Radiant floor design review

[Montyx5]

Looking for some input and validation of my hydronic heat system planning. Now that my barn has been up for a little while and the floor has been poured with pex circuits installed, I am looking to install the heat.

Here are details of the project:
Metal clad inside and out pole building, area to be heated is 60x64x16 feet with an unheated area 24’ wide attached on one of the 64’ sides (horse stalls). The slab is 5” min. with 2” R-10 insulation underneath and on the perimeter extending down 24” min. Walls are R-19 with house wrap on outside and Visqueen lined inside, ceiling is insulated to R-40+.

An online heat loss calculator gave 50,069 Btu/HR @ 0.35 Ach and 60,851 Btu/HR @ 0.5 Ach. The design temperatures were 0- and 65-degrees Fahrenheit for outside and indoor temps respectively. I believe the heat loss values are a little high due to the 64’ side with the attached unheated room acting as a buffer to outside temps, therefore I am using the 50,069 Btu/HR figure for heating design.

The 60x64 area has 11-300’, ½ pex circuits to a 11-circuit stainless manifold. The pex is placed at or just below middle of the 5” concrete slab. There is also a concrete pad outside the north end of the unheated horse area that will not see direct day light in the winter months with 1 circuit approximately 150’ of ½ pex encased, to clear icing when needed (measurable snow will be removed manually). Dhw required for two sinks and occasional car and horse washing.

For a heat source I have ordered a Westinghouse/HTP UFTC-140WLP combi boiler. I intend to have 1 in. pri/sec loops utilizing Webstone’s 1-1/4" Run x 1" Hydro-Core to interconnect the loops. I see four options for the outdoor slab loop, all installed on the return side of indoor secondary loop: 1. Connect with Webstone’s 1 x 3/4" Primary / Secondary Loop Purge Tee, partially closing the valve just enough to get a small pressure differential between the two sides to get adequate flow to heat pad when needed, 2: Connect with Webstone’s 1 x 3/4" Loop Purge Tee utilizing a small circulator on that loop, 3: Use a heat exchanger and isolate the indoor from out door, 4: use a diverter valve. My thoughts on the three options are: Option 1 would require a higher ratio of antifreeze in entire system and would slightly increase head pressure in system when using the valve. Option 2 adds the cost of a second pump, which may help in selecting the sec. loop pump. Option 3 adds the cost of a second pump and heat exchanger but allows for reduced antifreeze use in indoor heat system. Option 4 diverter valve is costly. Currently considering option 1 with the thoughts that a second pump could be added in if using the valve represents a problem. This loop would not be needed often, but when it is needed it will be worth the effort to prevent injury to any of the horses.

As for sec. circulator, I have calculated that on design day I will need about 0.52-0.78 gpm per circuit for a 19-12.5 deg. drop at 90.3-85.9*F supply temp. respectively (86*F is min. CH setpoint). Nibco’s 0.5 in. pex specifications give 1.0 gpm of water through 300 feet a head loss of 11.07, which changes to 13.18 at 30% pg. Since design gpm range is 0.5-0.8 with correlating head loss of roughly 6.59-10.54 with 30% pg. For circulator selection I am using parameters in the middle at 8.6 ft of head and 7.2 gpm, aiming for a 15*F delta between input/output temps. Using Grundfos selector these values fall just outside of 2nd speed of UPS 15-58 FC – 5989634, the H-10.54, Q-8.58 calculated values land on the high-speed setting as well.

This is where I am at in the design, I appreciate any input on whether I have made any mistakes, wrong assumptions or on the right track.

 

[Firebrick43]

Why such a grossly oversized boiler for your heat load?

Why a primary/ secondary loop? They are for micro zoning to maintain boiler flow(for boiler protection), non condensing boilers, or to use heat emmitters the need different temps such as baseboard and radiant together.

You have one big zone. Use a system pump, heat exchanger first and then the radiant manifolds.

Then use another pump on the antifreeze side of the heat exchanger for the ice melt loop. Use a timer switch on that pump. Simple, ice melt gets priority, less pumps.

If you do use valves to operate the ice melt use a diverter valve. Rigging up a system using fill/flush valves is ripe with mis operation. Don't expect your wife/children/trainer/whomever else have to do anything other than move a single valve or switch.

 

[Montyx5]

Why such a grossly oversized boiler for your heat load?

