pressurised systems and non pressurised systems, whats the difference?

[pybyr]

Books - generally the one book that gets pointed at is Modern Hydronic Heating, 2nd Edition by John Siegenthaler, P.E. - It is regarded as the "Bible" on the subject of hydronics, and while probably less ambiguous than the Bible, it is similar in size, complexity, and being a tough read... Also it's a relatively expensive book. There are other books that also get pointed at, but with far less concensus, most noticably the titles by Dan Hollohan of HeatingHelp.com fame...

Gooserider

Goose- I think you raise good points all around. Only thing I'll note is that I managed to get a copy of Siegenthaler's masterwork by way of interlibrary loan, as I was not ready or able to throw well over $100 to purchase a book sight unseen. I found it to be a GREAT book, and spent several weeks devouring it. Athough I did not need to buy it, on account of the fact that I gleaned what I needed for my system, I was really impressed with it and would say that it took my understanding of both the nitty gritty and big picture of hydronics from clueless newbie to bold and empowered designer and construct-er of my own system.

 

[Gooserider]

Glad to hear the positive review - Have to say I haven't read it myself, but was going by the comments of other folks here, and on some of the sites that sell it...

Ironically, there was a recent post pointing to a Canadian industry magazine that had a couple of Sigenthaler articles in it, and some ways I was less impressed by the layouts in those articles than I've been by some of the ones I've seen here... Hopefully the book would be better, or maybe there were subtleties I wasn't getting...

Gooserider

 

[Nofossil]

*meaning that I would keep all the plumbing I could in one pressurized zone - I would probably end up with three independent sets of fluid -
1. The pressurized water circuit (boiler, house heat loads, DHW heating, storage heat exchange, etc. which could be configured in multiple zones and other interconnected loops),
2. The glycol circuit (Solar heating panels, garage heating, and any other freeze prone items that would be better with the use of a glycol antifreeze solution instead of plain water)
3. The storage tank - non pressured, non circulating

You have described my system EXACTLY. Having run it in this configuration for three years, I'm contemplating a switch to pressurized storage. In my case, the issue is that any heat exchanger loses transfer capacity as the temperature differential decreases. That means that with boiler outlet water at 180, it's really easy to get storage from 120 to 130, really hard to get it from 160 to 170, and impossible to get it from 170 to 180.

The ability to have additional in-tank heat exchangers is the reason I went with open storage, and it's an important advantage. In addition to the charge/discharge coil, I have a DHW preheat coil and a coil for extracting heat from the solar panels. If I go pressurized, I'll probably do some large surface area sidearm variant for both of these functions.

I have the Siegenthaler book. It's invaluable, and I'd suggest that if you can't read and understand what he's saying you should spend some extra time getting system design help from someone who has that level of knowledge. It may not be your local plumber, though....

I don't disagree with Siegenthaler on any major points, but I find it interesting that in one section he goes to great lengths to show the total cost of a circulator including electrical power consumption over the system lifetime, making the point that the kWh add up over the years. Then, he completely fails to mention that primary / secondary systems consume two to five times the electricity of other designs, or that zone valves consume a tiny fraction of the power used by circulators.

 

[SolarAndWood]

You have described my system EXACTLY. Having run it in this configuration for three years, I'm contemplating a switch to pressurized storage. In my case, the issue is that any heat exchanger loses transfer capacity as the temperature differential decreases. That means that with boiler outlet water at 180, it's really easy to get storage from 120 to 130, really hard to get it from 160 to 170, and impossible to get it from 170 to 180.

Is your primary motivation to increase the effective capacity of your storage?

 

[Gooserider]

*meaning that I would keep all the plumbing I could in one pressurized zone - I would probably end up with three independent sets of fluid -
1. The pressurized water circuit (boiler, house heat loads, DHW heating, storage heat exchange, etc. which could be configured in multiple zones and other interconnected loops),
2. The glycol circuit (Solar heating panels, garage heating, and any other freeze prone items that would be better with the use of a glycol antifreeze solution instead of plain water)
3. The storage tank - non pressured, non circulating

You have described my system EXACTLY. Having run it in this configuration for three years, I'm contemplating a switch to pressurized storage. In my case, the issue is that any heat exchanger loses transfer capacity as the temperature differential decreases. That means that with boiler outlet water at 180, it's really easy to get storage from 120 to 130, really hard to get it from 160 to 170, and impossible to get it from 170 to 180.

