two stage parallel storage tank system?

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

semlin

New Member
Nov 23, 2009
28
okanagan, bc
i may be using the wrong term, but is there any discussion here about using a segregated and staged tank storage system to avoid mixing down temperatures? i am thinking of a system where the total volume of water in the system is, for example, equal to the system plus one tank, but two tanks are used connected to the system in parallel, so that one tank of hot water completely drains into the system and the return water fills the second empty tank which then switches roles to be the feed tank. i am assuming you could do this with float valves and pressurized air bladders in each tank that were connected together so that one would expand as the other contracted.

the advantage would be maintaining higher operating temperatures longer and hence higher radiator efficiency for longer.

i suppose the disadvantage would be higher initial cost and space used for the volume of storage achieved.

i may have completely missed the discussion, or it may be impractical or not useful for some reason i am not seeing.
 
Seems a good technique for maintaining high efficiencies while charging storage. I am aware of "sequencing the charging of thermal storage in some solar hot water systems. These systems make use of diverter valves to allocate which tank is recieving a charge.
 
Seems at first look not a bad thought, by it totally ingnores the "KISS" principal. I see the advantage of the idea, but it looks like a control and maintenance nightmare. I could see using a taller well stratified tank with layered coils to charge and discharge. There are controls available that have this allready in their standard program.
I would do some searches on stratification on this site and the net, you may find higher efficiency systems with a lower headache factor. On the farm we seem to have much interferance from Murphy (Murphy's Law) if can go wrong, it will go wrong but only when in a hurry or when no parts are available.

Good luck,

Henk.
 
hogstroker said:
Seems at first look not a bad thought, by it totally ingnores the "KISS" principal. I see the advantage of the idea, but it looks like a control and maintenance nightmare. I could see using a taller well stratified tank with layered coils to charge and discharge. There are controls available that have this allready in their standard program.
I would do some searches on stratification on this site and the net, you may find higher efficiency systems with a lower headache factor. On the farm we seem to have much interferance from Murphy (Murphy's Law) if can go wrong, it will go wrong but only when in a hurry or when no parts are available.

Good luck,

Henk.

thanks

i am just blue skying the idea here. i agree it would need to be done with robust commercially available parts and with a minimum of controls.

i understand stratification in general, but i do not closely understand its efficiency relative to a steady flow of optimum temperature water. i would think the key would be the radiator efficiency. when your radiator efficiency increases in a non linear fashion as temps increase then even modest temp drops in a stratified tank have magnified impacts. i think you also have to consider how often the tank/boiler would cycle. if the system is more or less constantly circulating until the boiler restarts, then i think the advantage would be less, but if the system idles then you have bigger temp drops in the system water idling outside the tank that will enter the tank as return water and more time for that colder water to diffuse and drag down temps inside the tank.
 
semlin said:
...

i am just blue skying the idea here. i agree it would need to be done with robust commercially available parts and with a minimum of controls.

i understand stratification in general, but i do not closely understand its efficiency relative to a steady flow of optimum temperature water. i would think the key would be the radiator efficiency. when your radiator efficiency increases in a non linear fashion as temps increase then even modest temp drops in a stratified tank have magnified impacts.
Certainly a radiator will be more powerful when supplied with hotter water, but I would think efficiency would be more a matter of the amount of heat energy delivered per kW-hr of pumping energy.

i think you also have to consider how often the tank/boiler would cycle. if the system is more or less constantly circulating until the boiler restarts, then i think the advantage would be less, but if the system idles then you have bigger temp drops in the system water idling outside the tank that will enter the tank as return water and more time for that colder water to diffuse and drag down temps inside the tank.

A stratified tank won't mix much, if any, except as a function of the conductivity of the tank walls. Consider a meromictic lake that stays stratified for centuries on end.

Some tanks use low-heat-conductivity composite material walls or an insulating layer on the inside the walls to combat heat conduction across the thermocline. To me a cheap tank is more important than worrying about it.

The most critical factor I can see is minimizing return temperature from the heat loads. Tekmar and others have got this quite well figured out with four-way mixing valves and variable speed pumps.

Cheers --ewd
 
ewdudley said:
Certainly a radiator will be more powerful when supplied with hotter water, but I would think efficiency would be more a matter of the amount of heat energy delivered per kW-hr of pumping energy.

i think you also have to consider how often the tank/boiler would cycle. if the system is more or less constantly circulating until the boiler restarts, then i think the advantage would be less, but if the system idles then you have bigger temp drops in the system water idling outside the tank that will enter the tank as return water and more time for that colder water to diffuse and drag down temps inside the tank.

