OK. one more thing.

I have never seen a residential hydronic system that needed an energy-wasting double-wall plate heat exchanger.

You may want to install a few more pressure gauges - pressure drives flow.

But Rob has convinced me that "the PSI only helps flow in DOMESTIC USAGE mode."

OK, to be more precise, pressure differential drives flow. That's why more gauges can show you why things are flowing the way they are. Whenever the flow is slow or the wrong direction, it's because the pressure differentials aren't correct.

It must be very frustrating to deal with amateurs like me who are so clueless! I get the same thing in my industry. People think it can't be that hard but they can't really anticipate cuz they've never done it before. They also don't realize how much knowledge you accumulate after doing it for years and years. Not surprisingly they often find themselves stuck and in need of help from pros! But I also really believe in learning by doing, and I definitely feel like I'm getting an education on radiant heating systems.

That said, advice on components for the HX system is exactly what I'm looking for! If you can tell me how, I'll do a heat load calculation. I did find a worksheet from the Radiant Floor Company, but we made some changes to the system after they did the calculations. Their worksheet shows:

Zone 1: 525 lf tubing for 600 sq ft. Est min BTU/HR = 11,000 Tubing BTU output = 21,000 (note that we put 600 lf tubing in 650 sq ft instead)
Zone 2: 300 lf tubing for 420 sq. Est min BTU/HR = 8,000. Tubing BTU output = 15,000 (note that we put 300 lf tubing into 300 sq ft instead)
They did not do a calculation for Zone 3 because we added it later but it is 200 lf of tubing in 250 sq ft

Please let me know how to adjust the calculations above or start from scratch so that I can get the right heat exchanger. The Radiant Floor Company will sell me a preassembled HX system but, honestly, I'd rather find the components myself and have my plumber assemble. Also, to clarify, we want to use the most efficient HX possible - heat loss is def a concern. Rob had mentioned the double wall HX for some places in CA but if we don't need it, great! In terms of the BFP - Rob recommended cold water makeup to the radiant side which will be connected to the rest of our domestic plumbing. In that case, it seems like we need a BFP. Please clarify.

In terms of the rest of the piping & pumps, I was thinking of using the same supply / return lines and zone circulators as in our existing system. How do I determine if that will work or not? And finally, how do I determine what pump to use as the system circulator?

Jon - Good point. Seems like there won't be much (any?) pressure differential once we switch over the HX system. No?

Enjoy the weekend everyone!

I found an online heat loss calculator. I come up with 15,000 Btu/hr which seems low - it is less than the estimate above. Perhaps because my house is well insulated and fairly tight. Anyway, I realize the online calculator isn't super accurate but using those numbers plus adjusting the tubing BTU output by doing some math I come up with:

Zone 1: heat loss EST 10,000 BTU/hr Tubing BTU output = 24,000
Zone 2: heat loss EST 2,500 BTU/hr Tubing BTU output = 15,000
Zone 3: heat loss EST 2,5000 BTU/hr Tubing BTU output = 8,000

Which seems to explain why my house is toasty warm when all zones working.

In terms of sizing the heat exchanger and getting the right pump to work as a system circulator, I think I'm going to consult with the salespeople at PEXsupply.com but if anyone has any recommendations or advice please do share.

Thanks!

we use cold water makeup on all closed systems. you can skip it if you are going to manually tend to your system pressure w/makeup water and hoses, but most people don't want to do that.

the double wall heat exchanger is required by code in california, last I knew. Unless that's changed, or you are in an area without serious inspection, that just is what it is. However I do not stay current on CA's codes so that might be out of date info... you should check locally on that. I agree it's a stupid requirement.

Also, there is no way you're getting 50 BTUs per linear foot of pipe out of a joist system. I don't know how you are calculating tubing output but that's "wildly outside the realm of possibility". In your application you would max out at around 15 BTUs, maybe 20 per square foot... and that, only at fairly high water temps.

Funny how one question / answer tends to lead to another....

I'm sold on the cold water make up on the closed side of the system but now I have what are probably some dumb questions. What solution should I use in the closed system? I'm assuming some form of anti-freeze and also assuming that the BFP will keep that solution from entering our DHW but I could use a little more explanation. Also, in term of monitoring system pressure, if the system is pressurized at 12-15 PSI when it is filled, will the cold water make up automatically maintain that pressure? (Again, regretting my lack of basic physics education.) Finally, I'm curious if I should increase the outflow temp setting on the Takagi when switching to the HX system? (to compensate for heat loss?). Our current setting is 131F.

When we first installed our open radiant system, the only inspection we needed was a pressure test. But our entire renovation is permitted and meets code so I will def check on heat exchanger codes requirements. Does anyone have ballpark in terms of heat loss on standard HX vs double wall HX?

