Simplest Pressurized Storage revisited

[Nofossil]

Way back in 2008(!) I sketched up a system plumbing schematic that became a sticky. It's been a while, and I've worked out a few other designs and variations since then. I'd like to take my current thinking and pass it by the collective wisdom of the boiler room wizards. Does this approach make sense, and does it accomplish the objectives listed below? Have I missed anything important? Looking for any thoughts, ideas, and suggestions.

Thanks....



Notes:

• All circulators have integral check valves.
• In this version, the bypass circ is a Taco Viridian or other temperature controlled variable speed circ, set to maintain 140 degrees at the wood boiler inlet whenever the wood boiler is operating. A Danfos or equivalent mixing valve could be used instead.
• The Load Circ and Oil Circ are Grundfos Aplhas or other constant pressure circulators.
• The wood boiler circ could be a Viridian or other variable speed circ set to achieve the desired target temperature at the wood boiler outlet. It could also be a fixed-speed circ controlled by the boiler controller.
• The oil boiler is disabled when the top of storage is hot enough to be useful.
• The load circ is disabled when the storage top is not hot enough to be useful.

Key objectives and ideas:

• The wood boiler / storage loop on the left side is completely independent from the loads on the right side.
• There's no opportunity for 'ghost flow'.
• The oil boiler will never charge storage.
• Each side (wood boiler and loads) can use different flow rates as needed to optimize performance.
  
• Stratification in storage is easily maintained.
• Controls can be very simple - native oil boiler and wood boiler controls plus a thermal switch on storage to disable the oil backup and enable the load circulator.
• Wood and oil are independent. Either can be valved off and worked on without losing heat.

 

[Kristen]

My DHW tank is heated directly off the boiler loop, I don't think my storage water would reliable be hot enough, "all the time", to do that job. Curious to hear how yours works / if you have to rely on some top-up heat etc. to always have hot DHW?

However, the cold feed to my DHW tank is pre-heated using a coil (**) in the storage tank, so when hot water is drawn off the DHW tank does not cool much/at all - indeed, when the storage tank is "properly hot" the replacement feed arrives already hot

(**) actually, there are two coils in the storage tank, low and high, to reduce de-stratification

 

[Nofossil]

My DHW tank is heated directly off the boiler loop, I don't think my storage water would reliable be hot enough, "all the time", to do that job. Curious to hear how yours works / if you have to rely on some top-up heat etc. to always have hot DHW?

Sounds like you're asking about my personal setup rather than the reference design in this thread. My DHW performance sounds like yours. My DHW tank also has much higher standby losses than my storage tank, so it needs to be topped off from time to time.

 

[maple1]

Looks pretty good to me, and actually pretty similar to mine. The main difference is I only have one tapping at the top & bottom of storage, so I T there rather than having two ins & outs.

Couple of comments though.

Are 3 expansion tanks necessary? Would likely be a situation specific thing, maybe, but thinking the one big one tied to storage should suffice? I have 3 tanks on my system, but that's a cabbaged up affair that I ended up with in a 'design-build' way. Started with just leaving the existing 20 gallon bladderless one in place in the joists, added a bigger one on the floor beside storage, than added another 15 gallon bladderless up high under the top of my stairs to second floor just this fall to give me a personally more comfortable range of pressures cold-hot. But if my big one on the floor was bigger, I think it would be enough. Also my circs are all pretty low in my system, on returns. Except my electric boiler circ, it's on the outlet of that boiler.

Then something a bit sideways. I don't have checked circs, I have separate check valves. But something else I did when I was into my system this summer, was take the flapper out of the one that prevents sending backup heat to my storage, drill a 1/4" hole in it, and put it back in. That allows a little bit of flow through storage when the backup electric boiler is being used - but it took a lot of the short-cycling out of the operation of that boiler (main concern). It works much smoother now. And also slows the drop in storage temps over a day or two - thereby also stabilizing my system pressures a bit better for the period I might run on backup. (Usually only a couple days a winter). That would likely not be a conderation for many. I think my backup boiler is bigger than I need (18kw), I would use something in the 12kw range if doing again - but the price was right.