A stratified tank won't mix much, if any, except as a function of the conductivity of the tank walls. Consider a meromictic lake that stays stratified for centuries on end.

Some tanks use low-heat-conductivity composite material walls or an insulating layer on the inside the walls to combat heat conduction across the thermocline. To me a cheap tank is more important than worrying about it.

The most critical factor I can see is minimizing return temperature from the heat loads. Tekmar and others have got this quite well figured out with four-way mixing valves and variable speed pumps.

Cheers --ewd

thanks for the thoughts

isn't efficiency the number of btus emitted by the radiators per btu of heat input? by that standard, if you heat water to say 180 degrees and store it and then deliver all that water to the radiator at 180 degrees, surely that is more efficient than using exactly the same btu input to heat the same water to the same temp but then having some of that water mixed down inside the tank with return water?

i can't express it as a formula, but, supposing your return temp drop was 20 degrees, my thesis would be that X amount of water delivered in succession at 180, 160, 140 and 120, would emit more btus than if you mixed that return water as you went along. i am presuming that the increased efficiency of the radiator at higher temps means there is a significant advantage in circulating the first storage tank all at 180 degrees.

and again, i am assuming that storage tanks do not have very strong stratification performance. there has to be diffusion between layers and the fact there is water circulation would create further diffusion. this graph below by another member shows that in a 24 hour period there is a gradual linear drop in temperature in the middle layer of the tank. this suggests significant intermingling between return water and original water at the mid layer of the tank. if the layers were discreet you would expect abrupt significant temperature changes as a colder layer rose up in the tank

https://www.hearth.com/econtent/index.php/forums/viewthread/48079/P60/#544049
 
not to get too complicated, but another option with this approach would be to have three tanks containing two tanks full of water, and then rotate two storage tanks and leave the third tank as a fast heat reserve to quickly heat the house. that would suit our usage.

one of the design issues i have been trying to figure out is how to deal with the peaks and valleys of our usage if we went with wood and storage. we have cast iron radiators so we could theoretically use lower temp storage water on a second (or third) cycle but not the way we heat the house now. with a ng boiler, we heat the house to 68 in the morning, let the house drop to about 60 degrees for most of the day when we are out, then ramp up to 68 in late afternoon.

we looked into a geothermal solution here, and were advised that the low btu output at the radiators from geothermal temps would require constant operation and maintaining 68 all day because it would take forever to warm up and be very inefficient. that constant use actually offsets the benefit of switching from ng.

so far, wood with storage presents the same problem. i am worried that to milk all the btus out of the storage water i will have to heat the house all the time after the first tank cycle which will mean i will burn more wood than just the equivalent to my annual ng btu consumption because i will be adding more heating time.

anyway, please feel free to tell me i am crazy...
 
I'm burning with a Tarm Solo 60 and 1000G pressurized storage. We just had some 0 to 10F degree days and nights. Where the 60 can burn without storage I would start a fire mid afternoon and keep it burning with a final loading at 10pm. At 10 PM I had 1000G X 185F in tanks plus a load of wood to burn off before system would start using from storage. Yes the Tarm would cycle on and off (idle) during the night but tanks had plenty of BTU's for the next day. I heat primarily with a fan coil and need relatively hot temps to circulate through coil to bring heat up in house in the AM.

During 30 degree weather I have enough BTU's in the tanks from burning a 1 or 11/2 loads for a 24 hr period.

Type of stove might be a one alternative / solution for you or, adjust the volume of storage. No matter where or how you store it you still need to burn X amount of wood to have those BTU's available for house.

For me to build a fire in my Tarm and to start pumping water at 170F is takes about 30 - 40 minutes.

rob
 
wouldn't two tanks in series like " the simpleest pres stor" be the best system overall? I cant see any issuses with this setup. You can heat the top tank to the max then you start heating the bottom tank. by djusting lfow rates you should be able to have some stratification within each tank and extreme stratification between tanks. One other idea, I see discusions obout stacking tanks to create this setup, i don't think they have to be physically stacked. IMHO if they are plumbed as if they are stacked and they are pumped the water will not know any better. I built my system before I discovered this site my tank has two upper ports water is piped over both ports cooler return water drops into first port an hotter water is drawn out of secon port, there is a bottom port that feeds back into boilerand loops bacint the upper ports. It works pretty well but kwowing what I know now I would follow the stickynote. I just looks perfect in function and simplicity.
 