Regarding the BTU output of the pipe, I just used the numbers from the worksheet provided by the Radiant Floor Company which showed 21,000 BTU output for zone one (525' tubing stapled under joists) then just recalculated for 600' of tubing in Zone 1 and 200' of tubing in Zone 2. They show 15,000 BTU for 300' tubing in the slab. I also checked their website this morning which states "The 7/8" Poly provides the highest heat output (50 BTU's per ft.) and can be spaced 16" on center." Also, they say "The 1/2" PEX ( 25 BTU's per ft.) can also be used in virtually any application, but it should be spaced 8" on center." I'm not suggesting the info is accurate, only showing how I got the numbers I did....

50 BTUs/sq ft would require you to wear insulated slippers to avoid discomfort. I try not to explicitly condemn my competition, but I have a pretty dim view of such... misleading... claims made by so-called professionals. You might, under some optimal testing condition like startup in a cold slab, get 50 BTUs/sq ft out of that pipe... but never in a typical heating situation. lucky for you your heat load is apparently low enough to be serviced by this system.

fill valves are pressure regulation devices, so they use the higher potable pressure and bring it down for the closed system.

I don't like antifreeze, unless you have a serious freeze risk. but if you did, you somehow avoided it thus far...?

I am not sure the heat loss difference between the heat exchangers, but the cost is very different, and the efficiency isn't as good. if heat loss is a major concern for you... i.e. if this is in cold space... insulate the pipes and HE. If it's in conditioned space, it's not heat loss, not really.

we haven't had a freeze problem because the system is always on. but now that both my husband and i travel for long stretches (months) for work and also because we live in an area with mostly 2nd homes where many people completely drain their domestic water and don't come at all in winter - for resale value we thought it would be an advantage. ALSO: we occasionally lose power (queue my personal frustration with above ground power lines). what are pros / cons of using antifreeze?

the radiant system is installed in an outdoor closet where any heat loss is lost. it's insulated from outside but the shared interior walls are also insulated. but i guess efficiency is really more of a concern.

sorry, more questions:

how do i determine the GPM flow on the radiant side using 3x grundfos UP-29su? the pump specs say Flow Range (GPM): 0-20.5 but i'm assuming head pressure is a factor and i don't know how to calculate. help?

also, i figured out that the HX that the Radiant Company recommends for my system is a 1" 5x12 - 20. i'm thinking it can't hurt to go with a HX with more plates but please let me know if there is a downside i'm not thinking of...

one thing i'm learning in this process is why we were so inclined to go with the Radiant Floor Company in the first place. we knew we were not experts on hyrdronics but felt we had skills to do the labor. the idea of purchasing a package with everything we needed - and that was partially pre assembled - was very appealing. i'm wondering if there are any pros out there that provide a design service for a price? ie that would be built on accurate loads and other info and would provide the customer with a design / installation plan and shopping list. or some such...

Posted By worthydocs on 16 Jun 2012 01:58 PM
what are pros / cons of using antifreeze?

how do i determine the GPM flow on the radiant side using 3x grundfos UP-29su? the pump specs say Flow Range (GPM): 0-20.5 but i'm assuming head pressure is a factor and i don't know how to calculate. help? 

i'm wondering if there are any pros out there that provide a design service for a price? ie that would be built on accurate loads and other info and would provide the customer with a design / installation plan and shopping list. or some such...

a) antifreeze?
Try to avoid, unless a new closed zone is not recommended without , and figure a 35 to 50% pumping penalty on flow, a 'packaged deal tech' should pick these ?'s off with you.
 

b) Standard pex pd's added to kinks and bends in your systems fittings and installation are in water known to have a specific pd per 100 ft of lengths and can point to GPM by reading pressure drops. ... find a drain port location that you can put a gauge on to (on a hose-cap fitted, sometimes common hardware stores carry these ) ...gauge differential readings, may be simple enough.

c) for a price?
loads, est
info, est
design with E&O insurance:  not a PE.

private message of envelope and internal zone equipment heat gains etc, can do that for the 3 zones, and a review before freely posting numbers is reasonable, here, to me, but here's a couple more things to be sure of right now:

1) 7/8 px if 3/4" id it is very common to hit 2-2.2 ft per sec velocity at ~3.4-3.5 GPM to make sure the air is out of that zone.  Every time you introduce new waters to the water loop gasses and other entraapped air comes in.

If normally there really is not 3.1/2 to even 4 gpm flowing at bursts for maintenance in EACH 3/4" id, then a blockage or cavitation arises at points and at the pumps. Antifreeze requires the 4+ for over 2.7 ft per second VELOCITIES, I have seen needed in pex.
Velocities over 5 ft per second flowing-speeds- are common in industrial designs.

