diagram of pybyr/ Trevor's proposed plumbing

[pybyr]

OK, here’s my try at drawing and posting a diagram of the plumbing of what I would like to achieve with an unpressurized tank, a shell-tube HX as a direct part of the primary, as suggested by Kabbott, and back-to-back reversible flow circulators as pionneered by WoodNotOil and others on here

I am not a draftsman, nor is my penmanship very good, but I hope this comes through well enough to get feedback.

[don't mean to duplicate to a ridiculous degree, but my other post based on discussion of these concepts was this was buried at the end of a thread on corrugated stainless tubing in unpressurized tanks]

 

[Willman]

I see your going with the Taco tees. I would want to know if they can be piped in and not used. Like a Monoflow T. You have two for future use with no piping to Tees. What does taco say about not using them right away ? Will the flow be negatively impacted ? Just wondering.
Will

 

[Willman]

You might have to loop a pipe between outlets to imitate the load.

 

[deerefanatic]

Looks great! Wish I had thought of something that simple.

 

[kabbott]

I like it, pretty simple.
I see no issue with the twin tee's for future use, just plug em or put a pair of valves on em.
My next biggest question is if you reverse flow in one side of the heat exchanger should it be reversed in the other side.(to have counter flow)
And if not which would be better brazed plate or shell and tube.
I would assume you would get better heat transfer if you reverse both(that would complicate things a bit)but it may not make much difference.

Looking at that diagram it could be much more simple than that if not for the to future connections.(are you sure your gonna use them?)

I posted about using 2 pumps and I think Joe talked about dual exchangers which may be a cheaper way to go even with the added plumbing.

 

[antknee2]

Looks great to me , just a couple thoughts . When is your boiler being delivered ??
Anthony

 

[deerefanatic]

Here's something I will inject........ I would put the twin-tee for the future DHW tank BEFORE the air-coil...... Reason being, DHW needs hotter water than the fan-coil does, so put it first in the flow circuit........

Also, yes, you will have a problem with your hx because you will need to reverse flow on both sides for optimum hx...... Now that someone else mentioned it, it makes very good sense..........

If I had to make a compromise in this regard, I would set it up so that you had counter-flow when you discharge the tank. (Right now you have counter-flow while charging the tank) Simple reason is this: You'll be able to draw your storage down further. Chances are, your burner could put out a hotter temp than your tank will charge to (say, 195F) and the lack of counter flow won't really hurt too much if you have a hi-flow circ in your primary loop..... But when it comes time to extract that heat out of the tank, you want to get all you can.

so, My advice is this: Plumb tank so counter-flow happens when discharging. Choose a circ for discharge that can deliver the gpm needed for your worst-case BTU load. (Remember, with a Primary secondary, you get some "free" btu's to work with because you use the same water over and over as it passes around the loop from load to load, extracting more and more energy as you go......) Then, choose a lower flow circ (say a 007, they're cheap and easy to come buy) for the charging circ..... This will give a lower flow rate with more "dwell" time inside the hx... Which should negate the problem of no counter-flow...... If you have a hi flow rate in the primary loop and a low one in the tank loop, you should be ok, even without counter flow.

 

[pybyr]

I see your going with the Taco tees. I would want to know if they can be piped in and not used. Like a Monoflow T. You have two for future use with no piping to Tees. What does taco say about not using them right away ? Will the flow be negatively impacted ? Just wondering.
Will

I checked this pretty thoroughly in talking with Taco; there's no adverse effect on flow of having them there, but dormant

 

[pybyr]

Looks great to me , just a couple thoughts . When is your boiler being delivered ??
Anthony

Hi Anthony- thanks for your suggestions

The boiler is, if all goes according to plan, arriving this coming week

 

[pybyr]

Here's something I will inject........ I would put the twin-tee for the future DHW tank BEFORE the air-coil...... Reason being, DHW needs hotter water than the fan-coil does, so put it first in the flow circuit........

Also, yes, you will have a problem with your hx because you will need to reverse flow on both sides for optimum hx...... Now that someone else mentioned it, it makes very good sense..........