I have my 2 X 500 Gal. stacked and joined in the middle. Your correct that if you plumb in series you effectively get the same result. If you have room I wouldn't stack. It just adds to the complexity of the project. Additionally if you were to stack 1000G tanks you would need to be careful of the weight and points of loading. Tank to tank, tank to floor and the floor itself. Just the water weight is 8300 lbs. per 1000G.
 
i see what you mean that two tanks in series would likely mean you enjoy a hotter first tank, but a cooler second tank, in comparison to one big tank. it would definitely delay diffusion in the first tank. i guess the impact would depend on how your rate of water consumption compares to the rate of diffusion.

i believe you could delay diffusion even better by using vertical cigar shaped tanks in series so there is a smaller contact area between water stratums.

it would be very interesting to see a number of graphs of different tank set ups side by side to understand and quantify the real world diffusion and stratification effects
 
I think the issue of tank stratification is probably less significant an issue than a whole lot of other things when using a wood boiler.
It seems that a lot of import is placed on stratification research done on solar DHW tanks. To relate this to a wood boiler setup is not quite apples and apples.

If you are heating a house with a huge heat load and are using low performance hot water baseboards, yes, stratification is going to help, a little. But you are still going to be limited to the amount of stored btus. You might buy a little extra time in between firings but if the system is that stretched, you are firing more or needing a bigger storage system or you should be thinking about changing to a more efficient heat distribution system.

IMHO, if you want to focus on storage issues, insulate as much as you can--both the tanks and your home.
The lower you can run the storage tank the better, but that means a radiant concrete floor and a backup hot water heater.
 
Tom in Maine said:
I think the issue of tank stratification is probably less significant an issue than a whole lot of other things when using a wood boiler.
It seems that a lot of import is placed on stratification research done on solar DHW tanks. To relate this to a wood boiler setup is not quite apples and apples.

If you are heating a house with a huge heat load and are using low performance hot water baseboards, yes, stratification is going to help, a little. But you are still going to be limited to the amount of stored btus. You might buy a little extra time in between firings but if the system is that stretched, you are firing more or needing a bigger storage system or you should be thinking about changing to a more efficient heat distribution system.

IMHO, if you want to focus on storage issues, insulate as much as you can--both the tanks and your home.
The lower you can run the storage tank the better, but that means a radiant concrete floor and a backup hot water heater.

the basic design issue i have is that i do not need or use full heat all the time so i have yoyo consumption right now. if i went to a system that had to provide full heat all the time to maintain temperature i would need more btus than i currently use. since i have calculated i need to burn at least 7 cords of hardwood a winter to meet the current btu usage by my ng boiler, i am at the point of the wood consumption being so high it is too great a chore.

so i have been trying to think of a way for wood with storage to recover as well as ng in a system that drops indoor temperature by 8 degrees for up to 8 hours a day plus overnight, then reheats. the only way that can happen is if i have lots of 180 degree water available right through the storage cycle between burns. that got me thinking about parallel storage tanks, with a stand by "hot tank" used to quickly reheat the house then switching to regular tanks of cycled water that would hold that higher temp and also maintain the 60 degree "idle" temp. that in turn got me thinking about the hot tank idling most of the time and being vulnerable to diffusion from the return water, at which point i started this thread...
 
semlin said:
...

so i have been trying to think of a way for wood with storage to recover as well as ng in a system that drops indoor temperature by 8 degrees for up to 8 hours a day plus overnight, then reheats. the only way that can happen is if i have lots of 180 degree water available right through the storage cycle between burns. that got me thinking about parallel storage tanks, with a stand by "hot tank" used to quickly reheat the house then switching to regular tanks of cycled water that would hold that higher temp and also maintain the 60 degree "idle" temp. that in turn got me thinking about the hot tank idling most of the time and being vulnerable to diffusion from the return water, at which point i started this thread...

But there is such a thing as a storage tank system that doesn't stir the tank up enough to worry about. You keep coming back to the notion that the water will go round and round, losing 20F or so each trip, which does not have to be the case. The idea is that you can have a system in which the hot water leaves the tank just once and returns as cool as possible, completely 'used up', just once. One way of accomplishing this is to use a computerized control that runs a variable speed load circ as slowly as possible, while drawing hot water off the top of the tank as slowly as possible.

In this manner when it comes time to re-heat the space after a setback period there will be unused hot water at the top of the storage system.

Have a look at the variable speed pump controls from Tekmar and others along with the motorized mixing valves. Some of these systems can even go ahead and figure out how much of a head start is needed to get the space up to temperature at the required time according to past data and the outdoor temperature.
 
ewdudley said:
But there is such a thing as a storage tank system that doesn't stir the tank up enough to worry about. You keep coming back to the notion that the water will go round and round, losing 20F or so each trip, which does not have to be the case. The idea is that you can have a system in which the hot water leaves the tank just once and returns as cool as possible, completely 'used up', just once. One way of accomplishing this is to use a computerized control that runs a variable speed load circ as slowly as possible, while drawing hot water off the top of the tank as slowly as possible.

