Newbie Radiant Install

josephny

I'm set on installing a radiant heating system in an old house (i.e., some, but poor insulation -- lots of heat loss). LP tanks.

From inside the basement, I have open access to the 2x10 16" OC floor joists under the 1st floor.

I understand Uponor pex and heat plates are very good, and I understand (I think) the looping process and runs under 300' and insulation under (lower) the pex/plates.

I understand a boiler is much smarter than a hot water heater (even a tankless), and I'll need an expansion tank, pump, gauges, sensors, drain/fill spigot.

The first floor is about 22x38, and I think 1 zone would be just fine. I'm having a very difficult time doing a heat loss analysis because the first floor extends past the borders of the basement -- that is, there are rooms that are supported by piers outside the footprint of the basement. I was going to just ignore those for now, and leave those for additional zones.

But that's about as far as I got (not that I've got that down pat).

I'm trying to understand the piping at the boiler, and reading lots of posts here and getting more and more lost. Specifically, using the straight through vs. bull (Tee) side of a tee, mixing valves and control circuitry is not something I have any grasp of.

Anyone have any advice on how to proceed? It's January and it's cold and the electric heat is working very hard and costing me a lot of money.

Thank you!

EzzyT

Did you have a Heatloss analysis and a design done? That’s where you need to start.
Where in NY are you located?
We just wrapped up a full house radiant project in Queensbury (Lake George Area).

josephny

EzzyT said:

Did you have a Heatloss analysis and a design done? That’s where you need to start.
Where in NY are you located?
We just wrapped up a full house radiant project in Queensbury (Lake George Area).

I have tried, but hit a dead end quickly.

I'm in Sullivan County, NY. Basement is 22x38 (836 sq-ft) but first floor is ~$1,700 sq ft (irregular shaped bump outs on all sides, constructed at various times over the past 60 years).

The exterior of the first floor has varying levels of insulation.

There is a 2nd and 3rd floor also, each 836 sq ft. But, the house is renovated, so I'd like to install the radiant just under the first floor (and probably just under the the main (over basement) area.

Thanks!

EzzyT

Are there crawl spaces underneath the 1st floor parts of the house with no basement?

EzzyT

Are you planning on doing the work yourself?

josephny

EzzyT said:

Are there crawl spaces underneath the 1st floor parts of the house with no basement?

Yes, some portion of the additions have accessible crawl spaces.

josephny

EzzyT said:

Are you planning on doing the work yourself?

Yes.

josephny

Don't know if it will help, but here's the first floor layout.

EzzyT

Are you using any software for the design or are just winging it based on the maximum loop lengths specifications as per manufacturer.

josephny

EzzyT said:

Are you using any software for the design or are just winging it based on the maximum loop lengths specifications as per manufacturer.

I've looked at basic formulas using delta-T/U/R/etc., and prebuilt spreadsheets, but they all require info I don't have. I have no idea what the R values of the walls/floor/ceiling/roof is, nor U values of the glass in use is, etc.

EzzyT

That info is must in order to insure the proper Heatloss is done. Then one can move forward on a design to determine if radiant alone will be sufficient enough for heating those areas of the structure.
The job in Lake George we just did two of the rooms required supplemental heat besides the radiant so we installed panel radiators along with radiant in those rooms.

hot_rod

josephny said:

EzzyT said:

Are you using any software for the design or are just winging it based on the maximum loop lengths specifications as per manufacturer.

I've looked at basic formulas using delta-T/U/R/etc., and prebuilt spreadsheets, but they all require info I don't have. I have no idea what the R values of the walls/floor/ceiling/roof is, nor U values of the glass in use is, etc.

You might consider hiring one of the pros here to at least do a load calc. That would at least tell if if it is a workable floor plan and what to expect for materials needed.
A few pics of the underfloor would help. If there are a lot of nails protruding it will not be an easy plate installation.
You may end up with a hybrid radiant and panel radiator system.

There are some assumptions that need to be made when calculating an existing building, someone that does load calcs knows how to estimate those unknowns.

josephny

The house has panel electric heaters now, which are doing the job, but not efficiently.

What can I do if I don't know those numbers? I know lots of the walls have a single layer of paper backed fiberglass batting, and the windows are just cheap big box store windows.

If we guess, and then oversize the boiler but use a modulating one, use 300' pex runs, would that work?

EzzyT

You needs to get those numbers close enough but if you build in some fudge factor and have a good design down along with a good mechanical room layout you can be in great shape.
I wouldn’t rely on the just the maximum 300’ loop length method, we rely on a system design and play around with numbers once that’s all been squared away.