 

[jebatty]

I greatly favor your new design.

Independence of the boiler/storage loop from the storage/load loop is an important improvement, but it does come with a cost which in a particular situation may outweigh the benefit. Regardless, a design which provides independence of these two loops currently is my only recommendation because it better assures avoidance of design issues which can arise with attempts to follow the original sticky design.

The cost would relate to a need for very high temp water to serve the load, but this also is a situation where usefulness of storage is limited. The need for very hot water can best be met by direct plumbing to the boiler, but keeping a wood boiler fired to meet this need is problematical.

I'm not sure why you show two expansion tanks, because pressure will equalize throughout the system. On the other hand, I don't see any disadvantage to having two expansion tanks.

When I first installed my Tarm in 2006 I carefully followed a design schematic, but did so with nearly complete ignorance of the need to plumb appropriately for the btus that needed to be moved, distances, pump head, flow rates, sizing of pipe and sizing of circulators. These remain the major problem areas for wood boiler owners which frequently are discussed in this forum.

 

[ewdudley]

The wood boiler circ could be a Viridian or other variable speed circ set to achieve the desired target temperature at the wood boiler outlet.

The Viridian VT pumps do not have a mode that would support this mode of operation. I don't know of any setpointT ECM pump that does.

t could also be a fixed-speed circ controlled by the boiler controller.

If boiler controller runs Wood Circ with "launch temperature" or similar on-off control then Bypass Circ Viridian would be hard pressed to maintain proper return temperature with the sudden changes in Wood Circ flow.

The oil boiler will never charge storage.

System designers may prefer to charge a limited upper portion of storage with the fossil fuel boiler to provide a buffer that can militate against short-cycling. Such a design would have the fossil fuel boiler draw its return water from somewhere near the top of storage and then would supply hot water directly to top of storage. This also keeps the load circuit design independent of the heat source(s).

Stratification in storage is easily maintained.

How so?

 

[ewdudley]

The main difference is I only have one tapping at the top & bottom of storage, so I T there rather than having two ins & outs.

Right. The two ports top and bottom serve no purpose.

 

[Nofossil]

Right. The two ports top and bottom serve no purpose.

They serve to pretty much absolutely prevent ghost flow through the wood boiler when the load circ is running. I've had some installations where the near-storage plumbing wasn't ideal and created enough of a pressure drop to induce ghost flow.
I agree that two ports should not be necessary, but they do provide insurance.

 

[ewdudley]

but they do provide insurance

No insurance is necessary if you draw the circuit and do the math.

 

[Nofossil]

The Viridian VT pumps do not have a mode that would support this mode of operation. I don't know of any setpointT ECM pump that does.

Unless I'm missing something, the Viridian 'Setpoint Heat' mode with 'Increase speed on temperature rise' setting would do exactly that.

If boiler controller runs Wood Circ with "launch temperature" or similar on-off control then Bypass Circ Viridian would be hard pressed to maintain proper return temperature with the sudden changes in Wood Circ flow.

True, but they do respond pretty quickly. In this situation a Danfos or similar return protection valve is probably better. I like the bypass circ because that lets me pre-warm the boiler from storage on a cold start.

System designers may prefer to charge a limited upper portion of storage with the fossil fuel boiler to provide a buffer that can militate against short-cycling. Such a design would have the fossil fuel boiler draw its return water from somewhere near the top of storage and then would supply hot water directly to top of storage. This also keeps the load circuit design independent of the heat source(s).

That's a pretty logical enhancement, and there are a few others. This design is intended provide a very simple baseline approach to the problem of adding a wood boiler and storage to an existing system.

How so?

Charging flow is top-to-bottom and discharge is bottom-to-top. Flow rates in and out of storage can be pretty easily minimized based on heat output and heat loads. No 4-way valves, opposing circulators, or other complexities. Flow through the heat loads can provide a much higher delta T to storage than most primary/secondary approaches as the cool/return water is separate from the hot/supply water.

 

[maple1]

I will add that I did have some ghost flow through my boiler at times with my 15-58 load pump - when enough zones were calling at once.

The ghost flow seemed to stop when I replaced the 15-58 with the Alpha, and reduced my zone flows.