If I had to make a compromise in this regard, I would set it up so that you had counter-flow when you discharge the tank. (Right now you have counter-flow while charging the tank) Simple reason is this: You'll be able to draw your storage down further. Chances are, your burner could put out a hotter temp than your tank will charge to (say, 195F) and the lack of counter flow won't really hurt too much if you have a hi-flow circ in your primary loop..... But when it comes time to extract that heat out of the tank, you want to get all you can.

so, My advice is this: Plumb tank so counter-flow happens when discharging. Choose a circ for discharge that can deliver the gpm needed for your worst-case BTU load. (Remember, with a Primary secondary, you get some "free" btu's to work with because you use the same water over and over as it passes around the loop from load to load, extracting more and more energy as you go......) Then, choose a lower flow circ (say a 007, they're cheap and easy to come buy) for the charging circ..... This will give a lower flow rate with more "dwell" time inside the hx... Which should negate the problem of no counter-flow...... If you have a hi flow rate in the primary loop and a low one in the tank loop, you should be ok, even without counter flow.

Thanks for the feedback; I was hoping that the reversible back-to-back flow on the tank side of things would achieve countercurrent flow for the primary side of the HX in both charge and discharge modes, but I'd been having a hard time taking a fresh look at the overall, etc.

One alternative I could do instead of the shell and tube in the primary loop would be to have the HX tied in as a secondary via a Taco Twin Tee. And since the Twin Tees have takeoff and return right side by side, rather than upstream/downstream, it might let me achieve the countercurrent in both charge and discharge. The drawback of that arrangement is that it'll add an additional circulator, with associated electrical load and control aspects. Maybe that'd be worth it to get best heat exchange. I'll need to give it some more thought and draw up the "Mark II" layout's potential diagram, but first, I'm headed to the woods for a while to cut some more standing dead hornbeam (my new favorite tree!)

 

[BrownianHeatingTech]

Posted in your other thread, and copied here:
------------
Okay, some notes:

Use IFC circulators on the Econoburn side, and delete the check valve in the bypass.

Leave a simple air vent on the top of the Econoburn, but install a microbubble remover in the primary loop, after the tees where the Econoburn connects. Plumb the expansion tank into the bottom fitting on the microbubble remover. Install the primary loop circulator after that (but before the tank heat exchanger). If you do end up going with a flat plate heat exchanger, you want the components in that order (boiler - air/expansion - pump - HX) to prevent low pressure pockets in the heat exchanger, which can cause cavitation.

The twin tees for the indirect should be first, so it gets the hottest water once you install it. No need to be taking cold showers.

I would suggest going with a reverse return piping scheme on the zones. That gives them equal access to the temperature in the loop. If you are sure you will go with in-floor radiant, and the design temp is significantly lower than the return temp from the air coil, then the twin tees will be fine. If you use panel radiators, or the design temp for the radiant is higher than the minimum return temp from the air coil, then the radiators/radiant will suffer in performance. In other words, delete the twin tees (except the one used for the indirect), and install four regular tees (supply, supply, return, return). That way, neither zone's supply is pulling return water from another zone. While you're at it, put in another pair of tees (three supplies, then three returns) - tees are cheap, and who knows if you will need to add another zone in the future, so having a pair of plugged tees would be helpful.
------------

One alternative I could do instead of the shell and tube in the primary loop would be to have the HX tied in as a secondary via a Taco Twin Tee. And since the Twin Tees have takeoff and return right side by side, rather than upstream/downstream, it might let me achieve the countercurrent in both charge and discharge. The drawback of that arrangement is that it'll add an additional circulator, with associated electrical load and control aspects. Maybe that'd be worth it to get best heat exchange.

I think that will end up being your best option. The shell-and-tube or flow-through plate heat exchanger have a "coolness" factor, but they will add significantly to your cost. A twin-tee and pump are no big deal to add. Two extra pumps (or a pump and a valve), and you can to the reversing thing and get better transfer. No extra control work, since you are already activating pumps for drawing from storage.

