3/4" vs. 1/2" copper tubing for heat exhangers?

[rfactor]

I have lurked on this board for a while and thanks to all for the incredible amount of knowledge shared here. However I now have questions but first a bit on my project. I have radiant floor heating driven by a propane cast-iron boiler. I'm going to tie that system in with a home-built hot water storage tank (approx. 550 gals) and a Tarm Solo 30 boiler. I'm going to make my own copper coils.

For the coils, if I have the chance, why wouldn't I use 3/4" copper for coils instead of 1/2"?

3/4" copper has a greater surface area for heat transfer (3/4" x 120' is 23.56 sq. ft. - 1/2" x 120 is 15.7 sq. ft.) and if I fed the tank with 1" pipe to the coils its area is roughly double that of 3/4". It would seem more efficient to me. What am I missing? Is 3/4" tubing significantly harder to work with etc?

Thanks!

Rurik

 

[Duetech]

Welcome to the forum Rurik,
I don't think you are really "missing" much but there are some details that would make your choice more effective. Whatever size tubes you use to make your coils you will have to match your supply line output and return or you will create a bottle neck (back pressure=head "pressure" or system resistance and the more system resistance you have the less efficient your system will be) and as you have stated you will need two 3/4" tubes to match the area of your supply line so you should have two tubes for charging and two for extraction or four 3/4" tubes. That means you would need four 1/2" to match the area of your 1" supply or a total of eight 1/2" tubes to handle the charging and extraction. That said two 3/4" or four 1/2" tubes have a greater surface area than one 1" tube so both will work well if you at least match your supply/return tube capacity. I don't know your btu requirements and a 550 gal. storage tank will help but is a bit on the small size for what I hear a Tarm S 30 can produce and either 1/2" or 3/4" coils can seem like a spaghetti mass in a smaller tank. If your Tarm heats your supply tank before the wood is used up you will go into idle mode so you will have to learn the burning curve for the amount of wood you use for your Tarm burn cycle.

 

[Dune]

I get confused easily, so ignore this if it is B.S., but I am pretty sure that 1/2 in tube has more surface area per volume of fluid than 3/4 in tube.

 

[heaterman]

It's all about flow rate. 1/2" will flow 1.5GPM, 3/4" will flow 4GPM and 1" will flow about 8 while still staying within the velocity generally accpeted by the industry. That being 4 feet per second in a residential application.

If you manage a 20* temp difference at those flow rates you can transfer 15,000 btu, 40,000 btu and 80,000btu respectively.

 

[CarbonNeutral]

I get confused easily, so ignore this if it is B.S., but I am pretty sure that 1/2 in tube has more surface area per volume of fluid than 3/4 in tube.

Surface area per volume of 1/2" = 8

Surface area per volume of 3/4" = 5.3333

So yes, 1/2" has 50% more surface area per volume.


It's been a long time since my thermodynamics classes.....

 

[deerefanatic]

He's right on this....... And if you do the headloss figures, enough coils of 1/2" will have no more head pressure than 3/4"........ The idea is many short (60' or so) coils piped in parallel.

 

[Duetech]

I get confused easily, so ignore this if it is B.S., but I am pretty sure that 1/2 in tube has more surface area per volume of fluid than 3/4 in tube.

As stated by CarbonNeutral you are exactly right about surface area and I have often thought that rather than go the tube route for charging an open storage tank that a person could use something like an old 4 core truck radiatior (or two) to charge thier tank and build an extractor using 1", 3/4" and 1/2" hard copper to beat the cost of all those coils. Flow rate would have to at least match the supply (see Heaterman's post for this topic) but could and probably should exceed the flow rate of the supply. It would be more labor intensive but the extractor would be isolated to the uper portion of the tank (where the hottest water tends to be) but because of equal/reduced flow rate with a greater surface area would seem to be more efficient and cost efficient than a mass of vertical coils. (all .02 without the proof)

 

[Gooserider]

I keep seeing people talking about having seperate heating and extracting coils - this doesn't seem efficient to me. From what I've seen the optimal design is to have the coil spiral down from the top of the tank to the bottom. If you search, you will see several posts about people using one coil for both functions by reversing the direction of flow through it. When charging, pump the hot water into the upper end and out the bottom, to discharge, reverse the flow.