I didn't do any math though, and the circuits were kind of loosely drawn but only in my head.

 

[Nofossil]

Way too many plumbers (and even system designers) don't get the importance of storage stratification or the potentials for ghost flow. I've seen really elegant designs where the installation resulted in real problems. Bottom line - if the installer doesn't understand the system, they can easily make mistakes that are expensive to correct. In single port storage connections, the orientation of the tees is important (see the sticky, for instance). About half the installations that I see have it wrong, regardless of how it's drawn. If I were doing it myself, I would certainly do the single port approach.

 

[ewdudley]

Unless I'm missing something, the Viridian 'Setpoint Heat' mode with 'Increase speed on temperature rise' setting would do exactly that.

Yes, you are missing something. Setpoint heat increases flow when sensed temperature falls.

 

[Nofossil]

Yes, you are missing something. Setpoint heat increases flow when sensed temperature falls.

You're right :-(
I was thinking 'Setpoint cool', but that only allows a maximum of 100. Have to be on/off or a Vesta controlled circulator.

 

[ewdudley]

You're right :-(
I was thinking 'Setpoint cool', but that only allows a maximum of 100. Have to be on/off or a Vesta controlled circulator.

I'm not right, Taco is wrong. To paraphrase Westly, "Setpoint Cool of Unusual Temperature? ...I don't think they exist.".

 

[DaveBP]

On the issue of multiple expansion tanks...

If you want to valve off one side of the system to improve or repair it during heating season, the side doing the heating (while you work on the other) will appreciate its own expansion tank.

Over on the heating help forum the subject of delta T vs. delta P circulators can resemble the OK corral sometimes.

 

[Nofossil]

EW - I was really surprised that Taco limited the temp range like that. It obviously limits potential applications. If I've learned anything about heating systems, it's that people will come up with uses for equipment that you would never expect.

Dave - that's exactly the reason for the multiple tanks. The little ones are really cheap, also. For purists, it also allows each circulator to be pumping away from an expansion tank. I don't think that's much of an issue in small residential systems, but it can't hurt.

 

[kuribo]

Regarding the taco viridian set point cool mode (reverse acting mode), I plan to use this for charging storage in a "one and done" set up. All you need to do is put the appropriate resistor in series with the temp sensor to "trick" it into working in the range you need. There is a chart in the taco literature that gives you the resistance vs temp which makes this relatively trivial to do...I have talked with taco about this and they foresaw no issue....

 

[Nofossil]

Nice hack - I like that.

 

[Beagle Dad]

They serve to pretty much absolutely prevent ghost flow through the wood boiler when the load circ is running. I've had some installations where the near-storage plumbing wasn't ideal and created enough of a pressure drop to induce ghost flow.
I agree that two ports should not be necessary, but they do provide insurance.

Separate ports are the best way to go. The tank(s) act as a low loss header so that the boiler circulator flow does not influence by the flow to the various load circulators and the other way around. "Ghost" flow can occur just by height differences between boiler and storage, and storage and loads. If motorized valves (to expensive to justify in most cases) are not used to isolate zones which are not in use then small flows can be established.

Tanks in series and separate connections for the boiler and the load is the best way to go. To further improve stratification connection returns from the loads at various points help stratification. I have four tanks in my system which allows the bottom of the tank system to hold 90 to 100°F most of the time and the top of the system to be kept at 160 to 170°F

 

[ewdudley]

Separate ports are the best way to go. The tank(s) act as a low loss header so that the boiler circulator flow does not influence by the flow to the various load circulators and the other way around. "Ghost" flow can occur just by height differences between boiler and storage, and storage and loads. If motorized valves (to expensive to justify in most cases) are not used to isolate zones which are not in use then small flows can be established.

Tanks in series and separate connections for the boiler and the load is the best way to go. To further improve stratification connection returns from the loads at various points help stratification. I have four tanks in my system which allows the bottom of the tank system to hold 90 to 100°F most of the time and the top of the system to be kept at 160 to 170°F

Again, this is nonsense. Study the schematic.