The ability to have additional in-tank heat exchangers is the reason I went with open storage, and it's an important advantage. In addition to the charge/discharge coil, I have a DHW preheat coil and a coil for extracting heat from the solar panels. If I go pressurized, I'll probably do some large surface area sidearm variant for both of these functions.

Sounds reasonable, though I assume that one can slightly improve on the in-tank heat exchanger performance by using more tubing and / or better performing exchange tubes... I was thinking of using some stuff like what these guys make - would that do better than plain copper pipe? In actual practice, I don't know that I have a lot of choice, as MA currently requires ASME stamps on pressure vessels, and I would have a great deal of difficulty in getting any kind of sizeable / affordable pressure tanks into our basement in any case...

I have the Siegenthaler book. It's invaluable, and I'd suggest that if you can't read and understand what he's saying you should spend some extra time getting system design help from someone who has that level of knowledge. It may not be your local plumber, though....

No problem with either idea, and didn't mean in my comments that his book was incomprehensible, just that many of the comments suggested that it was a technically oriented book written for those with some level of knowledge on the subject, and as such was not an "easy reader" book...

I don't disagree with Siegenthaler on any major points, but I find it interesting that in one section he goes to great lengths to show the total cost of a circulator including electrical power consumption over the system lifetime, making the point that the kWh add up over the years. Then, he completely fails to mention that primary / secondary systems consume two to five times the electricity of other designs, or that zone valves consume a tiny fraction of the power used by circulators.

Exactly! You put your finger on the biggest thing that bothered me about his suggestions in the articles I mentioned - he seems to be very fond of throwing lots of circs into a system, which IMHO defeats the purpose - why take money away from the oil / gas companies in order to give it to the electric companies? IMHO one of the desired design standards should be to minimize the amount of electricity that is needed to distribute the heat.

This is also a potential issue I have with some of the designs I see that seem to want to have circs running 24/7 - I suppose if one is talking a really tiny circ it might be possible for it to consume less power running all the time than a larger circ running intermittently, but I would really want to see some hard numbers on that, as from what I've seen even the larger Taco's don't seem to be all that high power draw...

Gooserider

 

[DaveBP]

I don’t disagree with Siegenthaler on any major points, but I find it interesting that in one section he goes to great lengths to show the total cost of a circulator including electrical power consumption over the system lifetime, making the point that the kWh add up over the years. Then, he completely fails to mention that primary / secondary systems consume two to five times the electricity of other designs, or that zone valves consume a tiny fraction of the power used by circulators.

Yeah. Ironic, isn't it.

I hope in future editions of the book he will give more attention to comparing the different overall strategies in terms like electric power usage and maintenance and repair costs.

(And I also hope that someone I know buys it so I can borrow it and not buy it).

 

[pybyr]

I was thinking of using some stuff like what these guys make - would that do better than plain copper pipe?
Gooserider

I'd looked into that product as it had been highly recommended by several knowledgeable people on here (Joe Brown and/ or Hot Rod Bob Rohr). The cost was pretty staggering. For my planned DHW loop within my unpressurized storage, I ended up getting about 70 ft of corrugated 3/4 stainless tubing through East Coast Metal Hose-

http://www.eastcoastmetalhose.com/home.htm

The 70 FT will run in loops around the upper perimeter of my 6x6 by 5 ft high tank

My aim/ expectation is that the corrugation of the stainless should increase the surface area, and also increase the turbulence (and thus surface contact of the fluid) within the tubing.

The HX between the boiler system and the tank will be a 5x12x70 plate FlatPlate.

I've got most of this stuff on hand and the tank already partially assembled- just need to re-ignite my motivation to finish the storage and connect it together. My Econoburn did a pretty nice job this past winter, once I put it on line on 1-15-09 even without storage, but it all ought to get even better with storage.

 

[Gooserider]

Interesting looking stuff, but the only stuff I saw looked like the braided stainless (or bronze) tubing I use for brake lines on my bike, wonderful stuff but it has a lining inside it, with the metal braid around the outside for strength, and I would expect it to have fairly poor heat transfer characteristics... If it is more of a corugated pipe like a longer version of the expansion joints they were picturing, it might work very well...

