Indirect Water heater as Pseudo Buffer Tank

thermodynski

Hello all,

I have a low temp zone in northern Minnesota that will only draw about 8K BTU's in the depths of winter and my existing boiler (Lochinvar 80) will modulate 5:1 down to 12.6K output. Based on my understanding of the system, I need a buffer tank to prevent short cycling of my boiler, even with ODR.

I'd like to use either an indirect or a reverse indirect water heater in a dual use buffer tank and "pre-heater" for my conventional gas fired water heater. I'm hoping my water heater will not have to run as much if it is supplied water temps above the 50F it's getting now. Seems like a good way to use some BTU's in my buffer with no need for any controls or pumps besides a TMV.

The tank will be on the end om my primary boiler loop (2 pipe) with a secondary loop containing my 3 zones. That way all the BTU's go to the zones first and any extra btu's will bypass into the buffer.

My thinking here is because the tank is not sitting at a specified temp (like a normal DHW setup) it's main purpose is to be a heat dump for when my zones can't take any more.

It shouldn't matter if the the water goes through the "tank" as in a reverse indirect setup (Thermomax or Ergomax), or the "coils" as in a conventional indirect (traingle tube, or many others), because the water in the coil would still dump heat into the domestic water prior to the water heater.

There's more benefits in doing a reverse indirect (decouling, dirt and air removal, stratification, bigger thermal mass for the system to pull off of if needed), but the costs I'm finding for those tanks are much higher, which makes a standard indirect more appealing.

What am I missing here?

GroundUp

So you want to preheat your DHW in this indirect and still have it feed the existing DHW tank in a series loop? I guess if it were mine, I'd eliminate the existing water heater and go with a dual coil indirect tank. This way the boiler could maintain a certain tank temp and feed both the DHW using the stored volume in the tank as well as the radiant via the upper coil. Such an arrangement would also act as that buffer you need.

thermodynski

Ground up,

Sounds like you understand.

I'm trying to keep this on the cheap if I can. We're finishing the basement and We'll be moving in 2022 so I'll never see a payback on improved efficiency, just want to not make a headache for the next guy. 

There was a certain appeal to finding a used indirect for a couple hundred bucks, and just running it as opposed to re-desiging both my hot water and hydronic systems for indirect hot water and tossing out my perfectly serviceable hot water heater.

Having a bottom coil like a typical single coil setup will have shed head well but not supply heat as well because the water will be colder down there, but I'm willing to accept that if I can save $1000. 

Hot_water_fan

What’s the low temp zone? Radiant floor? Radiators? 

And what is the total heat loss of the house? 

There’s a possibility if the total heat loss is high enough that even a micro zone won’t be a problem if there’s a high chance of overlap with other zones. Similarly, if it’s a high mass micro zone, run times could still be long.

if you do decide on the indirect, toss the tank heater, that money is gone. 

hot_rod

Which Lochinvar do you have? The Knights have a few functions that help minimize short cycles
Ramp delay keeps the boiler at low fire for a period of time, adjustable,

Anti cycling prevents the boiler from refiring after shut down for the period of time you select

If you already have a DHW tank, just add a plain steel buffer tank, if you really need one

An electric water heater or a Boiler Buddy would be less$$

EBEBRATT-Ed

What @hot_rod said

Just add something for a buffer and leave the WH as is or get an indirect and ditch the WH.

The lochinvar is probably more effecient that the gas fired wh

thermodynski

HWF & HR,

I have a Knight 80,000btu from 2010. Floor mount Mod Con with 5:1 turn down and their SMART control.

low temp zone is radiant in the walls.

design total heat loss (I designed for -40 in Duluth, MN) is about 80,000 btu/hr in my leaky old house. Looks like the previous designer was in the ballpark. I have a lot of south windows so it gains a lot fromthat in the day but higher losses at night, 

I had considered that the upstairs zones will likely call at some point and extend cycle time, especially once it's a little bit cold, but when it's in the fall and all 3 zone heat loads are pretty small I think I'd still have some level of short cycling even ramp delayed down to 20%. 

It looks Like there are some controls aimed at a min cycle time and ramp delay. so so maybe I can do something there and limit the setup to 10 minute cycles. 

Any recommendations for a steel buffer tank? I looked at boiler buddy and they cost more than an indirect where I was looking. An uninsulated steel tank would work for what I'm doing and serve as a heater for the utility/laundry room it's in. 

Would it be wrong to install without a buffer tank and then "wait and see" how short the cycles are? Seems like a lot of work to re-pipe it in if my calcs are right and it's needed. 

thermodynski

Ed, I have no doubt the boiler is more efficient than my WH, but my WH supplies all the hot water we need and it's an "if it ain't broke" situation. If it breaks I'll be putting an indirect in for sure. 

GroundUp

What is the BTU output for these small zones, exactly? Can you maybe tie them together and eliminate microzoning altogether?

thermodynski

My plan is to have one zone per floor on a 20x30 "foursquare" house.

