Tank piping question

[Eric Johnson]

I've decided to use a flat plate heat exchanger to get heat in and out of a non-pressurized 1,000-gallon hot water storage tank. But I'm not sure how to pipe it. It looks like a trade-off any way you do it, since it's either going to be more efficient at storing heat, or more efficient at recovering it, depending on how it's piped.

My original thought was that it's much better to have the efficiency skewed towards the recovery side, since charging up the tank can more or less happen in the background, as the opportunity arises, while fast heat recovery is the purpose of the whole exercise.

However, slowzuki's point about diminishing efficiency as the tank heats up has got me thinking. Suddenly, drawing water from the top of the tank when you're trying to extract heat from the wood boiler seems like a bad idea.

How bad? I don't know. Any thoughts would be appreciated.

Here's a link to a pdf of the piping scheme:

http://www.nefpexpo.net/DHWrepipe.pdf

Here's a jpeg of the same image:

 

[Andre B.]

Are you going for stratified or mixed storage?

 

[Eric Johnson]

Educate me on that, Andre.

If that means what I think it means, then I'm thinking mixed would make more sense.

 

[Andre B.]

Stratified. Starting with a hot tank you take hot water off the top and return the cold water to the bottom. You need to design the outlet and inlets of the tank piping to minimize turbulence so that the cold water going into the bottom does not mix with the hot water in the tank.
This way you get more even hot temp until all the hot water in the tank is used up and you will get more usable energy storage.

Mixed is just what is says disadvantage is the hot water coming out gradually gets colder as you take energy out of the tank.

I remember a big debate on which was better back when solar was first being pushed.

 

[Eric Johnson]

With the flow created by the circulator, do I have any choice but to mix it?

 

[slowzuki]

Eric, it will mix given your size of tank. If it was closed loop it might not mix so much.

 

[Eric Johnson]

So in a mixed situation, what's the best way to pipe it?

 

[slowzuki]

The temp will be similar throughout the tank, the short answer is it isn't going to matter a lot in your situation. If you were closed there is a bit more work to do or if you were going to design some nice laminar flow suction and discharge pipes. But straight pipes sucking and discharging is gonna mix that small of tank pretty good. I would be inclined to suck from the bottom and discharge on top if pressed for an answer only because I think you will have more problems charging it than discharging.

 

[Eric Johnson]

By odd coincidence, Ken (slowzuki) happened to be in my neighborhood--a 12-hour drive from his house--picking up a really cool looking wood gasification boiler from a guy who had it disassembled and stashed away in a storage shed. Ken emailed me, wondering if I'd be willing to lend a hand loading it onto his truck. Moving thousands of pounds of refractory, boiler plate and control panel on a rainy evening by hand is right up my alley, so of course I agreed.

Ken was nice enough to stop by my place while he was here, and I gave him a brief tour of my setup. I was especially interested in getting his ideas on how to proceed with the tank, since that's the only element I'm really not sure is going to work. It may not be the heart of the system, but I'd say it's one of the lungs. We discussed its structural integrity, which may be dubious considering my meager masonry skills. My plan is to rest a couple of floor joists on top of the wall so that the weight of the house adds support against the weight of the water. Ken suggested getting some of that fiber-based cement that you slather over cinderblock walls without having to mortar the joints. I'm thinking if it works in that application, it can do nothing but strengthen a conventional block and mortar wall. I've included a pic of where I'm at with it.

You can see the insulation and the edge of the liner above the top course of blocks. The white stud is just there to hold the liner in place. I'm going to put a layer of mortar on the top edge you see here (solid 4" blocks) and then bolt a 2-inch plank onto it. I think that's the best way to ensure even distribution of the weight I'll be putting on the wall. The top of the wall isn't entirely flat, in other words, and the layer of mortar sandwiched between the plank and the blocks ought to compensate for that.

Ironically, if the tank works, I have another 1,500 gallons of capacity right next to it in what's left of the original cistern. If that tank (part of which comprises the foundation) is full of water, the structural integrity of the smaller tank wall is a moot point. But I don't want to invest that much time, money and effort until I am confident that it will work. So the wall has to hold, which is probably just as well.

On the topic of getting heat in and out of the tank with a flat plate heat exchanger, the guy Ken bought the boiler from, who seemed to know what he was talking about, said that it would work "if it's engineered right." I didn't get a chance to get a more detailed explanation. Maybe Ken got more out of that statement than I did.

So I'm soliciting engineering ideas in addition to the ones already covered above. Along with the flat plate heat exchanger, we're talking about two dip tubes and one or two circulators, depending on how anal I want to get on efficiency, and assuming that it doesn't mix, which everybody seems to think it will. A mixed tank simplifies things, but if there's a way to get some stratification, I'd like to try it. My main question is, what would be the best location and length of the diptubes in the tank? For the sake of economy and compatibility with the foam and rubber tank lid, I'm planning to use one-inch CPVC for these tubes.