Plumbing this can be tricky, but it is possible - we even got a report of a tech support call to Taco that it is OK to have two circs facing in opposite directions as long as only one can ever run at a time, as there is little or no significant flow resistance to pumping through a non-running circ.

Gooserider

 

[Eric Johnson]

we even got a report of a tech support call to Taco that it is OK to have two circs facing in opposite directions as long as only one can ever run at a time, as there is little or no significant flow resistance to pumping through a non-running circ.Gooserider

I didn't know that. That's interesting.

 

[pybyr]

we even got a report of a tech support call to Taco that it is OK to have two circs facing in opposite directions as long as only one can ever run at a time, as there is little or no significant flow resistance to pumping through a non-running circ.Gooserider

I didn't know that. That's interesting.

WoodNotOil was the one that explored that, and hooked up his storage to work that way-- and he reported that it works well.

In the fine tradition of imitation being the sincerest form of flattery, I've decided to use that same approach; I just started ordering some of the rest of the stuff I need to finish my storage and get it on line before cold weather arrives (which'll be all too soon).

 

[julien]

we even got a report of a tech support call to Taco that it is OK to have two circs facing in opposite directions as long as only one can ever run at a time, as there is little or no significant flow resistance to pumping through a non-running circ.

Gooserider

A drawing here with a termomat and opposite circulateurs

 

[tom in maine]

Hey, I think this is how we have been building heat exchangers!
3/4" (and larger) is not the best choice for most heat exchangers, unless you don't know much about heat exchange.
Tom
www.americansolartechnics.com

 

[in hot water]

Consider an external plate style HX. You get a much better heat transfer with two pumped flows. A coil, regardless of tube size has some limits in a still water, stratified tank. Heat exchange is not just surface area, it's also about flows and turbulence, which happens on both sides of a plate HX, and really aids the heat transfer.

With plate HX you size the exchanger exactly to the load. Prioritize loads with multiple HXers and leverage the tank.

Different temperature loads compete with one another for energy quality and temperature. A small indirect on priority from the boiler would be a great way to produce and store (well insulated) DHW.

If the load for heating is mainly low temperature radiant, really no need to drive the tank so hot, other than storage of course. The higher the operating temperature of the tank, the higher the standby loss. It can never be completely eliminated.

Ideally the system gets it's BTUs from solar input as often as possible. With solar a single large tank loses the ability to operate the array at temperatures necessary to meet the needs of the various loads. Again cover the high temperature DHW recovery first in a separate HX tank, quickly and efficiently. Now the array, or boiler focus on the typical priority loads (DHW first) then picks off the other loads in order of temperature of energy quality.

hr

 

[Gooserider]

I feel a bit more comfortable with the notion of coils in the storage tank as opposed to a flat plate... It would seem to me like having a pump circulating water in the tank would make it harder to maintain stratification, and while you may get more efficient heat exchange w/ a plate unit, how much of that efficiency is going to be chewed up by the added energy needed to drive the additional circulator...

I will probably want to pick Tom's brains a lot more on optimizing coil design, but it would seem to me like the optimal configuration is several short, small diameter parallel tube loops, so that the total cross section area of the exchanger coils is as large or larger than the feeder plumbing - the larger the coil area, the longer the "residence time" that the fluid will spend in the exchanger, thus maximizing the opportunity for heat transfer. By using multiple small diameter tubes, you maximize the surface area for the exchange to take place in, and presumably get a more even distribution of heat input into the tank. By keeping the loops on the short side, you keep the head loss down, so as to not need as much energy at the circulator pump...

What I haven't quite figured out is what is the optimal way to arrange the loops in the tank... I would think the general idea should be to have the loops slowly spiral or zig-zag down from the top of the tank to the bottom, but I'm not sure about the details...