If the boiler circuit and the load circuit meet at a tee that goes to a short leg leading to a single port on both the top and bottom of storage, you're good to go.

As the Wood Circ charges storage there can be no unwanted flow against the oil boiler check valve, nor through any of the closed zone valves.

As the Load Circ draws from storage there can no unwanted flow through the Wood Circ until and unless the pressure developed by the Load Circ on that parallel path overcomes the weight of any cool water at the bottom of the parallel path.

 

[jebatty]

Way too many plumbers (and even system designers) don't get the importance of storage stratification or the potentials for ghost flow.

Isn't this somewhat system dependent? My shop system, plumbed like the new simple design, which would be typical in similar situations, uses mixed hot water supply at 100F to feed the concrete in-floor set at constant 61F, and I typically run storage down to about 110-120F before firing the boiler. Stratification is relatively meaningless, and with in-floor supply at 2.25 gpm vs boiler supply at 14gpm, mixing of the storage during boiler burns does not materially upset low temp return water from bottom of storage to the boiler. Any ghost flow, and there is some in my system, does no more than add a little extra heat into the heated space, since the boiler and storage are in the heated space. There is no wasted heat from ghost flow.

 

[Beagle Dad]

Again, this is nonsense. Study the schematic.

If the boiler circuit and the load circuit meet at a tee that goes to a short leg leading to a single port on both the top and bottom of storage, you're good to go.

As the Wood Circ charges storage there can be no unwanted flow against the oil boiler check valve, nor through any of the closed zone valves.

As the Load Circ draws from storage there can no unwanted flow through the Wood Circ until and unless the pressure developed by the Load Circ on that parallel path overcomes the weight of any cool water at the bottom of the parallel path.

The drawing at the beginning of the thread works just fine with one connection because of the zone valves isolating the loads which are not in use and hopefully people have a plan for speed control on the pump from a flow control and for electrical power consumption savings. My point was that not everyone has systems with zone valves everywhere and this is when they have problems. Many people who ask for help on the Forum have systems piped with undersized pipe at key point in the system especially around the storage tanks.

 

[jebatty]

Again, [separate ports] this is nonsense. Study the schematic.

If the boiler circuit and the load circuit meet at a tee that goes to a short leg leading to a single port on both the top and bottom of storage, you're good to go.

I'm having difficulty agreeing with the T approach and single ports. When both the wood boiler circ and one or more load circ are "on," the circs work in series, and there is a probability that hot water supply will bypass storage supply partially to completely and feed the loads directly, and based on return water temperature from the loads, return water may be "hot" and bypass storage return partially to completely, resulting in 1) returning "hot" water to the boiler and cause it to cycle or idle (even though water in storage is cool) and/or 2) failure to charge storage.

Maybe use of the control systems or plumbing suggested will mitigate or eliminate this result, but this is exactly the result which occurred in a system with which I am very familiar, and adding a new control system or radical plumbing changes was not economical or feasible. The fix was a fairly simple shift to a two port design: wood boiler direct supply to storage at top of horizontal tank at one end and direct return to boiler from bottom of tank, and similarly with loads, direct supply from storage at top of horizontal tank at opposite end and direct return to storage at bottom of tank. Storage is a 1600 gal horizontal tank. In this way the boiler only heated storage, hot water to top of storage and cool water from bottom of storage. Boiler cycling/idling was eliminated and storage charging was fully facilitated. And storage only served the loads.

 

[Beagle Dad]

Isn't this somewhat system dependent? My shop system, plumbed like the new simple design, which would be typical in similar situations, uses mixed hot water supply at 100F to feed the concrete in-floor set at constant 61F, and I typically run storage down to about 110-120F before firing the boiler. Stratification is relatively meaningless, and with in-floor supply at 2.25 gpm vs boiler supply at 14gpm, mixing of the storage during boiler burns does not materially upset low temp return water from bottom of storage to the boiler. Any ghost flow, and there is some in my system, does no more than add a little extra heat into the heated space, since the boiler and storage are in the heated space. There is no wasted heat from ghost flow.

True, if the heat ends up where it is needed then there is no problem. In some installations the area where the tanks are located (basement) can be warmer than the rest of the house