125 degrees F) over just the single tankless unit at its max flow rate of 3 gpm ?
0 -
You say "pumps" but tankless heaters usually just rely on water pressure. So does it have motorized valves or something similar to control which heaters get flow? And a pressure sensor to open more valves when the pressure drops?
0 -
yep! … all of them you mentioned and more. (a PLC doesn't know or care if its counting a digital pps output from a turbine water flow meter or counting chickens crossing a light beam) : )
0 -
"Regardless of the flow, you're going to be constrained by the BTU/hr that the burner is capable of producing. Increasing the water temperature doesn't increase the output of the burner" - DCContrarian
These are tankless "modulated" (ie: variable BTU flame output) units
0 -
Right, but they have a maximum output.
1 -
Closest analogy is a car engine. If the engine is 400hp, doesn't matter if you are in 4th gear or 5th gear, it can only ever put 400hp to the ground.
If you want more hot water, you need a bigger burner. Or you need a buffer tank somewhere.
With a buffer tank is where you can use this 180F output as now you can fill the tank up with 180F water. That will now run much longer to supply 120F water through a mixer.
Some Rinnai units you loose half the warrantee if you go above 140F, don't know if the case here.
0 -
Missed design, bad control logic, maybe scaled up?
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1 -
Probably up to about 6GPM, or whatever the unit is capable of delivering if it were set straight to 125
You still only have 199,000 BTU/hr (minus flue loss of course) of energy being input into the water so while your flowrate mixed from 185 down to 125 will be greater than the flow through the unit to achieve 185, it won't be greater than the flowrate of the unit just set to 125, there may be some already warm water in the piping of course which could change this value but since this is mostly theoretical and you would only do this during peak steady state output we can safely assume the piping system is already tapped of any theoretical capacity it may have held. With a tank set higher than demand you allow the system to get a head start, with constant flow set above the temp needed you don't have that luxury, unless you kept the entire piping system at that raised temperature, but that seems a bit sketchy at best, and possibly dangerous, at that point you would just want a storage tank to cover the extra needed capacity for peak demand
0 -
sure if you start out with a supply of 185 degree water you will get more hot water than if you just have the burner starting from cold. but without putting more energy (bigger burner etc) into the water, you are limited by the energy you are putting into the water. If you can find a way to take the gas fired tankless 199kbtu input and get 300kbtu out of it then patent that because you will be a billionaire
0 -
You need a real tank, not a virtual one, for that to work.
You don't seem to be getting that the burner is constrained in the number of BTU's it can put out. Heating the water to 185F and mixing it back to 125F takes the same number of BTU's as just heating the water to 125F in the first place.
The graph you posted above, 199DeltaT.png, tells the whole story.
0 -
"If you want more BTU, you either need a bigger boiler or more of them" - DCConttrarian
I don't want or expect more BTU generation than the tankless units are designed for……
I want to take the generated 185F and mix it with cold water down to 125F to emulate a larger deliverable volume of hot water (@125F) seen at the POU.
0 -
Yes if you STORE a volume of 185 degree water you can use that to get more deliverable volume, until that stored volume runs out of energy and then you are once again limited to the units energy output, which puts you at 125f at 6GPM deliverable water
0 -
It doesn't work like that.
With a fixed burner, doesn't matter if you make 120F water or 180F water. Once mixed down to 120F, you will have the exact same GPM out.
0 -
with a stored volume you can draw as high a flow rate as you want until the stored volume is gone (the energy is already in the water), an endless supply requires energy to be transferred as the water gets delivered, so you are strictly flow limited to the burner capacity to keep the water at temp during delivery. If you combine the two, a stored tank of hot water, with a tankless, you can enjoy "endless" hot water, and have that stored energy buffer to cover peak demand times. You can go back and forth on this all you want, but bottom line is without storage you are 100% limited to the burner capacity of your heaters. You really need to think about this like you are transferring energy, because that is what you are doing here, and with gas fired appliances you will not be getting more energy out that you put in.