Joe

 

[pybyr]

I think that will end up being your best option. The shell-and-tube or flow-through plate heat exchanger have a "coolness" factor, but they will add significantly to your cost. A twin-tee and pump are no big deal to add. Two extra pumps (or a pump and a valve), and you can to the reversing thing and get better transfer. No extra control work, since you are already activating pumps for drawing from storage.

Joe

Joe- can you remind me or summarize why you prefer a HX-based system (with a discrete heat exchanger, whether it be plate or shell and tube, outside the tank) to a coil in the tank (its own form of HX but within the tank?

The reason I ask is that, as I puzzle through some of these factors about counterflow, numbers of pumps, etc., for various options- if I instead just have a big coil ( or several parallel coils ) in the tank, with back-to-back circulators set to pump through the in-tank coil in opposite directions, and with a Twin Tee tying that "secondary reversible coil loop" to the primary loop, then the counterflow is easy to do, and the flow directions will also automatically favor maintenance of stratification in the tank. The primary loop will only need one circulator, in one direction. Plus, with coil-in-tank my overall secondary to/from the heat storage tank will have only two pumps, vs 3 or 4.

I don't want to seem either indecisive or argumentative regarding this "which type of HX" decision- it's just that with this forum and this project, every time I learn something, I realize there are still other things to learn and options to consider- that makes it both interesting (to learn and consider so many things) and at times, info overload (trying to consider, or in some cases go back and remember or find) the many valuable bits of info already contributed by so many

 

[BrownianHeatingTech]

Joe- can you remind me or summarize why you prefer a HX-based system (with a discrete heat exchanger, whether it be plate or shell and tube, outside the tank) to a coil in the tank (its own form of HX but within the tank?

Oh, I wouldn't say that I "prefer" it. Each system has its place, and I like both methods, when each is done right. You seemed to be leaning in that direction, so I gave my input on it that way. The stainless immersed heat exchanger is also a good option.

I'd say that the biggest advantage that external heat exchangers have is controllability/serviceability. Because the components are external to the tank, you can adjust them and maintain them without having to open the tank, disconnect a coil, and drag it out of the water.

More pumps and valves and such mean greater flexibility. They also mean more potential failure items, and more electric draw. It's all a balance, and the difficulty always ends up being that there is no "one right" answer." There are several right answers.

Some factors, for example: in a location with low headroom, the physical act of removing an immersed coil from a tank for maintenance might not be possible. The coil is usually almost as tall as the tank, so that means having headroom that is nearly double the tank height. Once a tank is installed, you're not going to want to partially collapse it in order to pull a coil if it springs a leak. If you find that the coil is not sufficient (either through a design flaw, or because of future expansion), and need to add another coil, the same issue can arise.

External heat exchangers avoid that. If a plate heat exchanger starts leaking, or is found to be inadequate, you can just shut off the isolation valves (always install isolation valves), loosen the unions (always install them with unions), and then replace it with another. On the other hand, if you have limited electric power (eg, a strong desire to use less, or a requirement to use less because of an off-grid power system), the extra/larger pumps may be an issue. Immersed coils use fewer pumps and a multi-parallel coil can have much lower head loss than most external heat exchangers, so smaller pumps can be used. You can also install parallel or variable-speed pumps, allowing you to use less pumping electricity when demand is low, especially on the discharge side.

There are other advantages and disadvantages to the two designs, and it really comes down to what will be best for a particular application. You won't "go wrong" by choosing one or the other. It's just a matter of weighing which you think will be the best for your needs. All heat exchangers involve compromise (and going to a pressurized system involves compromise in terms of needing a pressure-rated tank and a large expansion vessel). It's just a matter of choosing which of those compromises will be best for you.

Joe

 

[WoodNotOil]

Joe

Very well put! People can start second guessing themselves in reading posts here and think they have chosen the wrong method. I think the choice has a lot to do with the space the storage is going in and what skills and materials each person has available to them. The encouraging thing is that so many people on this forum are coming up with different methods that work!