 

[Nofossil]

I'm a big fan of stratified heat storage - the more stratified, the better. Major advantages: The hotter the water at the top of your tank, the more useful it is. If my tank top is 150 and my tank bottom is 100, I've got lots of useful heat. If my whole tank is 125, it's nowhere near as useful.

Next advantage: The cooler the water at the bottom of your tank, the more heat you can extract from any available heat source.

My storage tank is unpressurized, and has three heat exchanger coils in it:

1) Bidirectional coil with lots of extra copper top and bottom connected to boiler and baseboard zones. Flow is top-to-bottom when heating from boiler, bottom-to-top when drawing heat from tank to heat house.

2) Coil with 2-port inlet manifold for solar panel thermosiphon. Flow is always top-to-bottom. Manifold allows flow to heat bottom half of tank if solar panel outlet is not quite as hot as the tank top. The colder the bottom of the tank, the more solar gain I can get. The hotter the water at the top, the less chance that I'll need to burn any oil to get my DHW hot enough.

3) Domestic hot water preheat coil. Flow is always bottom-to-top. Water enters bottom at 55 degrees.

I can easily get 40 or 50 degrees of stratification. There's a picture on this web page - scroll down.

Comment: If I had to do this over, I'd include an extra tank - maybe 120 gallons or so - that I'd heat first and hottest so that I'd have a better chance of having really hot water when I need it. Ticks me off to have a good solar heat day - heat my tank by 80,000 btu - and still not have a high enough temperature at the tank top to avoid burning oil for DHW.

Hope this helps.

 

[Eric Johnson]

I may have to resort to copper coil or some other pressurized heat exchanger. I'll have to see how the pumped flow through the flat plate works. I've bounced the idea off of a few people who should know and they all said it should work, though obviously with some compromises that I may or may not be willing to live with.

I don't want to get too personal, but how long is your primary coil, and what is the size? It sounds like yours flows both ways. Do you do that with two pumps or some fancy bypass or mixing valve?

Somebody suggested that I get two coils--one for adding heat and the other one for taking it out. I don't know--that stuff's pretty expensive.

And finally, does a bigger tank demand a bigger heat exchanger? I imagine it couldn't hurt.

I thought about using pex-al-pex but as somebody pointed out, it doesn't transfer heat very well. But I know they use it for under-the-floor radiant and driveways, so maybe it would work. Or how about a bank of finned copper baseboard piped in series--one on the top of the tank and another on the bottom? I also considered a few cast iron radiators. With a good coat of paint, they might not rust out.

I can see why you would put a solar coil into a hot water storage tank, but I have a sidearm heat exchanger that keeps my 50-gallon electric water heater full of hot water. Less efficient, no doubt, but fast and effective.

 

[Nofossil]

Wow - lots of questions. I'll do my best, and rest in confidence that someone else will jump in if I'm wrong....

First - primary coil (the one that I use for heating the tank from the boiler, and extracting heat from the tank for space heating). It's 50' of 3/4 copper wrapped in a 40" diameter spiral about 45" high. I also have a rectangular array of 1/2" copper pipes 60" long on the top and the bottom. If you look at the picture in the embedded link on my previous post, you can see it. If I remember, I'll attach the picture itself to this post.

I do flow both ways. In my simple mind, I have three possible heat sources: The oil furnace, the wood boiler, and the storage tank. Each has its own circulator pump. The tank is funky because it also has a zone valve that allows the wood boiler or the oile furnace to treat it as another zone and circulate hot water through it. I think I already posted a plumbing diagram - let me know if you can't find it.

A bigger tank does not require a bigger heat exchanger. The size of the heat exchanger is determined by the amount of heat that you need to transfer per unit of time. I need to transfer a maximum of 80,000 btu/hr regardless of my tank size. A bigger tank will take longer, and a smaller tank will be quicker, but the size of the heat exchanger doesn't change.

I think the cast-iron radiators are a great idea. Immersed in water rather than air, I think they would have mind-blowing heat transfer rates.

I only use the solar (with oil backup) for DHW in the summer when the wood boiler is not running. I've gone as much as 30 days without using any oil at all, reaching a tank top temperature of over 150 degrees this past August.

I assume that this site gives you the ability to email me. If you're not too far away, a visit might be in order. Tours are available at the very reasonable price of one beer.

 

[Eric Johnson]

Very helpful. Thank you again.

Assuming the relationship is linear, I'm guessing I'd need something like 200 feet of copper coil to stash 205K btu/hour into my tank. Does that sound about right?

Actually, I only have a 1" line from the wood boiler to the tank. My other 3/4-inch line loops through the greenhouse. So it's probably closer to 150,000 K going into the tank, at least the way it's currently configured.

I go into and through Vermont on a regular basis on business, so I'll look you up one of these trips through the People's Republic. In fact, the trade association I work for puts on a logging/sawmilling equipment trade show every second year in Essex Jct. This year it's May 9-10.

Our fellow Hearth.com member keyman512 thinks the cast iron rad idea is a great one. I like working with ci rads, but getting a couple into my tank and mounting them in the right position would be a daunting challenge. Any opinion on the finned copper baseboard elements soldered into some sort of an upper/lower grid? Maybe 50 feet in each one. I don't know if you can buy the elements without the steel covers. I've never seen them for sale like that, but assume they're available somewhere.

