My radiant floor system has 1400 feet of 1/2 inch Zurn oxy barrier PEX in a 6 inch concrete slab.  There are 6 loops/circuits.  The maximum output of this floor at 3.5 gal per minute is around 35,000 BTU/hr. 

According to the Zurn specs, 100 feet of 1/2 inch PEX at 1 GPM has .6944 ft of head (1.6 psi pressure drop in 100 ft X 0.434 to convert to ft of head)

My circuit/loops are 181 ft, 258 ft, 270 ft, 218 ft, 231 ft, 242 ft.

At 1GPM, I figure ft of head at 1.257, 1.792, 1.875, 1.514, 1.604, 1.680

I have also figured ft of head for 2GPM, 3 GPM and 3.5 GPM and I have those numbers if they will help.

I know I also need to take the manifold, elbows, flat plate heat exchanger, valves etc into account when figuring ft of head, but just looking at the PEX part of the equation, WHAT is my foot of head for pump sizing?  Should I use 1.875 (the foot of head in my longest circuit at 1GPM)?  The foot of head for this circuit at 3.5 GPM (18.9832)?  Some other combination of GPM and ft of head?

All the PEX is on the ground floor, and will act like one huge radiator.  I am planning on using one pump for the entire system (well 2 - one for the hot water heater tank side of the heat exchanger) and like the looks of the Grundfos A series.  Any pump recommendations?

I am using a Vertex 100 LP 50 Gal for DHW and to heat the radiant floor (100,000 BTU, 96% efficiency condensing).  My heat load calculations call for 35,000 BTU/hr for the entire house, based on a delta T of 67 degrees F (2 story, 1399/floor, with the main living area on the second story and ground floor an at level basement).  I have been heating the house with a small space heater, and going by the performance of my passive geothermal, and what tiny bit of passive solar (we haven't had more than a few days of sun since 15 Sep - a VERY unusual year) I think my actual heating load will be much lower.

What size flat plate heat exchanger should I use?  3X8 14 plate? 35,000 btu?  40,000 BTU?  Any brand better than another?

As always, any help or info you are willing to share is much appreciated.

Thanks,
Rosalinda

Smart women have a singular quality.

I am impressed with your math and wish plumbers would follow your good example when designing a radiant floor.

First, you have stated the maximum "potential" output of the floor (which is dependent of water temperature, room temperature, heat load and surface temperature of the floor. You did not state the heat load which is where we start every radiant floor design.

Second, after performing a heat load for the entire radiated area to determine your heat source, a room by room heat load must be performed accounting for the extra load from windows AND which in turn determines the required surface temperature, response time considering the type of emitter, floor coverings and ultimately the design water temperature.

I use outdoor reset control on most of my designs, but in simple basement applications the "Bang-Bang" application you suggest is perfectly fine.

Since plate heat exchangers are sized by flow AND temperature drop for each side of the plates, the design water temperature (the temperature of the water needed to satisfy the design heat load) is indispensable.

Finally,the people that provide the products you buy have are ethically responsible for providing the answers to your questions.

If you want to DIY radiant floor heating-including procurement of all radiant floor heating PEX, pumps and heat exchangers- you would be well advised to find a "full service" radiant floor supplier (usually limited by their inventory) or hire a certified radiant floor designer to make sure you get what you ultimately want...perfect comfort.

www.badgerradiantdesigns.com

If the individual loop demands come out to be about equal, my *guess* is to size for ~3ft of head at the total (3.5?) gpm.

I'd also oversize the heat exchanger somewhat - you are running domestic water through it and there may be some decline in performance over time.

Somewhat? Overtime?

Soft water is a given, no degradation of performance in the first 10 or 20 years.

OK, sounds like you know that Rosalinda's DHW is completely soft.

Thanks everyone.

Assume the individual loop demands for heat are equal.

The heating load for the entire house, based on a room by room analysis, taking windows and doors and everything else into account, is 35,000 BTU/Hr. That does not include passive solar heat gain, or passive geothermal. I don't have modeling software that will work with my old MAC computer so the passive stuff is theoretical, or experimentation.

My water has a high mineral content - calcium. In the new house, I am planning on using an inline filter before the Vertex 100, once I can figure out exactly what I need. My current hot water heater in the old house, is an 80 gal Bradford White electric, that has lasted 24 years with maintenance, and still going strong (doesn't that sounds like an invitation for disaster?). The water on the radiant side will be distilled, so it will be the tank side that gets the mineral build up. I only believe the Vertex claims to minimize mineral problems just so far. I will keep up with the sacrifice anodes other tank maintenance. So yes, a bit of oversizing in the heat exchanger sounds like a good idea.

I like to design stuff on my own, and the radiant tubing is already in the concrete slab, based on my design to fit the house that I designed. For the most part I read endless books, then think things through, and pick the brains of knowledgeable people for the stuff I don't understand. I am confused about how one determines pump head. I understand the resistance part, but am not sure about WHICH number gets used for the radiant tubing in order to size the pump.

At 3.5 GPM (the max speed that should be required IAW my Delta T of 67 degrees F) my longest loop of 270 ft, has a feet of head number of 18.9832. But at that speed, isn't only one 6th of that 3.5 GPM going through each loop making the feet of head 3.17? Or is the water traveling at 3.5 GPM through each loop, so that I would need 18.9832 foot of head - which seems VERY high to me? I know I am missing something but I am not sure what. I guess I should have borrowed a book on fluid dynamics.... So if anyone can explain this in words of one syllable so I can understand it, it would be much appreciated.