 

[BrownianHeatingTech]

Very well put! People can start second guessing themselves in reading posts here and think they have chosen the wrong method. I think the choice has a lot to do with the space the storage is going in and what skills and materials each person has available to them. The encouraging thing is that so many people on this forum are coming up with different methods that work!

Not to get too far afield, but there are basically four ways to do storage:

1. Pressurized - pumped
2. Pressurized - gravity (we'll include the Garn system in here, since it's an odd duck)
3. Atmospheric - immersed heat exchanger
4. Atmospheric - external heat exchanger

Each has a place, and each has benefits and drawbacks.

Joe

 

[deerefanatic]

Yep, the in-tank method is the best for "off-grid" or "maximum energy efficiency" follks... BUT, the cost of the soft copper you'll need if you're building a hi-BTU system will get pricey in a hurry..........

I like having no HX at all..... To me, a little anti-corrosion chemical (if you have an open system with no hx like me) or big expansion tank (with closed systems) is a small price to pay for heating your storage water directly....

 

[pybyr]

Here is the "Mark II" design of my piping layout, based on a lot of the feedback above.

Again, please DO "fire away" with comments, suggestions, etc; in particular, I invite comments and suggestions on "how much of a difference" this MK II better layout with counterflow maintained in all conditions is likely to work compared to a possible layout like the original, above, at the start of this thread, which would use a "honkin' big" industrial surplus shell and tube heat exchanger directly in the primary loop.

I have not completely ruled out a shell and tube HX in the primary loop, as I may have bumbled onto an economical source of some healthy -sized un-used industrial ones,

__BUT__ it does seem clear that a pump-fed flat plate HX on a secondary, as shown in this second drawing offers an inherent advantage of maintaining [I think] automatic counterflow temperature gradients across the HX in both charge and discharge mode of the hot water tank...

I am still leaning towards the Taco Twin Tees for the places where primary and secondaries meet, on account of the ability to shoehorn a lot of activity on a primary loop into some relatively compact but uncluttered space, via a bunch of 45 degree elbows.

As always, thanks to everyone for all the knowledge and experience and willingness to share it.

 

[pybyr]

unhhh- ugh- as I am looking at my diagram, it's now occurring to me that perhaps, in order to maintain counterflow in the plate HX in all modes, I may need an extra circulator- for back-to-back reverse circulators- over there, too (in the plate HX's secondary's connection to the primary loop).

or am I just overthinking due to overtiredness?

I do "get" the physics behind counterflow and its inherent efficiency at transferring anything like heat or chemical concentration gradients across any exchange medium, but I am wondering about people's opinions of how much effort and expense, in both construction (extra circulators, flanges, etc) and operation (power to run circulators) it is worth to chase this holy grail of countercurrent.

Is achieving countercurrent flow in the HX between the primary loop and the storage tank in __both__ tank-charge (boiler heating tank) and tank-discharge (tank heating house) likely to gain me big noticeable improvements in higher house comfort, and lower wood consumption? I definitely want to "do this right the first time" but also don't want to get hung up in chasing concept-based ideals that may or may not make a huge difference in results as actually experienced.

And can someone confirm that the innards of plate HX's flow fluid and transfer heat equally well regardless of direction of fluid flow on each/ either side (assuming same countercurrent flow is maintained and both sides are reversed)?

I also realize I left out any depiction of controls like an automag in the dump zone- that's only because I am focusing on the main system at the moment- I understand that as depicted, with no controls, the open loop shown for the dump zone would mess up all the flows in and around the boiler- so again, ignore the dump zone for now.

thanks

 

[kabbott]

I am not convinced that counter flow is all that important but I do not no this from experience, that is why I brought it up
in the other post.would be nice to hear from someone who has done/tried it.

If it is important(counter flow) than rather then all those pumps I would lean toward using a pair of three way valves or four EBV's as described
in another thread to switch between upper and lower tank taps.