2 -
Hi, This has been an interesting discussion to watch. Now I'd like to offer up some thoughts. First is to know what you're aiming for. I'd want 100% guest satisfaction first. I never want them to come to the office and ask for a discount because there was a hot water problem. Next I want lowest life-cycle cost, so when all costs are added up over the life of the equipment, (or some given time period, like ten or twenty years) it has been less expensive than any other approach. Also, I want the system to be simple and easy to understand and maintain. If you can never leave because nobody else understands it… that's not great.
With that understood, I'd start by making distribution efficient. Good 1.5 gpm showerheads can be had. Extra heavy insulation for the recirc line can be added so you can downsize the pump for low piping heat loss… Just to throw some crazy into this, If you have 60 psi, it's possible to use 1/4" tubing to supply your 1.5 gpm fixtures. I have a letter from an engineer saying it meets code! That would save you energy and water while giving faster hot water delivery. Now, put a data logging water meter on the supply to the existing heater/s and actually measure gpm, minute by minute. You'll see maximum flow and be able to better size recovery, knowing what the actual numbers are. With that, I'd install two tanks in series, valved to isolate either one. They would be sized to handle the peak. Heat source depends on lots of things. If you could use commercial heat pumps, that would probably be nice. If you could reduce load further by using drain water heat recovery or solar thermal, good. So much of this depends on local energy rates now and future, and physical conditions.
Anyway, the hot water system shouldn't be a job. It should be as simple as possible and give the best service with the lowest lifecycle cost. That might be three cents 😏
Yours, Larry
3 -
It will be easier to see if you think in terms of BTUs and BTU/hr. The CX199 you included the chart for can provide 11 GPM at a 35F rise, 11*35*500=192,500, it can provide 192,500 BTU per hour. Eight of them can provide 1,540,000 BTU/hr.
Your short term peak demand is 65 GPM at 125F, you say your water comes in at 60F so that's a rise of 65F. 65 GPM *65F*500 = 2,112,500 BTU/hr.
There is no way to get 2,112,500 BTU/hr out of a collection of water heaters rated at 1,540,000 BTU/hr. It doesn't matter what the water temperature is.
Now, you could store some BTU's during low demand periods for use later. Let's say you need to be able to cover two hours of peak demand. 2,112,500-1,540,000=572,500 so you're short 572,500 BTU per hour, two hours of that is 1,145,000 BTU. That's how much you need to store.
Let's say you do figure out a way to store it at 185F, which is 125F above incoming temperature. 1,145,000BTU/125F= 9,160, you need to store 9,160 pounds of water at that temperature. At 8.3 pounds per gallon, that's 1100 gallons of water. To put that in perspective, if you know what a 275 gallon fuel tank looks like, you'd need four tanks that size.
I think you would run into a lot of practical difficulties maintaining that kind of storage. Another alternative would be to look at conventional storage tanks, although they'd be big ones and a lot of them.
Or, you could just have enough tankless capacity to meet the peak load. You'd need 11 of them instead of eight.
1 -
An energy management engineer I know, specializes in retirement homes, first thing he does is put a portable ultrasonic flowmeter/ data logger on the supply to the water heating device. Leave it over a period of time. Fo retirement homes, 30 days is adequate. For a hotel you would want to cover peak season.
This gives you hard, stable data to design around.
All the water heat manufacturers have sizing tables. Some based on Hunters Curve developed in the 1950's!
So using a more current sizing program helps get the numbers more realistic.
As @Larry Weingarten mentioned, lower the load first. Un or poorly insulated recirc lines are basically a hydronic heating circuit, a huge energy consumption.
Smart circulators with thermal balance valves, if you have multiple loops or floors maximize recirculation energy, both the pump and piping heat loss. Quality low flow shower heads are a no brainer.
Keep in mind the hotter you run the boiler, storage the greater the mineral precipitation. So more scale removal maintenance. Storage tank life shortens with higher temperatures also.
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1 -
" …. but bottom line is without storage you are 100% limited to the burner capacity of your heaters" - GGross.
My proposed paradigm for this setup never stated I intended nor could exceed the maximum BTU design production capability of the tankless unit!