Gooserider

 

[in hot water]

it's a fairly recent concept, AST "Active Stratification Technology." The use of a small ECM circ, 30W or less consumption to layer or stratify the tank better. Several manufacturers have these tanks with internal coils, or external plate style HX available.

In addition to extremely low power consumption these new "smart pumps" have delta t functions built in the watch what is going on in the tank or loop.

Some wild stuff hitting the market to enhance the designs and principles we have used for many years.

Most of the Euro tanks with multiple coils stack the coils in order of temperature requirement. With a hot tank you may be able to use the upper coil as the DHW provider. ErgoMax and ThermoMax do this with 3 copper coils inside the "boiler water" tank. With enough surface area, and correct flow rates, and enough "A" side temperature, you could generate the DHW instantly as needed. Take the storage factor our of the DHW equasion.

Then the coils are stacked below that according to temperature requirement. I have seen tanks with 4 sometimes 5 coils. often the top coil is corrugated stainless for the DHW production, enameled steel or copper coils below.

hr

 

[rfactor]

So to summarize a few of the good ideas I have heard:

- small pipes are good but make sure I have adequate flow
- variable speed delta-t pump is good to match flow to transfer rate

While I like the idea of the flat plate HX, one of my design goals is to keep the number of pumps a minimum.

From what I've heard I'm going to try to have 4-100' 1/2" coils fed by 1" and driven by a delta-t pump.

Does the Grunfos Alpha qualify? Anyone have experience with the TACO variable delta-t pumps (008 or other)?

Thanks!

Rurik

 

[in hot water]

soft coil copper generally comes in 60 or 100 foot coils. I'd steer you towards the 60 foot loops, why not?

I've used the Grundfos MixiMiser delta T pump on wood boilers before. I'm sure the Taco brand is just as good. It really a basic Taco circ with a tekmar variable speed circuit board built on. It can be done with separate components from tekmar and any pump you can find, e-bay for example.

The external controls offer more features like digital display, etc.

The Grundfos Alpha is a delta P circulator. It modulates based on pressure, not temperature.

More and more ECM (high efficiency) circs are on the market. Currently Laing, Grundfos and Wilo offer these in small residential sizes. Same pumping performance with 40- 60% less energy consumption, plus the built in delta p function to adjust the output to the everchanging loads, with zone valved systems for example.

Here is an example of the multi coil tanks we see in Europe. Notice the multiple coils and plenty of ports into the tank to access the various temperature levels.

hr

 

[rfactor]

soft coil copper generally comes in 60 or 100 foot coils. I'd steer you towards the 60 foot loops, why not?

I've used the Grundfos MixiMiser delta T pump on wood boilers before. I'm sure the Taco brand is just as good. It really a basic Taco circ with a tekmar variable speed circuit board built on. It can be done with separate components from tekmar and any pump you can find, e-bay for example.

The external controls offer more features like digital display, etc.


hr

So if I used a setpoint controller like a Tekmar 157 with Taco 007 pump already have I would have the same as the Delta-T pump capabilities but with a view on the actual temps?

 

[pybyr]

HR-- is there any big advantage to a variable speed differential T controller over a fixed-speed differential T controller (on a non-ECM cic)? I was looking at the Tekmar 156 and 157, and then ran across this unit from Azel:

(broken link removed to http://www.azeltec.com/catalog.0.html.0.html#DST-932)

I've liked the Azel items I have worked with so far.

And with a single phase non-ECM motor, I expect that the variable speed won't save appreciable energy in actually running the pump.

But if the variable-speed differential-T will help achieve appreciably better results (with my system which will use an unpressurized tank and a 5x12x70 FlatPlate HX), then I'd be glad to know that (and gain an understanding of why) so that I can go in that direction.

Thanks

 

[in hot water]

soft coil copper generally comes in 60 or 100 foot coils. I'd steer you towards the 60 foot loops, why not?