There seems to be no "easy" way. I am seriously considering dropping 1000 feet of pex in the tank for this winter and see how it works.
I am planing on a bunch of radiant anyway so if its not up snuff I have a use for the pex.

 

[BrownianHeatingTech]

unhhh- ugh- as I am looking at my diagram, it's now occurring to me that perhaps, in order to maintain counterflow in the plate HX in all modes, I may need an extra circulator- for back-to-back reverse circulators- over there, too (in the plate HX's secondary's connection to the primary loop).

or am I just overthinking due to overtiredness?

I do "get" the physics behind counterflow and its inherent efficiency at transferring anything like heat or chemical concentration gradients across any exchange medium, but I am wondering about people's opinions of how much effort and expense, in both construction (extra circulators, flanges, etc) and operation (power to run circulators) it is worth to chase this holy grail of countercurrent.

Is achieving countercurrent flow in the HX between the primary loop and the storage tank in __both__ tank-charge (boiler heating tank) and tank-discharge (tank heating house) likely to gain me big noticeable improvements in higher house comfort, and lower wood consumption? I definitely want to "do this right the first time" but also don't want to get hung up in chasing concept-based ideals that may or may not make a huge difference in results as actually experienced.

Being able to have counterflow in discharge is the important thing. So it's set up fine as is.

Counterflow will always improve heat transfer, so it's a "good idea" in any system.

Adding a second circulator would not be a major hassle - just one pumping away from the twin-tee on each side.

On that note, the tank circulators should be pumping away from the tank, rather than next to each other on the bottom. In other words, each tapping should have a circulator, and that circulator should point away from the tank. Since the tank is atmospheric, maintaining positive pump pressure in the piping is important to avoid potential boiling.

Joe

 

[jebatty]

I've wondered how well this might work to maintain counter-flow at all times in the hx with a single circulator.

1, 2, 3, and 4 are valves. When the tank is charging, 1 and 2 are open, 3 and 4 are closed. Hot water to top of tank, cold water drawn from bottom.

When the tank is supplying, 1 and 2 are closed, 3 and 4 are open. Hot water drawn from top of tank, cold water to bottom.

 

[jebatty]

For info on parallel and counterflow, this might be helpful:
http://www.engineersedge.com/heat_transfer/parallel_counter_flow_designs.htm

 

[Nofossil]

I've wondered how well this might work to maintain counter-flow at all times in the hx with a single circulator.

1, 2, 3, and 4 are valves. When the tank is charging, 1 and 2 are open, 3 and 4 are closed. Hot water to top of tank, cold water drawn from bottom.

When the tank is supplying, 1 and 2 are closed, 3 and 4 are open. Hot water drawn from top of tank, cold water to bottom.

I contemplated such a scheme, but valves are about the same cost as pumps, and typically impose a big flow restriction.

 

[jebatty]

Didn't check prices or reliability, but Taco and others make a ball zone valve that might be considered, for example:
(broken link removed to http://www.taco-hvac.com/uploads/FileLibrary/EBV_100-7.7.pdf)

 

[BrownianHeatingTech]

Didn't check prices or reliability, but Taco and others make a ball zone valve that might be considered, for example:
(broken link removed to http://www.taco-hvac.com/uploads/FileLibrary/EBV_100-7.7.pdf)

The EBV's are nice, but primarily for the design of the electrical actuator (which draws extremely little power). The internal ball valve is smaller than the pipe diameter, and they don't have much better flow characteristics than other zone valves.

For flow-reversing with a valve, I'd look for a simple 4-way valve. Single valve body and a single actuator, and you have flow reversal with a single pump.

It will be more expensive than a pump, so it's of dubious value. Personally, I'm liking the idea of using pumps, and adding a switch to make either pump the "only" pump. That way, if one fails, you flip the switch and go to single-pump operation. You'll lose the stratification effect on one half of the cycle (either charge or discharge, depending upon which pump failed), but can take a few days to obtain a replacement pump and actually install it, instead of having a non-functional heating system at 9PM on a Friday night.

Joe