For the DHW part, combi boilers that I found start around 100 Mbtu. The ones I priced were higher in cost than this one. Other options I considered included two separate water heaters for CH/DHW and CH with indirect DHW tank, but this would have been more costly than the one unit. One of the local hvac Co. wanted to install a 250 Mbtu cast iron boiler in, as that is what he believed would be the minimum size I needed. Three other contractors were between 150-200 Mbtu and only one was at 100 Mbtu (he was the only one who asked specific questions about the building construction and then sat in his van and did some math).

Why a primary/ secondary loop? They are for micro zoning to maintain boiler flow(for boiler protection), non condensing boilers, or to use heat emmitters the need different temps such as baseboard and radiant together.

That is what the manufacture explicitly recommends. Also from every thing that I have read that has been published in the last few years also suggest that this is the preferred way unless the boiler is designed work in that capacity. If the manufacture would publish the specifications on the internal pump and it's operating parameters along with the other needed info then a confident workaround could be achieved. Of what seams like 100 manuals that I have read most of the combos recommend pri/sec loops, whereas the heat only boilers may or may not.

You have one big zone. Use a system pump, heat exchanger first and then the radiant manifolds.

Then use another pump on the antifreeze side of the heat exchanger for the ice melt loop. Use a timer switch on that pump. Simple, ice melt gets priority, less pumps.

If you do use valves to operate the ice melt use a diverter valve. Rigging up a system using fill/flush valves is ripe with mis operation. Don't expect your wife/children/trainer/whomever else have to do anything other than move a single valve or switch.

I have given this alot of thought about using HX, additional pump and controls and am having trouble justifying the cost since it would be used so infrequently across many years. The other problem I also see with the HX is that the outdoor loop is just a single 0.5 in loop, trying to push more than 2 gpm will result in really high dynamic head and would be unable to pull enough heat possibly causing short cycling. This is also the a potential problem I see with an expensive diverter valve. At the time of the install I had not had the time to research the subject adequately, in hind site I would have been better off using a few more loops and/or larger tubing. My thought with using the more simpler purge tee ($44) was to experiment with small movements and try to get about 1.5 gpm through that loop. I could then make a simple bracket that would prevent further moment. When deicing I could just run secondary pump all the time using the stored energy in the slab when not in heat cycle. I realize that this will work best if I start the process before certain weather events occur since the temps through the outdoor loop would be low requiring time to work. If it does not work out it would not be hard to incorporate a HX setup.

Once I have the heat operating properly indoors I am considering leaving the secondary pump on all the time initially. I have other thoughts on controlling the secondary pump post heating cycle to continue equalizing the heat in the pad. I have access to thermal imaging equipment and will better be able to determine if and/or how much post circulation would be beneficial.

Thank you for the input

 

[Montyx5]

If you do use valves to operate the ice melt use a diverter valve. Rigging up a system using fill/flush valves is ripe with mis operation. Don't expect your wife/children/trainer/whomever else have to do anything other than move a single valve or switch.

You got me thinking about this Firebrick43. If I were to simply put the outdoor loop in parallel with the manifold loops I could simply put a ball valve in to isolate that loop when not needed without the potential hazard of the purge tee being accidentally closed completely. This would give me the hotter water while in the heat cycle and would not raise the dynamic head as the purge tee would. I could also leave the valve cracked open a little in extreme cold weather, when deicing ice not needed, to prevent any potential slushing.

 

[finn]

Maybe I missed it, but the building inspector made my contractor install a thermal break between the 32x24’ heated portion of my garage floor and the unheated 32x30’ unheated storage section before he gave the ok to pour.

Sounds like you have a similar design.

 

[Montyx5]

Maybe I missed it, but the building inspector made my contractor install a thermal break between the 32x24’ heated portion of my garage floor and the unheated 32x30’ unheated storage section before he gave the ok to pour.

Sounds like you have a similar design.

There is a wall between the two and the floors poured independently. The perimeters of both floors have 2" insulation on them.

 

[danski0224]

A primary/secondary loop is absolutely required with mod/con boilers.

Best advice I can give is to explicitly follow the manufacturer recommended piping diagrams for near boiler piping. If you follow this, the system will work.

A hydraulic separator can eliminate the requirement for a secondary loop pump, depending on the piping design. The hydraulic separator can also give you the same water temperature at all supply tees, with proper piping. It can also combine air separation and dirt separation in 1 device.

