buffer tank piping

hoosierdaddy

We are doing an Engineered LEEDS radiant heating system that has the attached piping diagram for the boiler / "buffer ' tank / manifold piping. It is my understanding that the the main purpose for the tank is to add the solar heat which is by a seperate contractor. My concern is that the diagram shows the water off the top of the tank supplying the zones but also going to the inlet of the boiler. Will that not send the hottest water to the inlet of the of the boiler forcing it to heat water from 140 upward which defeats the efficiency of the Knight boilers? On the other hand if I take the water from the bottom of the tank will the tank heat from the top down to the sensor located in the center of the tank? Maybe I'm just overthinking the process but with all the emphasis on efficiency I want to be doing what works best not just what somebody drew from older diagrams? 

NRT_Rob

forgot something

no diagram attached?

Ben_12

try again

Fortunately I'm a better heating guy than computer guy. I am going to try again to post the diagram again.

NRT_Rob

hmm.

I don't like this very much. They are doing several things I don't like, including heating a solar vessel with fossil fuel, and maximizing the return temp to the boilers.



I don't know who is doing the solar, but in this case, it can't even transfer into the system unless it's hotter than the supply water temperature requirement for the radiant... which is apparently fairly high temp, at 140.



Trying to do double duty as a solar and a buffer tank is pretty iffy in my mind. I'd much rather see a dedicated buffer tank, and a dedicated coil in a solar tank for heat extraction, piped into the radiant return. Any sharing of the buffer tank and solar would then mean that solar can only operate at radiant temperatures or higher, which will reduce its collection ability and efficiency.



but it saves the cost of a tank to do it this way, I presume. seems like a lot of penalties to pay for first cost economy though.

Ben_12

more info

I really don't see the need for a "buffer" tank on this job. The only time I've used one is with a bang bang boiler that was feeding four very differently sized reheat coils. This job has five fairly equal radiant zones with two 105,000 btu Knight boilers. The building management system controls the outlet temperature of the two mixing stations based on outdoor temperature and runtime. The bms also controls the boiler temperature by the requirement of the highest operating mixing station. These are the reasons I'm especially concerned by the bms probe being located in the middle of an 80 gallon tank.The bids are already in so I have these parts to accomplish the sequence of operation. Moving connections and parts around will not change my costs but adding anything else would. So the puzzle is the most efficient and flexible way to use the listed parts to accomplish the listed sequence. The best response to this wins nothing..

Thanks for all interest and plans of action

NRT_Rob

can the BMS

synchronize zone firing? if so, maybe skip the buffer tank.



or if the zones are large, or high mass.



buffer tank just controls cycling, so any time you have lots of little demands... say, 5 low mass zones in the shoulder season... it can help efficiencies.

Karl_Northwind

how about

it seems that the circulators on the floor loops are pumping towards the mixing valves, doesn't seem right,  I've never done that. 



I'm concerned at the way the boilers are tied into the solar portion.  if a buffer tank is required, plumb it up and stage the boilers into it, floor loops out, and have the solar (whatever means you have of moving the solar heat into the heating water) staged as the first stage in the heating system.



not having any idea of how the solar is configured, and the mass of the heating system (120F to air handlers or staple up? 140F to panel radiators or air handlers?)

it's hard to tell what the best integration method is.  is it a drainback solar with a smallish(compared to the panels) storage tank that means we have to get rid of the heat (to the building) or lose it, or a pressurized solar system with a big atmospheric tank that will store a lot of heat to be used later?



seems like a sub- optimal heating integration situation for solar.  get the heat into the lowest temperature portion of the system (IE the return from the floors) and use a MMV with ODR to use the lowest temp possible to meet the heating need.



on second look, it's possible that the drawing simply has the connection points for the boiler in the wrong lines.  if you reverse the supply and return connections from the boiler to the distribution system it seems to make more sense.  it would be possible that the floor circulators would have enough oomph to move the water thru the buffer tank and not the boilers (flow checks) and only heat from the buffer tank(heated only by solar) unless the boiler circulators turned on.



if you still circulated water thru the floor loops at whatever the solar can provide, not mixing it down, because it will most likely be lower than 120, you can still deliver BTUs although at a slower rate, and extend the time between boiler cycles.



can you give us more info on the building (SF? heat Loss?) and solar system? (how big? how many gallons storage? )



Karl

NRT_Rob

return mounted mixing valves

work just fine.



also if you flip the connection points around you'd still need a solar tank transfer pump and p/s to the tank, or you have the same issues you have now... slightly reduced as you heat the "buffer" with return water instead of supply water.