-Rosalinda

First things first, soft water extends the life and performance of everything from shampoo to the heat exchanger.

Distilled water is not used or needed in residential heating systems (soft water is good).

If you were engineering commercial processing equipment you might be concerned with head pressure to the .0001 but not here. Pumps are sized to the longest loop and highest flow. Example: 4 gpm system, 20° delta T, 8 loops, .5 gpm each.

Look at your chart to find head loss at required flow and size pump accordingly.

Grundfos UP1558 heating side and equivalent in bronze on the "softened" potable side.

Merry Christmas

Thanks Badger,
I have ready access to distilled water, so that is an easy solution on the radiant side.

My potable water is from a spring, and is absolutely wonderful, with a great taste. I really don't want to soften it, just get some of the mineral out for the hot water heater and clothes washer. I have never had a problem with sudsing or much else, other than needing vinegar once in a while to remove build up in kettles or the shower heads.

The Grundfos 15 58 is a relatively inexpensive way to go as long as the one pump will handle the whole system. I think my local supplier even carries it. Thanks for reminding me that I needed bronze for the open potable side.

Anyone have an opinion on the Grundfos A 15 55 ? Is it worth it's high price (3 times or more than the price of the 15 58) in energy efficiency?


-Rosalinda

Unfortunately, there isn't a simple filter that will remove hardness. So you either run a softener for DHW or live with the performance degradation and cleaning of your heat exchanger.

Don't forget the head loss from the heat exchanger.

I cna't tell whether or not you have read Modern Hydronic Heating, but if you are planning on (correctly) engineering a radiant heating system, I'd heartily recommend it. It'll cost you about $125, and you'll gloss over about half of it, but it's an invaluable resource in places where you least expect it to be. I'd also recommend, if you do design the sytem yourself, that you find a professional installer to do the actual installation. That person will know about all of the stuff that you don't care to know about - shutoff ball valves to make isolating parts of the system easier, flanges around pumps, heat source installation, etc. That is also invaluable.

And finally, I have a document from FlatPlate that treats the subject of szing flat plate heat exchangers in some depth, including using "close approach" systems where the delta T between the "hot input" (fomr the heater) and the "hot output" (to the floor) are close in temperature, which is presumably what you desire. For example, it speaks of designing to 6 degree, 8 degree, and 10 degree approaches - very interesting. If you'd like a copy of that document, drop me a note at jeff at baronanddangelo dot org.

Jeff

The ICC make hard water to the kitchen tap a must, the rest is softened.

I have the same arrangement. You won't live without soft water (or spring water for drinking) once you have it.

Thanks folks - I was afraid of that. I guess I will have to add a softener before the hot water system. I don't use tap hot water for cooking or food prep anyway, so I guess it is not the end of the world, and I would like to have less than more maintenance work.


Hope everyone had a great holiday, and has a happy and healthy New Year to come.

-Rosalinda

A demand based water softener (that measures usage before it cycles) will minimize salt use. For example, one based on a Fleck 5600SXT. Not sure what is less expense/work - a 2x oversized heat exchanger and occasional cleaning or installing and filling a water softener.

Wow, that is a lot of money to soften the water, to replace the couple of gallons of vinegar I currently use each year to remove excess mineral from things.

According to the Flat Plate doc Jeff sent me, the Flat Plate folks recommend in a well water/high mineral situation you use their Marine series plates. Their FP 5X12-10, the exchanger I think would do the job for me, runs $300, and the comparable MPN 5X12-10 $878. I wonder if I am just better off going with the FX plate, skipping the water softener, and just plan on cleaning the plate every year or two, and planned replacement when performance drops, along with an annual flush out of the Vertex 100 and close monitoring of the sacrifice anodes.

In addition to the expense of the water softener, the thought of adding all that salt to the water, that will take a turn through my septic system, filtrate into my soil and eventually work its way down to Cayuga Lake - the water shed my farm feeds in a fairly direct path, does not thrill me either. Any thoughts on the ecological impact of water softeners?

-Rosalinda

Is there a reason to not use distilled water as long as the pH is around 7?

-Rosalinda

Anyone have any tips on getting the @@@#$%$ 3/4 npt plugs out of the side outlets of a Vertex 100 hot water heater? There is no way to put any heat on them, though I guess I could try my hairdryer. I called AO Smith and they said brute force - but after spending 2 grand on this baby, brute force seems like a last resort. I assume the plugs have pipe dope on them.

-Rosalinda

Never mind - I wacked the wrench a couple of times before adding a length of pipe for more torque, and managed to get them loose

Is there a problem with using some galvanized fittings in the manifold system such as elbows or couplings? All the galv will be separated from the copper with a brass fitting, and there wont be too many of them. The price of all brass is prohibitive, and I am really trying to keep solder out of the system altogether if I can.

-Rosalinda

Galvanized pipe and fittings should not be used for any residential purpose, more especially hydronics. "Solder" on the other hand when properly defined, is universally accepted for all hydronic applications. Brass is used but not necessary.

Yeah, you are right. I talked myself into using them, and started to think I had made a mistake on the way home from the supply house. Guess I need to fire up my torch.

-Rosalinda

Badger - would you not limit 1/2" pex to 250 feet per run? Also would it not be better to have all runs with in 10 % of each other in length? Assuming single zone.