Edit: This particular manufacturer claims that no primary/secondary piping is needed in the brochure here: http://www.htproducts.com/UFT-Boiler.html

At 140,000 btu, I'd make sure that your 1" piping choice is correct. At the industry standard of 20* delta, it isn't (14 gpm, 140,000 btu). Your 15* delta will require more water flow, not less.

RTFM and follow the piping instructions.

 

[Montyx5]

A hydraulic separator can eliminate the requirement for a secondary loop pump, depending on the piping design. The hydraulic separator can also give you the same water temperature at all supply tees, with proper piping. It can also combine air separation and dirt separation in 1 device.

This has me confused. If there is no pump in the secondary loop you would then be required to place pumps in the individual circuits. I fail to see why this is only possible with a hydraulic separator, could this not be done with tees also. Granted you wouldn't combine air separation and dirt separation in 1 device. I suspect a hydraulic separator is beneficial in systems where numerous cast iron components are prevalent and scale and magnetite are a potential issue and/or the pressure differentials between loops are high.

Edit: This particular manufacturer claims that no primary/secondary piping is needed in the brochure here: http://www.htproducts.com/UFT-Boiler.html

Your link is for their heat only boiler. This is the brochure for the unit I have: http://www.htproducts.com/literature/UFTC-Brochure.pdf

At 140,000 btu, I'd make sure that your 1" piping choice is correct. At the industry standard of 20* delta, it isn't (14 gpm, 140,000 btu).

You are correct. One inch copper will carry 10.9 gpm @ 15* delta gives 81,750 btu/hr, which should be more capacity than I ever need.

Your 15* delta will require more water flow, not less.

Not sure what your suggesting here, are suggesting that the 7.2 gpm/11=0.65 gpm per circuit would have a higher delta than 15*. Perhaps you could clarify a little. Thanks for the input, I appreciate it.

 

[danski0224]

This has me confused. If there is no pump in the secondary loop you would then be required to place pumps in the individual circuits. I fail to see why this is only possible with a hydraulic separator, could this not be done with tees also. Granted you wouldn't combine air separation and dirt separation in 1 device. I suspect a hydraulic separator is beneficial in systems where numerous cast iron components are prevalent and scale and magnetite are a potential issue and/or the pressure differentials between loops are high.



Your link is for their heat only boiler. This is the brochure for the unit I have: http://www.htproducts.com/literature/UFTC-Brochure.pdf




You are correct. One inch copper will carry 10.9 gpm @ 15* delta gives 81,750 btu/hr, which should be more capacity than I ever need.



Not sure what your suggesting here, are suggesting that the 7.2 gpm/11=0.65 gpm per circuit would have a higher delta than 15*. Perhaps you could clarify a little. Thanks for the input, I appreciate it.

The hydraulic separator is an engineered product vs building your own with closely spaced tees. It may also allow piping to be done in a more efficient manner. Caleffi has plenty of piping diagrams on their website and some videos.

If I was planning a system, I would size the primary loop to the capacity of the boiler, and I would put at least one extra set of tees and isolation valves in for future expansion. Limiting the primary piping size may show problems later with the indirect DHW side.

There are 2 methods of primary secondary piping- series and parallel. In series, your possible supply temperature changes at each tee location and parallel gives the same temperature at each tee location.

A hydraulic separator can give you the same possible temperature for each zone and eliminate the secondary loop circulator if the standard pri/sec parallel piping method is used. Much less piping than standard pri/sec in parallel. Yes, each zone on the secondary supply side of the hydraulic separator needs its own circulator, and there should be a balancing valve (yes, an actual balancing valve, not a standard ball valve standing in as a balancing valve) on the return. B&G has direct read balancing valves.

Small circulator pumps are available. It is quite possible to do each loop with one circulator... or 2 loops with one circulator. It's also possible to split off your 10 loops into 2 zones of 5 for example, depending on how the tubing was laid out. This could get your circulator size down. Two or more small ECM circulators will draw less power than one big one that isn't ECM. Upfront cost is higher, but long term electrical bill savings will be a factor.

Floor loops with more outside wall exposure would technically require more GPM than a loop that is mostly interior floor space. Again, this depends on your tubing layout. Your .65 gpm/loop is assuming that each circuit is equal in all respects. Maybe they are. Maybe there are some loops that have more exposure to the outside perimeter than others.

If the space can be viewed more as a commercial space with the floor system more for heating than barefoot comfort, a 20* delta will require less gpm and therefore smaller circulators. Your tubing layout may impact this. Lots of stuff online about 10* vs 15* vs 20* and even 30*.