Right now it's 1 zone pumped by a taco 007 for the main floor and 2nd floor and the basement is un-heated. ODR is not enabled so I've been manually turning up and down as weather demands. when we moved in it was set at 160 all the time. I get a 10 degree DT with this setup.

I have old gravity piping that I'm pulling out and converting to pex in the basement to get my headroom. I'll be using 2 Taco 007 VDT's on the cast iron rads and moving my 007 to the radiant loop. Looking at 20 DT on the rads and 7 DT on the radiant.

1st floor 44,500 (higher ceilings , 2 doors and a sunroom bump out) - cast iron rads
2nd floor 26,500 - cast iron rads (3 bedrooms and a bathroom)
Basement 8,500 - in wall radiant (1 bedroom and a bathroom)

=79,500 or the 80,000ish I mentioned earlier.

I'd like to be able to independently control the 3 floors, I could add some loops to the ceiling of my tuck under garage and shed some heat in there, but I'm a little worried about those loops freezing when it's -40 as it's pretty much uninsulated and I'm not planning on insulating.

I could maybe just add on the basement zone to the main floor but It seems like the heat loads are so different it'd be too hot/cold somewhere.

Seems like "Just try it" with ODR, some ramp delay and cycling control might be the right answer if such a thing existed.

Hot_water_fan

Hm how much gas did you use last year? Seems unlikely your heat loss is above the boiler output, which is what the calculation says. That smallest zone might be smaller than you think. That’s a lot of pumps too! 

hot_rod

It sounds like you will need two temperatures
One for the radiant in wall, maybe around a 120 set? Then a high temperature for the cast radiators.  The delta t circulators do not regulate SWT, not sure what they add to a system like yours?

imIf you ever plan on TRVs on those radiators a delta p circulator is what you want,
Either a buffer tank or a hydro Sep would be the best piping option

Only rule of thumb for when to add a buffer. Calculate the smallest load that could be running. How does it match the lowest turn down of the boiler?

A goal would be to have at least a 10 minute run time under that low load condition

thermodynski

Yep 2 temps.

High temp: 180 max and odr. Ill set my odr to the min (180 at -30) and see how well it matches up 

Low temp: parallel the high temp with a max of 130 or less. (Manual or Thermostatic mixing valve) leaning Tekmar or danfoss manual because CV is higher. 

I'm set on pumps because my father in law is helping with the piping and that's his preference. but yes dletaP and TRV's is a slick solution. If I was starting from scratch that's how I'd do it. I do have uponor EP manifolds that have flow meters. That's how I was planning to balance this thing.

I got the DT's because they can speed up or slow down to try to hit the right DT. They should save a little on power but I think they should help keep my return temps lower and boiler condensing more than 007's. Hopefully they also can help with some of the educated guesses I had to make on my actual heat loads. That's the idea at least. 

Hot rod, my calcs show I would be on for about 3 minutes in my small zone, which brought about this post.

I'll have to look into my gas costs and get back. The boiler is sized for no basement heat and I added that in. which puts output the same as demand. Also my design day is -40 which is usually 1 or 2 nights a year up here. 

Anyway to get back on topic, sounds like what I originally proposed could work, but it may not be needed since my boiler has some smarts. 

GroundUp

Frankly, it sounds like you're going to do whatever you want to do regardless of what we tell you. Your boiler has only about 70,000 BTU to give at design temp, for an 80,000 BTU load. That's not enough, if your calcs are accurate, which I doubt because 74 BTU per square foot is unheard of. And yes, I work in Duluth all the time- I know how cold it gets with the lake effect. It would be easy enough to simply add some more emitter to the basement zone and eliminate all the buffer crap. 5000 extra BTU into the basement is not going to be noticeable from a comfort perspective, but the boiler will thank you.

hot_rod

So, you have 3 levels of 1200 sq. ft. each? Your main level needs 36 btu/ sq ft?
Which boiler do you have, the Knight 80 has an input of 85,000, 79,000 output in condensing mode. At higher operating temperatures, like 180, more like 72,000 output.
So cutting it close, depending on you heatload accuracy.

When you pipe it in plan on the possibility of needing a buffer, if in fact you see that smallest load operating by itself often?

I’m not convinced a constrained delta t operation is a good fit for a high mass, multi temperature multi pump system on ODR.? Maybe take a read through Idronics 23 for thoughts on the concept.

mattmia2

If you haven't actually put anything in the basement yet you could use higher mass emitters sized for a low water temp so they would heat more slowly and wouldn't overshoot once they got hot. Maybe a combination of radiant and panel rads or salvaged ci.

thermodynski

I had a look at changing over to valves for zones and TRV's with a deta P pump. I had been seeing each TRV costing $100+ when on reality they are much more affordable. I'll kill this buffer tank thread and re-start with a whole system design for everyone's review.

mattmia2

There is also a valve for panel radiators with a bypass built in that would let you divert some flow to the radiator will maintaining a minimum flow through the loop, if the TRV closes it becomes just the bypass flow.