HomerJSmith

When I have done under floor pex with plates and there are nails protruding thru the sub-floor, I have cut them off with a 4" grinder which can be purchased from Harbor Freight for peanuts.

It was easy and worked great. Just wear eye protection and have some water in windex bottles, just in case.

josephny

hot_rod said:

josephny said:

EzzyT said:

Are you using any software for the design or are just winging it based on the maximum loop lengths specifications as per manufacturer.

I've looked at basic formulas using delta-T/U/R/etc., and prebuilt spreadsheets, but they all require info I don't have. I have no idea what the R values of the walls/floor/ceiling/roof is, nor U values of the glass in use is, etc.

You might consider hiring one of the pros here to at least do a load calc. That would at least tell if if it is a workable floor plan and what to expect for materials needed.
A few pics of the underfloor would help. If there are a lot of nails protruding it will not be an easy plate installation.
You may end up with a hybrid radiant and panel radiator system.

There are some assumptions that need to be made when calculating an existing building, someone that does load calcs knows how to estimate those unknowns.

I've attached some (bad) pictures of the under side of the joists.

My local distributor has generously offered to the analysis, but unless I can provide the data, he would be just guessing.

josephny

EzzyT said:

You needs to get those numbers close enough but if you build in some fudge factor and have a good design down along with a good mechanical room layout you can be in great shape.
I wouldn’t rely on the just the maximum 300’ loop length method, we rely on a system design and play around with numbers once that’s all been squared away.

That is very interesting Is there an online calculator that could let me play around with the variables?

I do really want to learn this stuff.

josephny

HomerJSmith said:

When I have done under floor pex with plates and there are nails protruding thru the sub-floor, I have cut them off with a 4" grinder which can be purchased from Harbor Freight for peanuts.

It was easy and worked great. Just wear eye protection and have some water in windex bottles, just in case.

Absolutely. While adding some time to the install, I'm not worried about nails interfereing with the plates sitting flush. I might be concerned about the pex hiting a sharp nail, so I'll have to be extra careful.

EzzyT

Slant/Fin has a free Heatloss app and Loopcad has 30 day free trial on design along with Heatloss calculations.

josephny

Youngplumber said:

I was going to post pictures of the info you need. It's a lot of info though. I do suggest you hire someone to get you over the humps you are stuck on. I took a week long class and did many examples of different buildings /layouts/ designs. To get a grasp of what is needed to estimate a job like your doing. 

It takes time and experience to accurately do what you want and people get paid to figure out the things you want to know. This is a site to help and and people do get lots of help on here, but spilling the beans about every aspect doesn't help the business side of things. Not that everything is about business all of the time. 

Keep in mind I'm new here but that is my opinion. I'm sure @ChrisJ will comment and tell me I'm way out of line, lol. 

I do understand that many (or nearly all) of you are professionals and you have a reasonable need to protect your valuable assets -- one of which is the expertise you've worked long and hard to acquire.

josephny

Okay, very cool!

I used the slant/fin (as best I could).

I made room 1 the area over the basement and room 2 is one additional area/extension and room 3 is another additional area/extension.

Here's what I got (and I probably erred on the high side of the heat loss):

Floor 1 TOTAL FLOOR BTUs: 20820
Room 3 SELECTED BASEBOARD: BTUs: 7524
Room 2 SELECTED BASEBOARD: BTUs: 1541
Room 1 SELECTED BASEBOARD: BTUs: 11755

Does that mean that room 1 (836 sq-ft with 11,755 BTU/HR loss) needs 14 BTU/Sq-ft? And if so, that's not too difficult to achieve?

If my calculations are in the ballpark, and I eventually would like all 3 "rooms" heated, then I'll need a boiler/heater that can provide 21000btu/ht, which is the net amount after taking into account boiler efficiency, right?

And, a modulating one (Lochinvar 10:1?) means that I can't go very wrong getting one that can handle 50000BTU/hr, right?

I'm also interested in the possibility of a combi to produce DHW.

Okay, fire away -- how many errors have I made (;-)?

Ironman

If you have and old, poorly insulated house, then I seriously doubt that 14 btus per square foot is anywhere close to the actual heat loss at design temp. It's probably at least twice that much.

Load calcs are only as accurate as the data that's entered.

If need be, post the square footage of your walls, Windows, doors, exposed ceilings and floors with their respective construction data, as well as your locale, and we'll see how close your numbers look.