Gooserider

 

[pybyr]

I've already got it in my cellar-- it is corrugated 300-series stainless, with no braid, nor any non-metallic lining- and the vendor that I mentioned brazed NPT pipe thread fittings on the ends of the 70 FT of corrugated for me.

 

[tom in maine]

I am not quite getting the rationale of using corrugated pipe or other materials instead of smooth copper for heat exchangers.
Maybe it is because I have spent a lot of time and energy over the past 30 years checking out these alternatives.
There are a lot of research papers done on this whole issue and every time (I would say except where someone is trying to sell their product)
the cost/benefit of smooth copper is much better than any extended surface tubings.
Since we manufacture smooth copper hx, I guess you can say the same about us, but NREL and a bunch of engineering schools have done the testing.

Now, if you own the stuff, an extended surface material might make sense, but most folks are buying this stuff--for a lot of money. I think when you factor in the ease of fabrication, smooth copper is a great immersed hx. Smooth PEX can work also at 3-6x the lineal footage.
Extended surface materials are usually using refrigerants or are other applications other than water/water.

The configuration of the heat exchanger is significant, but when properly designed in a regular tank, performance can be very good.
Corrguations do not do as much as we would like to believe in this application.

Give it a try, but I would suggest trying to research it more, if you are planning on spending a bunch of money in high hopes of gaining some great performance gain.

 

[Gooserider]

I am not quite getting the rationale of using corrugated pipe or other materials instead of smooth copper for heat exchangers.
Maybe it is because I have spent a lot of time and energy over the past 30 years checking out these alternatives.
There are a lot of research papers done on this whole issue and every time (I would say except where someone is trying to sell their product)
the cost/benefit of smooth copper is much better than any extended surface tubings. <snip>

Interesting - I will admit I haven't done any testing, but I would have thought there would be benefits to using the increased surface materials - ideally with surface extensions on both sides... On the outside I would have expected the benefit to come from increased surface area for heat exchange, and on the inside, from the surface increasing the turbulence in the flow, bringing hotter water into contact with the tube walls, and breaking up any laminar flow patterns, again increasing the heat transfer rate - presumably reducing the amount of tubing length needed in the exchanger.

Of course, I guess that if you put enough feet of smooth pipe in the tank, the above advantages would cancel out... Depending on the relative costs the smooth pipe might be cheaper.

Which case is it, that the extended surface stuff doesn't give as much increase in heat transfer as one would expect, or that the cost differential isn't good enough to justify it?

(Would there be advantages in hitting the salvage yards to see if one can find used stuff, or is it flat not worth the effort?)

Gooserider

 

[pybyr]

Tom- not to disagree with you- you obviously have a depth in this subject that exceeds mine- I'd be interested in links to the NREL and academic studies you mention, and I expect that others would also.

I sourced the stainless corrugated tubing for the DHW loop partly because my tank is also stainless and I didn't want to get into issues about dissimilar metals setting up long-term funky dielectric/ corrosion factors over the long haul.

One commercial application I can think of that does use corrugated stainless is the Haase heat tank. But oh my goodness, is that expensive

 

[tom in maine]

The one seminal paper is one that I have in my office. I will get it scanned in and email it to you.
Davidson, at the University of Minnesota has done a lot of research for NREL on immersed heat exchangers.
You might Google some of them. Their focus has been on inexpensive smooth polymer hx, but after all the work done, NREL has collaborated on a warm climate solar collector with integral storage that uses a smooth copper hx. And for good reason, as you said, corrugated exchangers are expensive.

As I recall, there is a barrier effect in corrugated tubing that negates the extended surface benefits in immersed water/water applications and also induces more flow restriction than one would want to see. My engineer friends get into Reynolds numbers, etc. and I start to glaze over. Bottom line is with Dick Hill, my retired ME friend, who developed the first gasifiers, who says that corrugated tubing is not worth looking at in our applications.

Once I found some papers corroborating what he told me, I have focused on using smooth copper.
Is it perfect? Absolutely not. But nothing in this world is.

 

[Gooserider]

Don't know just why but I have less trouble believing what you say about the outside of the exchanger than I do about the inside... Just out of curiosity do you know if anyone has tried doing anything with "turbulator" type inserts in the exchanger pipes?