 

[Lonnies Performance]

Here are my concerns...

I have the 200K BTU version of that boiler operating at 16gpm flow with 30% glycol in a 6700sq. ft. building.

Initially I would not say a 140K boiler is grossly oversized for your application.
The reason is you may want/need a higher supply temp to satisfy your outside slab.
These typically need a higher water temp & return a correspondingly lower temp because the slab outside is so cold.

Currently I am running my boiler at 86 supply temp & the secondary is actually supplying at 85 due to a slight temp loss at the T to T connection. It works well, but I see the boiler has to run quite a while to get up to temp on initial startup.

The boiler operates more efficiently at lower temps, which is good, but this may be too low of a temp for your outside slab.

If you have 11gpm max flow (1gpm/loop), you will only be able to get a 25deg rise out of a 140K boiler (35deg at 8gpm), no matter how high you set the boiler setpoint. I see more than a 20deg drop on the return when my system first comes on. I hold my building at 54deg & I get a 56-60 return on initial startup. It ramps up to approx 65 shortly before the space temp is satisfied. My point in all of this is the boiler is running flat out for quite some time before it hits the 86deg setpoint. My heat calc said a 13deg temp differential, but it is in actuality much higher.

 

[danski0224]

Here are my concerns...

I have the 200K BTU version of that boiler operating at 16gpm flow with 30% glycol in a 6700sq. ft. building.

Initially I would not say a 140K boiler is grossly oversized for your application.
The reason is you may want/need a higher supply temp to satisfy your outside slab.
These typically need a higher water temp & return a correspondingly lower temp because the slab outside is so cold.

Currently I am running my boiler at 86 supply temp & the secondary is actually supplying at 85 due to a slight temp loss at the T to T connection. It works well, but I see the boiler has to run quite a while to get up to temp on initial startup.

The boiler operates more efficiently at lower temps, which is good, but this may be too low of a temp for your outside slab.

If you have 11gpm max flow (1gpm/loop), you will only be able to get a 25deg rise out of a 140K boiler (35deg at 8gpm), no matter how high you set the boiler setpoint. I see more than a 20deg drop on the return when my system first comes on. I hold my building at 54deg & I get a 56-60 return on initial startup. It ramps up to approx 65 shortly before the space temp is satisfied. My point in all of this is the boiler is running flat out for quite some time before it hits the 86deg setpoint. My heat calc said a 13deg temp differential, but it is in actuality much higher.

As an example, a cast iron boiler needs to have a return temperature no less than 140*F, even though the system may be returning water that is cooler.

Something must be put into the near boiler piping to ensure that the return water temperature is within the desired range. This something needs to be able to monitor the water temperature and adjust accordingly. A pipe bypass with a couple of valves is better than nothing, but it really only works as intended within a narrow range.

I put a diverter valve into a system that keeps the boiler loop at no less than 140*F, and the boiler loop will run up to 140-180* (depending on what the setback does) and the rest of the system can be returning 90* from the radiators when the setback is on but the primary loop is at 140*.

It isn't my project, but I am sure that something should be done to a mod/con boiler that is connected to a high mass radiant system. The concrete just ***** the heat out of the water way faster than the boiler can add it. Probably a diverter valve and a buffer tank and another circulator to feed out of the storage tank. Just guessing, up to the OP to do the research.

 

[Firebrick43]

As an example, a cast iron boiler needs to have a return temperature no less than 140*F, even though the system may be returning water that is cooler.

Something must be put into the near boiler piping to ensure that the return water temperature is within the desired range. This something needs to be able to monitor the water temperature and adjust accordingly. A pipe bypass with a couple of valves is better than nothing, but it really only works as intended within a narrow range.

I put a diverter valve into a system that keeps the boiler loop at no less than 140*F, and the boiler loop will run up to 140-180* (depending on what the setback does) and the rest of the system can be returning 90* from the radiators when the setback is on but the primary loop is at 140*.

It isn't my project, but I am sure that something should be done to a mod/con boiler that is connected to a high mass radiant system. The concrete just ***** the heat out of the water way faster than the boiler can add it. Probably a diverter valve and a buffer tank and another circulator to feed out of the storage tank. Just guessing, up to the OP to do the research.