In this manner when it comes time to re-heat the space after a setback period there will be unused hot water at the top of the storage system.

Have a look at the variable speed pump controls from Tekmar and others along with the motorized mixing valves. Some of these systems can even go ahead and figure out how much of a head start is needed to get the space up to temperature at the required time according to past data and the outdoor temperature.

very interesting! i have been assuming the flow rate (and thus return temp) is not something i can mess with. if i can slow down the flow so as to milk the btu storage in one cycle then that would indeed solve most of the problem. can you do that with an older cast iron radiator system though, or would i lose a lot of heat in the feed pipes in the walls? another possible issue is that my system is a converted gravity feed with large main pipes. it will thus have some natural flow that i would need to stop. presumably i can use some kind of restrictor valve at the return.
 
semlin said:
very interesting! i have been assuming the flow rate (and thus return temp) is not something i can mess with. if i can slow down the flow so as to milk the btu storage in one cycle then that would indeed solve most of the problem. can you do that with an older cast iron radiator system though, or would i lose a lot of heat in the feed pipes in the walls?

In an older house with cast iron radiators, especially one that was designed originally to work with inherently slow-flow gravity feed, you can achieve dramatically low return temperatures. They generally liked to use a lot of radiator back in the day to begin with, and hopefully the house has been improved with better insulation and tightened-up as well, so that with any luck your radiators are now effectively several times as large as they need to be.

In which case you might not even need computerized variable-speed circ pump, maybe one that simply isn't any faster than it needs to be or maybe a flow rate regulator. Plus you can adjust the anticipators in your thermostats to shorten the cycle time of your radiators.

As for feed pipes, I would hope that heat lost from feed pipes would be heat gain somewhere useful within the house.

another possible issue is that my system is a converted gravity feed with large main pipes. it will thus have some natural flow that i would need to stop. presumably i can use some kind of restrictor valve at the return.

It is almost certain that when your house was converted from gravity feed some sort of 'FLO-CHEK' valve was incorporated into the system. This will prevent unintended gravity flow.
 
There was a recent link to a VERY high level theory paper looking at stratification in tanks... I'm still reading it, but so far it seems pretty clear that if one pays proper attention to the diffuser design at the suction and return ports so as to minimize turbulence mixing (something that many here have NOT done judging by their descriptions) you can maintain an extremely high level of stratification. I suspect that you would get FAR more usable heat out of two full tanks of storage, than you would from some sort of dual tank setup. The intro to that paper (use the search to find it, the discussion was on diffuser design and was within the past few weeks) mentioned the system you describe, and seemed to be of the opinion that a simple stratified system worked better.

Gooserider
 
If I recall correctly, that was a study done on a solar DHW system.
You might have feed water in such a case that is 40F in the winter into such a system.
This is not as significant an issue with a large wood fired system.

I would be skeptical believing that there are major savings with diffusers, etc on a wood fired system.
Btus is btus. The are more factors at play here than one might think of.

How is the DHW interfaced with two pressure tanks in series? Are you using a radiant floor or cheap hot water BB?
How frequently are you firing the tank?

I can tell you that I have significant stratification on a 345 gallon tank. But half of my heating load is DHW. My installation is only 3.75' tall.

An educated guess is that you might pull another 10-15% out of a properly designed stratified system. But I'll bet that
cold in the bottom and hot out the top is all the stratification design you need.
Baffles and diffusers are not going to give the kind of systems we discuss here even a 10% performance bump.

If someone is going to prove or disprove this theory for wood boilers, they need to do both parallel and series pressure tanks, with degree day data
and some serious thermocouple installations to sort it out.
 
Some guys are using propane tanks with the stock fittings at the top of the tank. If they are using high flow circulators, for whatever reason, they might be jetting hot water into the tank like an upside-down geyser and mixing significantly. But even in that extreme case when the hot water has filled the tank at the top after a while the incoming hot water is going to be mixing with hot water: no large deal. Only a phenomenon at the beginning of the charge cycle.

I think the more difficult design problem is what is returning from the system. If the water coming back into the tank from the loads is always colder than what is in the tank and it comes in at the bottom, there is no issue. Brain-free operation is my ultimate design goal. The easiest way to get this situation is with radiant floors, I think.