Gooserider

 

[tom in maine]

Turbulators on the inside would work well, except how do you put them in a normal hx?
You need to get turbulent flow in the coils since that is part of the equation.
I suspect extended surface on the outside is worthwhile, it is inside where the problem arises.

I think much of this issue is plain old dollars and cents. Extended surfaces are too expensive and do not give the advantage
that everyone thinks they do in a water immersed water to water heat exchanger.

There is a lot of subtle physics taking place in tanks with coils, between volume of the coils, temps, stratification, flow, and
coil thermal performance. The only way to test this is to cut and try.
The engineers I dealt with had weird ideas about heat exchangers. We had many coils that did not work and sorted it out initially by trial
and error in solar drainback tanks.

There is no greater motivator than a pissed off customer who cannot get enough hot water in January in Maine!

You don't have to believe me, but if you are looking at a big bill to buy extended surface heat exchangers, you might want to compare it to
smooth copper. There are enough people here who are using them that you can easily calculate the costs.

Trendsetter and Haase use extended surface heat exchangers and they cost a lot of money. I will bet they use them to look impressive
and imply some sort of high tech performance. Or they never did any homework and met a smooth corrugated sales person.

 

[DaveBP]

When I was scheming on how to make my copper tube heat exchangers with all the tube benders available at work I had the notion to mount some 1/4" dowel pins at opposing 45 degree angles on the jaws of a vise and indent the tubing 1/8" or so every few inches to create a helical pattern along the length of the tubing. The idea was to make a spiral turbulator effect along the water path that would produce a spiral water flow inside the tubing without restricting the flow too much ( through 3/4" tubing). Going to revolutionize the industry, I was.

Checked into the price of corrugated tube and fin tube and figured you could buy a whole lot of plain old copper tube for the same money. That's when I thought of "enhancing" the interior turbulence of the copper tubing.

Of course, like a number of other times I was going to revolutionize the industry with a bright idea of my own, I got distracted (in this case by a pair of 500 gallon propane tanks) and never did manage to change the world. Alas, next time....

 

[in hot water]

A lot of changes in heat transfer design these days. I noticed several of the European tank manufacturers switching some, if not all of the coils to corrugated. It's a surface area game, no question. Same with the turblators inside the tube. Check out copper tube boiler tubes and you will see the devices they insert into the tube to "stir the water.

I read an article on Delphi recently. They build radiatiors, oil coolers and condensors for the automotive industry. They have been adding turbulators to their products to increase transfer, just as we do for the EKO tubes. Same with the HVAC industry. Trying to squeeze any performance increase is what heat transfer engineers do, the quest goes on.

The Triangle Tube tank in tank indirect and solar tanks use a corrugated inner tank to get more surface area.

To compare fairly you would need coils built from the same materials. I've not seen corrugated copper coils used in tanks. Finned copper, yes, but smooth coils have been shown to out-perform fined if the surface area is exactly the same. The German Solar Energy Society has some good data on smooth vs finned copper heat exchanger tubes. that subject. Copper is a better conductor, than stainless so comparing copper to stainless would not be fair.

I've been working with a company in Ohio that builds HX tube and turboltors as a way to increase performance of solar coil tanks. See the picture of one of the sample tubes they sent us for testing.

Here is another company that deals with turbulators www.alcotwin.com/turbulators.html

Yes, it comes down to what is not only do-able, but at what cost, for what performance increase. And certainly the pressure drop and increased pumping power needs to be considered.

I suspect high copper costs drive manufacturers to look at other materials, like stainless.

Yet another example of surface are can be seen in flat plate exchanger. Now if you really want performance, skip the tubes in a tank and consider a small plate stye. Lots of performance from a small block. Surface area, turbulance, and two pump flows are the key.

hr

 

[Dune]

Sorry to jump into this so late, but I think the primary consideration would be the heat conduction rate of the metals involved. Since copper's rate of heat conduction is aproximately 6(six) times that of stainless steel, there is not much of a question. As far as finned conductors, these would make more sense in a water to air system, since most of the conduction is from the edge of the fins, effectively adding very little surface area in a water to water system. I have to agree with Tom, the most economical and practical system is smooth copper tube.