2) I may have missed it: is the 600ft "linear" or 3 x 200 off parallels?

3) At 15 btuh's per sq.ft, not 50- (steady-state BTUh per square-foot), floors should get pretty warm, eventually for zero-deg.F days heating loads  met

4) Can you simply restrict the flows at the outlets of pumps, again isolated at other zones not being tested, one at a time, and then open one, listening for any return- pitches, crackles or other than quiet flowing sounds? and  see if a cavatation sound can even be heard-  not necessarily meaning none exists. Cavitation in a pump body with tornado-eddy currents, can bore tiny holes through the cast iron.

Nice catch jon- "PD".
Great discuss's , and don't miss Mike's notes.
If those are iron pumps, then sell them or get the closed loop to a HX   NOW !
Some HW tankless require a minimum differential of TEMPERATURE; and some need over 25-deg cooler fluid hitting the inlet.
Is that a case?

-DIY tech links seem helpful:
http://www.insulation.org/articles/#fn1

A measured pressure drop across anything in the system in line with the pump or the pump itself, can get crossed to a chart and a known be estimated. I see drain ports and hopefully those can be where you put a couple of matched gauges .
for GPM
 Measure that drop "on" to a 1/4th of a psi can be compared across the pump at 68-75 deg, or on what the chart is based.

Work/Performance other than GPM here:
Another useful formula BTUh = gpm x temp-diff x 500 for water
gpm=
can be double checked by work.
         BTUh (if work can be, or is to be compared to loadings/outputs at the zones) / ( 500 x dT )

If 2 gpm was used at a HW tap, and dT is measured to tenth, you can know some performance in BTUh. 
A standard meat Thermometer for 9 -11 bucks reads dT perfectly well enough.
ie) 2 gpm x 500 x dT = BTUh

---and again what I mean is if the dynamic heat of the system can be identified one way or another, then other zone loads can be observed for understanding outputs by work, and checked against GPM estimations.

hey knotET,

i hate to admit it, but most of what you are saying is completely over my head.

what i can glean is that you agree with rob to avoid antifreeze if possible. i think you are suggesting that without enough flow (as measured by velocity or GPM) air that comes into the system with fresh water can gather in points of tubing or in pumps and cause the problems we've been experiencing. assuming that will be resolved by switching to HX. zone 1 is a total of 600 lf tubing in 3x 200' circuits. it is super simple to isolate the zones. whenever we've made any changes / additions to our plumbing i've purged air from the system zone by zone - i think that is what you are describing. i do that for 10 mins per circuit in zone 1 and 10 mins ea for the other zones. after 10 mins i don't hear any signs of air. all my pumps are stainless. our takagi is activated by flow - i don't believe there is a require temp differential.

otherwise, i'm pretty lost on the various calcs and various other info you describe. it may simply be that i don't have the right vocabulary - i might be able to understand better in plain english. for example i've read this sentence perhaps 20 times and still can't make sense of it:

Also any way you can get a 0-30 , or 0-15 psi gauge, drop the loop pressure in that range to be reading across the radiant isolated loop at the lines only, pump "on" w/ all other zones isolated?

since i know that my tubing produces enough heat to warm the house in the existing system (so long as all three zones are working together) i'm inclined to leave all the existing tubing / piping / pumps when converting to a HX. but i'm not sure if my logic is sound? leaving the existing piping i'm thinking a 1" 5x12 24 or 30 plate HX - hopefully not double wall. i'd like to get more advice on sizing the HX but not sure if i have all the required info to make an informed decision. i also plan to use my existing expansion tank but add an air separator/elminator as well as another pressure gauge, pressure relief valve and a BFP / fill valve on the cold water makeup. will also add a drain and circulating pump on the takagi side of the HX and need advice on selecting the right pump. hoping PEXsupply.com will give some advice.

i'm also not sure if i will need to raise the outlet water temp setting on the takagi from it's current 131F when switching to HX?

i'm pushing my limited aptitude for math and physics but i'm willing to learn with more remedial instructions / explanations!

I've been away for a couple of days so i haven't commented. First, put the pump on the pipe leading INTO the HX, then into the Takagi (and don't bother with a double wall, waste of money and only there so the municipality can have extra insurance against litigation).

Instead of putting glycol (propylene not ethylene which is car antifreeze), have an alarm set to 45F or something like that which will alert a neighbour.