Karl_Northwind

mixing valves

funny, one of the first things I learned from a mechanical engineer who designs large commercial HVAC systems is you pump away from the mixing valve, and I've never seen it drawn or installed any other way.   the other was that every heat source needs a PRV or T&P valve that can't be isolated from the heat source.



well, there's my new thing learned for today. 



I'm just trying to make sense of what the designer of that system is thinking.



Rob, do you prefer to stage the solar as stage one into a primary loop and then feed all the low temp loops off that? what do you do when there are multiple temp requirements on that primary loop? shut off the solar source and fire up the boiler?



I've always just made sure that the coolest water from the floor loops goes thru the solar tank, and reduces the time between the heating calls for the boiler to heat the buffer tank.





karl

NRT_Rob

it's not as ideal

kind of like "pumping away", raising the pressure of your loop field is better than lowering it. but with modern air separation and good valve Cv it works fine.



things might be different with commercial size equipment though.

Karl_Northwind

you beat me

you beat my inevitable 2 minute later edit to clarify and add to my post.



what thinks you about integration, see above>

karl

NRT_Rob

I'm a dull boy.

I always do the same thing: non-maintained solar tank, boost for DHW with electricity as needed, "return water reheat" for solar contribution to heating.



I want to keep my solar tank as cold as possible, to collect as much as possible, and use every single BTU it collects to the max.



the only thing I've seen that can do it better is "overcharge" techniques... then you don't need a heat demand to dump heat into the envelope. But without a dedicated emitter it can get tricky.



For this system if he doesn't need a buffer tank, I'd simply pipe it to the boilers like normal, and use a dt control and 3-way diverter to either run return through the buffer or not if the return water were cold enough to pull heat from the solar tank. High limit valve on the buffer extract loop maybe, just to make sure you don't fry a floor or anything.



then ditch the motorized mixing valve and let the boilers do the reset they are built to do directly.



this is forum design though, maybe there are design characteristics I don't know. But that would be my starting point I think.

Ben_12

more info

This building is constucted with some foam form concrete walls and the rest of the walls and roof are structural panels on timber frame. All of the heat is slab radiant and the 5 zones are similar in size. The two Knight boilers are 105k btu and the storage tank is an 80 gal. and when bid the coil in the tank was for future solar but they have since selected a contractor for that. The reason the mixing stations are shown pumping away from themixing valves is because that's the way Viega builds their manifold mountable mixing stations. I have since sold the customer on Alpha pumps and will build my own "stations" pumping away from the valve. The bms sets the temperatures shown at 0 modulating to 80 degrees @ 60 degrees. If a zone runs for more than an hour the station will bump up 5 deg per 1/2 hour.

I am sure the bms/boiler temperature probe should be located near the thermometer above the tank and just ahead of the inlet to the mixing stations. I am also confident the outlet from the boilers should be where the present inlet to the boilers is shown. My big question is the best location for the inlet to the boilers. Which would keep the tank the coldest, a closely spaced tee on the outlet of the tank or where the outlet to the boilers is presently shown at the inlet to the tank? I know that the colder the tank the more value they will get from the solar.

I'm sorry I wasn't able to interject ealier but 9 hours of running pex plus 4 hours of driving and an hour helping my 80 year old dad work on his truck kinda shot the day.

Thanks again for your thoughts. 

      

Ben_12

any opinions?

I just wondered if anyone thought there would be an advantage to the closely spaaced tees after the tank or just reversing the connection points shown. It seems the tees would only allow the water returning from the slab to enter the tank. This would also take the most advantage of any solar heat since that heat would tell the boiler not to run once the sun had produced enough heat to meet setpoint. On the other hand, if the water from the slab could go directly to the boiler inlet there would be an even smaller amount of boiler produced heat entering the bottom of the tank but at any time the primary pumps were flowing more than the secondary pumps the tank would start heating from the top down. Which do you think is best?