 

[DaveBP]

I want to see the machine that makes the stuff in Hot Rod's photo.

 

[Nofossil]

Sorry to jump into this so late, but I think the primary consideration would be the heat conduction rate of the metals involved. Since copper's rate of heat conduction is aproximately 6(six) times that of stainless steel, there is not much of a question. As far as finned conductors, these would make more sense in a water to air system, since most of the conduction is from the edge of the fins, effectively adding very little surface area in a water to water system. I have to agree with Tom, the most economical and practical system is smooth copper tube.

Heat transfer is an astonishingly complex process, and almost any first-approximation mathematical model is wrong - sometimes shockingly wrong. In a fluid-to-fluid heat exchanger, turbulence and convection are both important. Here's a way to look at it. First, a static model of a coil-in-tank system:

Imagine a hot water molecule on the inside of the pipe, at an average distance from the pipe surface - say 1/4". We need to transfer the heat from that water molecule to a cold water molecule that's an average distance from the outside ot the pipe - say 6". There are 5 thermal barriers to deal with:

1) The 1/4" of water between the hot molecule and the pipe wall.
2) The boundary layer between the water and copper (or stainless)
3) The metal of the pipe itself
4) The boundary layer on the outside of the pipe
5) the 6" of water between the outside of the pipe and the average 'cold' water molecule.

Of these, the water itself is by FAR the biggest insulator. My guess is that the boundary layers are next, and the metal pipe is last, whether it's copper or stainless.

There is another factor, of course - how much time you have before the hot molecule is circulated away.

On the outside of the pipe, this model is too simple. Still fluids naturally establish convection currents as heated fluid rises and cool fluid falls. This serves to bring fresh cooler fluid into contact with the heated surface and reduce the average distance between the hot surface and the average cool water molecule. However, there is a maximum convection velocity which is much too slow to accomplish optimum heat transfer. For that reason, anything that increases the amount of external surface area will help dramatically by bringing more cool water in contact with the heat source.

On the inside of the pipe, the static model is also too simple. For the diameters and flows that we're going to have, flow is very turbulent and water molecules are continuously bouncing around from the center to the outside. The only thing that can help here is more surface are and more time. A larger diameter pipe helps in both cases. Because the flow is already turbulent, additional turbulence enhancers probably don't make much difference.

If you wanted to get hard data, what you'd want to do is measure temps at multiple spots and look at the temperature differences:

1) fluid, center of pipe
2) fluid, at pipe wall
3) pipe wall, inside surface
4) pipe wall, outside surface
5) fluid, at outside of pipe wall
6) fluid, short distance above pipe

I suspect that you'd find virtually no difference in the first two, showing that the fluid is turbulent and well mixed. There would be a larger drop between the fluid at the pipe wall and the pipe wall itself (both inside and outside), but virtually none between the inner and outer surfaces of the pipe. The biggest drop would be the fluid at the outside pipe wall vs. fluid an inch or two away.

If I'm right, that argues for the largest diameter pipe that you can do, with fins on the outside if possible.

 

[tom in maine]

Nofossil, you have hit the nail on the head, mostly.
There are a lot of variables that are constantly changing in a seemingly static tank environment.

Many folks just throw a lot of copper into the tank until it works. That is pretty expensive in the best of times and suggests lack of engineering by the manufacturer.
I believe many of the extended surface heat exchangers suffer from boundary layer effects and fluted and corrugated heat exchangers can affect flow
rates. I think a lot of these extended tubes were originally designed for condensing or evaporating refrigerants--a whole different animal than that water to water heat exchange environment inside a tank.

Fins might help, but when one factors in the cost and the possibility of improper orientation of the fins or more importantly, the damage that fins can reek on a liner, I suspect barrier effects and cost cannot overcome the cost effectiveness of smooth copper.
As one who has analyzed this for over 25 years, I keep coming back to smooth copper. Would like to think there are other options, but other than plate hx with extra pumps, there are no real good choices other than a well designed smooth coil in an immersed environment.

The one time I think larger tubing diameter might be more helpful is with PEX, but that might only afford you some preheated volume before the rules of thermodynamics come and bite you and force you to use 3-6x the surface area of copper.

Have a Safe and Happy Memorial Day.
Let's Remember all our Veterans and the 9/11 people, too.