I'll try to summarize it again in a different way,
(lets take the return "keep alive loop" out of the equation ….. as we know this is perpetual heat (BTU) loss demand 24/7 designed into the system to afford an instant hot experience at the POU)
No demand = no firing
.5 gpm to 6 gpm demand = Only one tankless out of the eight fire (modulated) (changing to one of the others on a "round robin" rotation per each new call for firing)
6 gpm to 12 gpm demand = Two tankless units firing (modulated) + (group rotation)
12 gpm to 18 gpm demand = Three tankless units firing (modulated) + (group rotation)
……. this same group expansion continues out to all eight tankless units maxing out at 48 gpm @125 degrees F. (mixer contributes nothing at this time)
V (amended)
All the tankless units are "set-temped" at 185 degrees F. …… but I dont allow the combined firing output temperature to rise over 125 degrees F at this time (via external temp sensor on the mixer ….. shutting down the tankless programmed set temp firing)
Now, my normal week long (and plenty of hot water for everyone) peak demand (48 gpm) is exceeded by a 15 gallon per minute additional demand for an hour or two on a busy Saturday night.(63 gpm demand)
Note: all the tankless units are full firing (modulating) at this time.
Note: the gpm and temps listed below are for simple explanation only (not for a true correlation)
If excessive demand flow is 1 gpm = allow the combined tankless temperature to rise to 126 degrees F. (mixer adds cold water to drop temp to 125F)
If excessive demand flow is 2 gpm = allow the combined tankless temperature to rise to 127 degrees F. (mixer adds cold water to drop temp to 125F)
If excessive demand flow is 3 gpm = allow the combined tankless temperature to rise to 128 degrees F. (mixer adds cold water to drop temp to 125F)
…….. this same pattern continues up to the max. temperature rise of 185 degrees F maximum. (the mixer contributing gpm flow to equal the total excess demand and maintain a 125 degree F POU temperature at all times)
…… a "virtual" hot water storage tank" ?
0 -
just a a side note, thermostatic mixing valves need at least 20 degrees hotter hot side than the mixed outlet to perform accurately. Some models want 20 degrees
So you wont get 125 outlet with 126 supply The valve will hunt all over a temperature range and you will get customer complaints . Cold, hot, lukewarm the valve will never settle in
Shower valves that are pressure and temperature balance would help they are $$
An electronic mixing valve that doesn’t depend on a thermal “motor” is a better option
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1 -
A small side note. You don't want to be mixing down the output of a tankless unit. Set it for the temp you need, you'll get less scale buildup in the unit. There is no GPM boost benefit by doing so.
The only time to add a mixing valve is for some combi heat applications that need hotter SWT for the air coil.
0 -
"Set it for the temp you need, you'll get less scale buildup in the unit" - Kaos.
Scale build up is no longer an issue for me nor the required 6 month pumped vinegar scale removal maintenance.
I've designed a special Carbon + TAC scale treatment media filter that doesn't allow scale to even build up in the first place and removes any scale deposited earlier. (Media design life expectancy is 5 to 8 years completely unattended)
0 -
Is this copper tube piping? Hopefully those peak DHW flows are very rare. 40 gpm for 2" copper is about the maximum flow.
100 gpm in 2" tube, is off the charts! That tube must be screaming at velocity fps (feet per second) around 10 fps with 100 gpm flow in 2" tube.
3" copper tube at a 100 gpm flow rate is around 5 fps
At around 5 fps in copper tube you start to hear the flow noise.
These guidelines from The CDA appear in most plumbing codebooks now.
Pinholes and fitting erosion would be the indicator of excessive flow velocity. Also water hammer when you close off a valve against high velocity flows, fast acting valves like solenoids for example.
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream1
Categories
- All Categories
- 86.6K THE MAIN WALL
- 3.1K A-C, Heat Pumps & Refrigeration
- 54 Biomass
- 423 Carbon Monoxide Awareness
- 100 Chimneys & Flues
- 2K Domestic Hot Water
- 5.6K Gas Heating
- 101 Geothermal
- 158 Indoor-Air Quality
- 3.5K Oil Heating
- 66 Pipe Deterioration
- 934 Plumbing
- 6.2K Radiant Heating
- 384 Solar
- 15.2K Strictly Steam
- 3.3K Thermostats and Controls
- 54 Water Quality
- 43 Industry Classes
- 48 Job Opportunities
- 17 Recall Announcements