I've used the Grundfos MixiMiser delta T pump on wood boilers before. I'm sure the Taco brand is just as good. It really a basic Taco circ with a tekmar variable speed circuit board built on. It can be done with separate components from tekmar and any pump you can find, e-bay for example.

The external controls offer more features like digital display, etc.


hr

So if I used a setpoint controller like a Tekmar 157 with Taco 007 pump already have I would have the same as the Delta-T pump capabilities but with a view on the actual temps?

Yes, here is the data sheet and info on that control

www.tekmarcontrols.com/literature/acrobat/d157.pdf

I like Taco products, just feel for about the same $$ it's nice to have a display not just indicator lights.

 

[in hot water]

HR-- is there any big advantage to a variable speed differential T controller over a fixed-speed differential T controller (on a non-ECM cic)? I was looking at the Tekmar 156 and 157, and then ran across this unit from Azel:

(broken link removed to http://www.azeltec.com/catalog.0.html.0.html#DST-932)

I've liked the Azel items I have worked with so far.

And with a single phase non-ECM motor, I expect that the variable speed won't save appreciable energy in actually running the pump.

But if the variable-speed differential-T will help achieve appreciably better results (with my system which will use an unpressurized tank and a 5x12x70 FlatPlate HX), then I'd be glad to know that (and gain an understanding of why) so that I can go in that direction.

Thanks

The variable speed function can take a lot of the bang/ bang (off/on) cycling out of the operation. By watching the delta T it attempts to match the circulator speed to the temperature gain, or loss. Much like a cruise control on a car. True the motor is not as efficient as an ECM style, but I believe running any circ at 40% speed reduces the energy consumption.

I have not used that Azel control. I have used, and like their temperature display and setpoint controls.

There are dozens of solar differential controls on the market. Resol is the brand you see most often. It is sold under a lot of manufacturers brands. Steca is another nice one. The nice thing about these is they have multiple outputs that can run multiple pumps or zone valves in either the differential or setpoint mode. Steca has one that will control 6 loads. Most Resol are two output.

I have a sample of one brand (prototype) that includes a data logger with a SD card to stick into the control. A driver fits into the USB port. You can configure the control on a laptop and transfer the info to the control. Or pull data from the control to the computer. Soon it will work via wifi!

 

[rfactor]

I have not used that Azel control. I have used, and like their temperature display and setpoint controls.

There are dozens of solar differential controls on the market. Resol is the brand you see most often. It is sold under a lot of manufacturers brands. Steca is another nice one. The nice thing about these is they have multiple outputs that can run multiple pumps or zone valves in either the differential or setpoint mode. Steca has one that will control 6 loads. Most Resol are two output.

I have a sample of one brand (prototype) that includes a data logger with a SD card to stick into the control. A driver fits into the USB port. You can configure the control on a laptop and transfer the info to the control. Or pull data from the control to the computer. Soon it will work via wifi!

The Steca looks like tne "one controller to rule them all!"

Rurik

 

[in hot water]

Stecca and Resol have controls that could do all those functions. I had a couple Resol E and M series. Never could figure them out with the "Genglish" translation in the manual. It really is more like a programable logic control where it helps to understand the control platform, piping, and exactly what needs to happens when, in your system. Everything read in C not F to complicate things even more.

hr

 

[rfactor]

Stecca and Resol have controls that could do all those functions. I had a couple Resol E and M series. Never could figure them out with the "Genglish" translation in the manual. It really is more like a programable logic control where it helps to understand the control platform, piping, and exactly what needs to happens when, in your system. Everything read in C not F to complicate things even more.

hr

The Resol units look just the Stiebel-Eltron SOM 7 unit I have.

 

[rfactor]

Update: I decided not to fabricate my own coils and bought them instead from "Tom of Maine" and they are working like a charm. I just got my boiler online (that will another thread with pix) and right now the coils are full flow and transferring heat to my tank. There is little stratification when charging (within 6 degrees at this moment) but the tank stratefies (sp?) nicely when the flow stops.