A mod con does not need to be kept above 140. The reason for this in CI boilers is condensing(which happens in the 120 ish range) will corrode a CI boiler out in short order. A mod con (modulating condensing) is designed to condense to get its efficiency ratings. The HX is made of stainless to resist the mild acids created by condensing. Running a mod con as low of temp possible is better for efficientcy.

I looked at the UFT boiler installation manual and even though it's a medium mass fire tube boiler it has a built in primary loop pump so yea I guess not doing a pri/sec loop system is moot

However for "explanation" purposes only, just because a boiler is a mod/con does not mean it has to be a pri/sec loop set up especially with a single large zone.

Page 28 fig 5.5 goes into the math of why. Low head mod cons dont exhibit this issue on a single large zone.
https://www.caleffi.com/sites/default/files/file/idronics_15_na.pdf

Few pages down on 34 it shows your 1" tubing to be correctly sized.

 

[Firebrick43]

Again, most layouts in manuals are "recommended" and if you notice almost always involve mixed temp heat emmitters. Some modcon manufactures have much better technical documentation.

For example viessmann vitodens 100 shows a single zone system with out any form of hydraulic seperation. (Page 15)
https://www.viessmann.ca/content/dam/vi-brands/CA/pdfs/wall-mount/vitodens_100_tdm.pdf/_jcr_content/renditions/original.media_file.inline.file/vitodens_100_tdm.pdf

If you do multiple zone then they recommend a low loss header which is an alternate name for a hydraulic seperator but they do allow closely spaced Tees.

 

[Firebrick43]

Here are my concerns...

I have the 200K BTU version of that boiler operating at 16gpm flow with 30% glycol in a 6700sq. ft. building.

Initially I would not say a 140K boiler is grossly oversized for your application.
The reason is you may want/need a higher supply temp to satisfy your outside slab.
These typically need a higher water temp & return a correspondingly lower temp because the slab outside is so cold.

Currently I am running my boiler at 86 supply temp & the secondary is actually supplying at 85 due to a slight temp loss at the T to T connection. It works well, but I see the boiler has to run quite a while to get up to temp on initial startup.

The boiler operates more efficiently at lower temps, which is good, but this may be too low of a temp for your outside slab.

If you have 11gpm max flow (1gpm/loop), you will only be able to get a 25deg rise out of a 140K boiler (35deg at 8gpm), no matter how high you set the boiler setpoint. I see more than a 20deg drop on the return when my system first comes on. I hold my building at 54deg & I get a 56-60 return on initial startup. It ramps up to approx 65 shortly before the space temp is satisfied. My point in all of this is the boiler is running flat out for quite some time before it hits the 86deg setpoint. My heat calc said a 13deg temp differential, but it is in actuality much higher.

His slab loop is only 150' of 1/2 pex. He won't push much through that circuit.

 

[danski0224]

A mod con does not need to be kept above 140. The reason for this in CI boilers is condensing(which happens in the 120 ish range) will corrode a CI boiler out in short order. A mod con (modulating condensing) is designed to condense to get its efficiency ratings. The HX is made of stainless to resist the mild acids created by condensing. Running a mod con as low of temp possible is better for efficientcy.

Right, but there may still be merit to a buffer tank in a high mass radiant system and a mod/con boiler. Another poster commented about the "boiler running forever" on a cold start.

The buffer tank could be kept at say 110*F and the return would be 90* assuming a 20* drop. On startup, the return could be lower, unless variable speed circulators were used with temperature sensors.

But, the boiler should be able to maintain that buffer tank at 110*, supplying 110* to the floor. This should make for faster warm up times and less boiler run times.

I wouldn't be surprised if there was a reset control that could be set up that would adjust the tank temperature automatically.

The system I did doesn't have a buffer tank, but the primary cast iron boiler loop is always at 140*F even if the supply to the radiators is 110* and the return from the radiators is 90*F when the reset kicks in. It works pretty good

 

[Firebrick43]

Right, but there may still be merit to a buffer tank in a high mass radiant system and a mod/con boiler. Another poster commented about the "boiler running forever" on a cold start.

The buffer tank could be kept at say 110*F and the return would be 90* assuming a 20* drop. On startup, the return could be lower, unless variable speed circulators were used with temperature sensors.

But, the boiler should be able to maintain that buffer tank at 110*, supplying 110* to the floor. This should make for faster warm up times and less boiler run times.

I wouldn't be surprised if there was a reset control that could be set up that would adjust the tank temperature automatically.