But if you have different types of emitters in the house and under various conditions have different temps returning to the tank you might be mixing from the bottom up, so to speak. And even with low temp return water to the storage tank if you want to add solar heat to it that can vary widely in temp and get you in the same pickle. I went through a number of attempts at designing a diffuser for my return water in case I wanted to add solar heat in the future. I finally gave up and decided the KISS principle was more important and if I did want to add solar hot water some day I would do it with a separate tank altogether.
I guess what I did was come to the same conclusion that Tom is diplomatically suggesting. I just did it without all of Tom's practical experience.

You can read Jebatty's test results of using a only-slightly-modified propane tank here. He gets pretty good practical stratification with a simple setup.

https://www.hearth.com/econtent/index.php/forums/viewthread/36503/

Everyone has to find their own point of diminishing returns, I guess. Just make sure you share your results with the rest of us. I'm going to if I live long enough to get this thing done and running.
 
I like simple. And a little common sense goes a long way.
And 30 years of stupid mistakes helps (sometimes) to figure out what not to do.
 
Tom in Maine said:
If I recall correctly, that was a study done on a solar DHW system.
You might have feed water in such a case that is 40F in the winter into such a system.
This is not as significant an issue with a large wood fired system.
The paper I was referring to is quite a monster, it was 44 pages, and seems to be more of a "survey of the field" - I actually misremembered - it is chapter 6 of a book entitled "Thermal Energy Storage Systems" - seems to be one of those tomes intended to substitute for sleeping pills - lots of equations and "this author said this, and the other guy found that..." It seemed to be an effort to create a theoretical model of how the tanks worked, compared to real world results... However they appeared to be using a wide range of studies, using systems applications ranging from chilled water coolers to solar, to boilers. About the only thing shared in common was that they were all using water, or some sort of water based solution, with no phase changes involved. They said that the different systems behaved similarly, with the big difference being that the greater the ΔT, the easier it was to maintain good stratification. The other common factor appeared to be that the tanks were circulating the fluid as in a pressurized system as opposed to the "coil in still tank" method that the SolarTechnics tanks use.

They typically found 3 "zones" in a tank - the hot zone on top, cold zone on the bottom, and a "mixed" zone at the boundary - and that the narrower that mixed zone could be kept, the better the tank would perform. This was both because of the obvious idea that the narrower the mix zone, the wider the other two, and because it seemed to give a more constant temperature in the entire depth of the hot and cold zones, thus the desired zone would stay closer to the target temperature.

I would be skeptical believing that there are major savings with diffusers, etc on a wood fired system.
Btus is btus. The are more factors at play here than one might think of.

The article seemed to say the biggest design controllable factor in the way a tank was constructed that influenced the thickness of the mixed zone was the way the fluids were introduced into the tank - i.e. diffusers. It seemed more important that there was a gentle and distributed flow that discouraged mixing, than the specific design of the diffuser... The other big factor was the heat transfer in either direction through the tank walls, and thus a tank with walls of low thermal conductivity (insulation and plastic) would maintain better stratification than one with highly conductive (metal) walls - obviously this would be a negative for the popular LP tanks...

How is the DHW interfaced with two pressure tanks in series? Are you using a radiant floor or cheap hot water BB?
How frequently are you firing the tank?
They weren't going into this - just what happened inside the tank, and they generally only used one tank... However the suggestion was that it was best to get the biggest ΔT possible - thus you were better off to circulate the water around as few times as possible, so as to get more ΔT out of each pass... They were also just using a single heat change - water was pulled out of the tank, temperature changed, and then returned in most cases, which would be different than the multiple I/O coils that we like to use if we can...

I can tell you that I have significant stratification on a 345 gallon tank. But half of my heating load is DHW. My installation is only 3.75' tall.

Their data seemed to confirm what we have been saying for a while - that tall and skinny is slightly preferable to short and fat, but that there were definite limits before the increased losses due to larger surface areas exceeded the benefits from increasing the stratification. I believe they said ideal seemed to be around 2-1 height vs. width...

An educated guess is that you might pull another 10-15% out of a properly designed stratified system. But I'll bet that
cold in the bottom and hot out the top is all the stratification design you need.
Baffles and diffusers are not going to give the kind of systems we discuss here even a 10% performance bump.
10-15% more out of a system would be a BIG improvement - on the order of one or two additional hours of heating between fires at peak load - significantly worth the effort. How close to that ideal we can get with our setups can be debated, but presumably it's worth pursuing as much as we can get....

Gooserider


If someone is going to prove or disprove this theory for wood boilers, they need to do both parallel and series pressure tanks, with degree day data
and some serious thermocouple installations to sort it out.
 
Status
Not open for further replies.