If your heat load is 15Mbtu then you will only need 1.5GPM at a 20 deg temp difference. This is really hard to do with 3 circulators as they will be barely moving the water to meet the heat load. The total restriction on the suction side of pumps from the fittings and from the mix valve, at the high velocity creates the problems with cavitation. I will bet that if there was a pressure gauge on the suction, it would be outside the comfort range of the circ even at 25psi static.

Look at a watts 9D backflow preventer and an watts 1156 pressure reducing valve for keeping 15psi in the radiant loop. Exp tank is always on suction side of pumps, as is the water feed.

The pumps are SS and I use the same ones a lot. Don't like B&G, stick to grundfos.

Add up the flow rates of the pumps because I'll bet they are going full bore. I need to look at the drawing again to see if the valves are balanced at all, if possible to see. If going full bore, almost ALL the PD will be between the pump and the mixing valve. There will be a torrent in that pipe.

I am not expert, so I would ask the experts to comment, but I see check valves (BFP) on the inlets of your pumps in your pictures. I used one of these valves on my system, and if I remember correctly they were supposed to be mounted horizontally, not vertically. The valve is just a flapper, and it seems to me that if you mount it vertically the insertion loss is going to be much higher at low flow (it has to push the flapper up against gravity) then at high flow (once the flapper is vertical the weight just sits on the pin). Conversely, if you mount it horizontally, the loss will be low at low flow (where the flapper is just opening, and most of the weight is on the pin), and higher at high flow (where the flapper is being held horizontal by the fluid flow). So my thought is that this could be the additional loss that you overcame when you added the booster pump, and that you might have better luck getting the 3 zones to share flow without the pump if you changed the check valve orientation.
On the rust stains it sounds like you didn't use any ferrous parts in your system, so you shouldn't be introducing iron. Perhaps you should check with you neighbors and see if they have issues with rust stains, perhaps it is an artifact of your water.

Here is what I think I know based on the discussion so far:

- the flow problem in our open radiant system will be resolved by converting to a HX system
- avoid using antifreeze in the radiant side
- use cold water makeup and a combo fill valve / BFP (i'm looking at the 1156/9D combo) to keep pressure at 15psi
- make sure the expansion tank / air elminator is on supply (suction) side of pumps
- put system circ pump between Takagi outlet and HX (which is going to be challenging spacewise - that's why i put on other side of HX in sketch)
- will adjust the Takagi outlet temp until the radiant side falls in the 120-135F ideal water range for the PEX tubing in my system.

Please let me know if I got any of the above wrong.

Here is what I still don't know:

- can i use the existing pumps / pipes / tubing on the radiant side of the HX system?
- sizing of HX itself. is there any downside to adding extra plates?
- selecting system circ pump on takagi side of HX
- do i need to be concerned by the GPM flow thru my radiant system? (seems like an important factor but i'm still totally lost here on how to calculate)

Other things I'm not thinking about from my simplistic perspective?

Dave111- the zone manifold came preassembled with the check valves in that orientation. will defer to the pros about whether that could be a factor?

First, check valves should be on the discharge side of the pumps. As little restriction as possible should be placed on the suction.

Second, these check valves are OK on a vertical pipe, with flow going up.

Third, if the heat load for the area is only 15Mbtu (which I doubt, I think it will be double that given the R-values), allow 1GPM for each 10Mbtu of heat loss, you will need a more powerful pump to PUSH through the Takagi and the HX. I would suggest a UPS-15-55. This will give you 3 speeds so you can experiment.

Fourth, Use the valve on the discharge side of the radiant pumps to balance down. Start at 50% closed.

Fifth, over time reduce the temp on the mix valve to to point where you are still comfortable. Then reduce the takagi temp a bit as well. Many people can have 105F and still heat and have comfortable showers.

Hey Mike,

Thanks so much for all your advice. To make sure I understand:

Are you recommending the UPS-15-55 for the Takagi side of the HX only? Or are you suggesting I replace the 3x UP15-29su zone circulators?

Are you also recommending that I close the ball valves above the zone circulators by 50%? assuming we should reconfigure the zone manifold so that each zone has a ball valve, then pump, then check valve, then another ball valve at 50% closed. correct?

The UPS 15-55 is for the takagi side and it is a bit of an art finding the right pump sizing for this which is why i like the 3 speed pumps best.. Keep the UP15-29s where they are but they are, individually, too big for each loop. You only need a 30w pump for each one and turning each valve down to 50% is just a start as ball valves don't start to throttle down untill they are 50% closed. The goal is to have the pump running as slow as possible to get the desired dT across the loop (20 deg) and still be within the pumps comfort zone. The boiler needs a minimum temp drop to fire properly as well and this would help.

The system SHOULD have the valve, pump, check valve and ball valve but it may not make that much difference to move it at this time.

got it. thanks!