The system I did doesn't have a buffer tank, but the primary cast iron boiler loop is always at 140*F even if the supply to the radiators is 110* and the return from the radiators is 90*F when the reset kicks in. It works pretty good

I would disagree. In a low mass system(to prevent short cycling) or a system with a CI boiler sure(to prevent condensing temps) But the higher temp rise possible the better efficientcy. Also an ideal heater size, in any system, is better to run 100% of the time at the low design temp. Now ideal and real world are slightly different but there is nothing wrong with running 70 percent of the time when its zero outside. (or in the far north it may be -20).

Heating equipment wear and tear is more from the number of heat cycles as opposed to hours ran. Really not different than your car (city miles vs highway miles).

Startup times are irrelevant in high mass systems. Why would it matter if it takes a day or a couple of days? High mass systems shouldn't and really can't be used with day/night offsets and other features "smart thermostats" use to save energy. They save by lowering btus into the envelope do to the human condition of needing less heat if thier feet are warm.

 

[Montyx5]

Here are my concerns...
The reason is you may want/need a higher supply temp to satisfy your outside slab.
These typically need a higher water temp & return a correspondingly lower temp because the slab outside is so cold.

The boiler operates more efficiently at lower temps, which is good, but this may be too low of a temp for your outside slab.

I see this as time related issue. Higher temps give faster response time and lower temps would take longer, exponentially getting longer as the input nears output of energy in the slab. Its the weather event of rain into sleet and then perhaps little snow as the temps drop relatively quickly to 10* or lower that concern me. The outdoor slab is 350 sq/ft semicircle with a fairly decent slope. If I am proactive and start the warming process 6 hours or so before the expected temp drop I believe I will be good. With these events being so random and infrequent I don't see it worth while to run at elevated temps compromising efficiency 99.9% of the time. I have considered separating that circuit and use an electric micro-boiler if I find that what I have planned does not work out.

If you have 11gpm max flow (1gpm/loop), you will only be able to get a 25deg rise out of a 140K boiler (35deg at 8gpm), no matter how high you set the boiler setpoint. I see more than a 20deg drop on the return when my system first comes on. I hold my building at 54deg & I get a 56-60 return on initial startup. It ramps up to approx 65 shortly before the space temp is satisfied. My point in all of this is the boiler is running flat out for quite some time before it hits the 86deg setpoint. My heat calc said a 13deg temp differential, but it is in actuality much higher.

I'm curious what your temp delta is near the end of your heat cycle and what size/length your circuits are. The calculated temp delta is for steady state condition on design day input. In theory at the calculated supply temp on design day conditions, the system would run nonstop and hold the room temp at design input with all other things being equal.

Appreciate the input and help Lonnies.

Some modcon manufactures have much better technical documentation.

This is what drives me nuts. It would be nice if manufactures would give an equation or even a simple graph of H vs gpm through the boiler. The documentation that was supplied with my unit says repeatedly that the sec. loop gpm must be higher than the boiler loop and never gives the actual gpm value or data for the boiler pump.

Right, but there may still be merit to a buffer tank in a high mass radiant system and a mod/con boiler. Another poster commented about the "boiler running forever" on a cold start.

The buffer tank could be kept at say 110*F and the return would be 90* assuming a 20* drop. On startup, the return could be lower, unless variable speed circulators were used with temperature sensors.

But, the boiler should be able to maintain that buffer tank at 110*, supplying 110* to the floor. This should make for faster warm up times and less boiler run times.

If the heat source is unable to maintain a 110* temp directly than it would not be able to maintain a same temp indirectly, tank temp would diminish at some rate dependent on a lot of variables.

Post 16

I concur

Thanks to all for the input. Can I assume that no sees any major/minor problems in my design? Pump selection appear OK?

 

[Firebrick43]

Minor issues are not enough tubing in the slab(both ice melt and building) but that is already done and it will be ok probably so I didn't bring it up. Also it would desirable to lower length to 250 feet for the loops but you can't change it either. Also while I don't know your specific needs, i would not buy a combi boiler for an occasional use bath but a small I demand or small (7gal) electric water heater and a standard modcon. But that is more for simplicity and easier maintenance.

Your primary/secondary loop flows look right for 1" tubing,towards the high end of 4 fts velocity.

While I would still do a plate HX in series and a pump on a timer, I think a simple ball valve to isolate the ice melt circuit is probably the safest alternative without being complex. Due to the short loop length, flow and therefore heat will naturally choose that loop.