 

[Nofossil]

I am terrified of the unknown, and in my case that would include galvanic corrosion. I think aluminum in a water bath with steel or iron anywhere nearby is really bad, but I will gladly defer to someone who knows what they're talking about.

 

[Eric Johnson]

I'm starting to get cold feet on the flat plate concept. The tank isn't quite finished and I still have room to get a pressurized heat exchanger in there. I can use the 150K flat plate for the greenhouse when I switch it over to glycol. I don't think my old 1.25-inch sidearm is really up to the task.

First I thought about using a 1" pex-al-pex coil in the tank, but I think that's obviously a loser when you compare its heat transfer efficiency to copper. But then it occurred to me that I have about 200 feet of 1" copper currently supplying and returning water from the EKO to the gas boiler in my basement. So with a little work and a few fittings, I think I could replace the copper with pex-al-pex and build a heat exchanger pretty much guaranteed to work and stratify the tank. All for probably less than $500.

So I'm thinking about it. Building the thing would be fun, of course. Running the pex somewhat less so, although since it runs through unheated space, I'd be more comfortable with plastic and aluminum than copper anyway in case things ever froze up. The pex has the added advantage of fewer leak-prone joints and, as we've discussed, is to some extent self insulating. Plus it allegedly handles freezing better.

So, nofossil, can you point me in the direction of some hx designs for non-pressurized tanks? Yours looks really nice, but I get confused by the solor and DHW components. All I really want to do is dump heat from the EKO into the tank and recover it as needed, using a couple of fairly compact heat exchangers made out of rigid 1" copper and fittings.

Your enthusiasm for the cast iron radiator idea has me thinking that if I put in the second tank, maybe I'll try that strategy there. Other than the corrosion potential, I don't see how you could lose doing that--especially with the right radiators, and there's about a million designs to choose from. Around here, old steam rads are pretty cheap. That's because they can't be used with hot water--no top pipe and thus no venting the columns. But turn them upside down or mount them horizontal in a tank, and they vent just fine out of the supply and return tappings.

Too bad they didn't make them out of stainless steel. I think in a sealed tank in the basement, the oxygen level in the water would eventually get pretty low. I know I've pumped tap water through cast iron circulators for years and never had a problem. Of course, we always have to try to look out for the next generation.

 

[elkimmeg]

What about daisy chaining truck radiators in the center of that tank and let them be your exchangers you would get plenty of surface area.

In my days of remodeling you would not believe the stuff getting thrown away including radiator cores and even cast iron radiators

 

[Eric Johnson]

I've been through all that, elk, and while there are plenty of good potential hx designs, I'm back to the basics, which is essentially a couple of copper loops in the tank. This approach has rather dramatically changed the design on parts of my system, but I've finally concluded that simple is best and in this case, most efficient and responsive.

I hope somebody who knows how to make the calculations can tell me specifically:

How much 1" copper for each coil (top and bottom)
How it should be shaped
Where it should be located in the tank.

The tank is roughly 4x6x7 and I need to dump a max of 150,000 btu/hour into it. My water temps going in will be 180-190. Most of my zones have cast iron radiators, so there will be lots of flow through the heat recovery coil and the temps should be relatively low. The supply coil will be supplied by a 1" line piped directly (100 feet each way) from the wood boiler with a big pump (Grundfos 26-96).

 

[slowzuki]

Eric I've just got caught up with the posts and I'll offer up my opinions.

-The wall should reinforced, the total force the water will exert on the wall is about 9000 lb if you fill it up to the top. The bottom is under more load than the top, the first foot of block is holding back 2500 lb roughly.
-The engineer in me doesn't like multiple pumps only because of the pumping costs, but it is possible to size the flat plate correctly to achieve your goal.
-You will have to be careful in adding coils, you may end up needing a larger pump at your boiler if it is a lot more resistance than the flat plate.
-I like the truck rad idea. There are tank additives that can protect aluminum and iron components. A big pickup rad dumps perhaps 150 kw of heat into air at highway speed when towing a big load. In water without forced circulation it should do almost as well. A cast rad would work as well.
-I will try to scan the coil design for the 120,000 btu/hr Jetstream.

I'll have to take a look over your schmatic to see how exactly it is setup now.

 

[Eric Johnson]

Thanks. Forget the schematic--it's obsolete. And forget the flat plate.

I'm going conventional now--a pressurized loop for heat storage coming from the EKO and a pressurized loop for recovery going to the zones. Still only two pumps on the system and the one doing all the work is the big one. Actually, it's the biggest Grundfos in that popular series.

Thanks for the wall tips. I'll do my best and see how it works.

Hope you had a good trip home. I get dopey after a few hard days on the road.

 

[slowzuki]

The trip was good, we were home at 3 am monday morning! That night we made it up to the 1000 Islands crossing and found a hotel. The frame and axles the next day are a whole other story but we eventually loaded them.