Maybe look into a grundfos alpha2 pump? It's higher cost is offset by savings in electricity and it self adjust to a constant pressure. ECM pumps save 40-50% electricity off the batt from a standard circulator and more due to self adjusting

 

[Lonnies Performance]

For calculation purposes.... GPM x Delta T x 500 = BTU required.

Based on your boiler part number it looks like you are using Propane.
Cost to operate is approx 140,000 BTU/91,500 btu/gal. = 1.53gal/hr.
Mine runs about 4hrs/night then its off most of the day. Based on run time & propane usage, mine averages 80% output.

 

[danski0224]

If the heat source is unable to maintain a 110* temp directly than it would not be able to maintain a same temp indirectly, tank temp would diminish at some rate dependent on a lot of variables.

I recall seeing a bunch of piping diagrams online pertaining to high mass radiant and how the floor pulls the heat out of the water faster than the boiler can add it, even though the heat loss calculations are supposed to be good.

Obviously, this is more relevant on a cold start, or some other event like leaving a door open that cools the floor.

It probably involves a buffer tank at a higher temperature, with some of the return water being used to bring down the temperature before being sent through the floor, and the boiler gets the rest of the return water to heat up and add back into the buffer tank.

I believe the objective is to reduce the boiler run time at start up. I would have to do more digging, and I would if it was my project

I'm sure that there are many, many different possible piping and control arrangements.

 

[Montyx5]

Minor issues are not enough tubing in the slab(both ice melt and building) but that is already done and it will be ok probably so I didn't bring it up. Also it would desirable to lower length to 250 feet for the loops but you can't change it either.

Unfortunately I didn't research enough before installing the floor as I was racing to beat winter weather at the time. As I started to do the dynamic head calculations the benefits of shorter lengths became apparent. I ran 250' just to see the difference and found that there were many more pumps to select from. I also think I would have benefited by running supply and returns the length of one of the walls and tapping in the circuits as needed, instead of all of them entering and exiting in one area.

Also while I don't know your specific needs, i would not buy a combi boiler for an occasional use bath but a small I demand or small (7gal) electric water heater and a standard modcon. But that is more for simplicity and easier maintenance.

I had considered many options. While in the winter months my DHW needs are small, in the summer months the wife's only requirement for this project was the ability to wash the 4 horses back to back weekly. Especially in the hotter months the water used to wash them needs to be fairly warm or they will not stand still. I wish I had the time in the summer months to have started this conversation to planned better but I am still unable to find a way to get 30+ hours out of a day. If it were not for the horses, I would have used an electric point of use water heater.

Maybe look into a grundfos alpha2 pump? It's higher cost is offset by savings in electricity and it self adjust to a constant pressure. ECM pumps save 40-50% electricity off the batt from a standard circulator and more due to self adjusting

That sounds good. I think I will get the UPS15-58FC ($80) for now and get this system up and running. When I get a chance, I will look into this and replace it with an ecm pump giving me a back up pump.

For calculation purposes.... GPM x Delta T x 500 = BTU required.

Based on your boiler part number it looks like you are using Propane.
Cost to operate is approx 140,000 BTU/91,500 btu/gal. = 1.53gal/hr.
Mine runs about 4hrs/night then its off most of the day. Based on run time & propane usage, mine averages 80% output.

For the calcs I used 450 instead for 30% pg. If my setup works anywhere near yours I will be happy. I suspect, weather dependent, mine may be close given the smaller area but higher room temps.

I appreciate both your inputs.

I have read a lot about the benefits of post and constant circulation in slabs. I can clearly see the benefit to at least have some post circulation to continue redistributing the heat in the slab after a heating cycle. As of this time, I have not found a product that gives this type of control. One possible option that I have not dug into that may work is a temp differential controller. It would be nice if there was a timed controller but have not found one yet. Any thoughts on this or products that you are aware of that can so this.

As always I appreciate the input.

 

[Montyx5]

I have read a lot about the benefits of post and constant circulation in slabs. I can clearly see the benefit to at least have some post circulation to continue redistributing the heat in the slab after a heating cycle. As of this time, I have not found a product that gives this type of control. One possible option that I have not dug into that may work is a temp differential controller. It would be nice if there was a timed controller but have not found one yet. Any thoughts on this or products that you are aware of that can so this.

I am curious if anyone has experience with AquaMotion's AM55 circulator. It's night setback feature could be utilized for post heat cycle circulation without the need for additional controllers.

As always I appreciate any input.