EzzyT

What exactly @Ironman has said.
I would stay away from a combi unit and go with a boiler and indirect tank. 

josephny

Room 1, which is the space above the basement:
Room Height (ft) 9
Room Length (ft) 38
Room Width (ft) 22
Door (sq ft) (Add manually OR click arrow to calculate) 3
Doors Factor (Add manually OR click arrow to calculate) 0.75
Glass (sq ft) (Add manually OR click arrow to calculate) 150
Glass Factor (Add manually OR click arrow to calculate) 0.65
Exposed Wall Length (ft) 100
Exposed Wall Factor (Add manually OR click arrow to calculate) 0.1
Cold Partition Length (ft) 0
Cold Partition Factor (Add manually OR click arrow to calculate) 0
Ceiling Factor (Add manually OR click arrow to calculate) 0
Floor Factor (Add manually OR click arrow to calculate) 0
Infiltration Factor (Add manually OR click arrow to calculate) 0
Indoor Temperature (F) 72
Outdoor Temperature (F) 5

josephny

Room 2 (modified a little from original):
Room Height (ft) 8
Room Length (ft) 8
Room Width (ft) 10
Door (sq ft) (Add manually OR click arrow to calculate) 21
Doors Factor (Add manually OR click arrow to calculate) 0.75
Glass (sq ft) (Add manually OR click arrow to calculate) 40
Glass Factor (Add manually OR click arrow to calculate) 0.65
Exposed Wall Length (ft) 20
Exposed Wall Factor (Add manually OR click arrow to calculate) 1
Cold Partition Length (ft) 0
Cold Partition Factor (Add manually OR click arrow to calculate) 0
Ceiling Factor (Add manually OR click arrow to calculate) 0.1
Floor Factor (Add manually OR click arrow to calculate) 0

4064 BTUH

josephny

Room 3 (modified from original):
Room Length (ft) 40
Room Width (ft) 16
Door (sq ft) (Add manually OR click arrow to calculate) 44
Doors Factor (Add manually OR click arrow to calculate) 0.75
Glass (sq ft) (Add manually OR click arrow to calculate) 200
Glass Factor (Add manually OR click arrow to calculate) 0.65
Exposed Wall Length (ft) 20
Exposed Wall Factor (Add manually OR click arrow to calculate) 0.1
Cold Partition Length (ft) 0
Cold Partition Factor (Add manually OR click arrow to calculate) 0
Ceiling Factor (Add manually OR click arrow to calculate) 0.1
Floor Factor (Add manually OR click arrow to calculate) 0
indoor 72
outdoor 5

13641 BTUH

josephny

Modified room 1 to include .1 ceiling factor and correct door sq footage:

18097 BTUH

Total:

35802 BTU/H

josephny

Yea, I didn't realize it was door sq-ftage. I modified it for post #22.

I don't know how to account for window or door infiltration.

Basement is not heated -- mineral rool between joists.

josephny

I changed room 1's window/door infiltration factor to .018 and now the htu/h is 27000. Huge changes.

Thank you for pointing these out.

Ironman

I would have guesstimated 32 btus per square foot from your earlier description and that's what your latest numbers put you at.

Here's the long and short of it: it's generally recommended to keep the supply water temp at 120* or less for wood floors. Using good aluminum heat transfer plates, you'll get about 20 btus per square foot output from a radiant floor at 120* SWT, depending upon your floor covering. This was proven by a scientific study at VA Tech about a dozen years back and we have means of calculating it now too.

So, the output of a radiant floor would fall short of heating the house from the lower 20's and below. If you choose to door the floor, you're gonna need supplemental heat when it's colder outside. Panel rad's or low temp baseboards are a couple of options.

josephny

Ironman said:

I would have guesstimated 32 btus per square foot from your earlier description and that's what your latest numbers put you at.

Here's the long and short of it: it's generally recommended to keep the supply water temp at 120* or less for wood floors. Using good aluminum heat transfer plates, you'll get about 20 btus per square foot output from a radiant floor at 120* SWT, depending upon your floor covering. This was proven by a scientific study at VA Tech about a dozen years back and we have means of calculating it now too.

So, the output of a radiant floor would fall short of heating the house from the lower 20's and below. If you choose to door the floor, you're gonna need supplemental heat when it's colder outside. Panel rad's or low temp baseboards are a couple of options.

Keeping the electric for supplemental, radiant would be 120* max water temp, 300' runs, 8" OC, Uponor heat transfer plates? Flow rate? Boiler size?

HomerJSmith

Is that wood floor plank or engineered 12mm flooring?

As far as insulation in the joist bay under the pex, this is what I do. I use a polyurathane 1" solid insulation board with the foil side up against the pex which creates about a 1" space between the insulation and the sub-floor and then put fibre glass insulation under the insulation board. This spreads the heat pattern over the whole joist bay and not just where the plates are and it really prevents backloss.

You really have heat energy resistance thru the subfloor and flooring. Spreading the heat across the whole joist bay would increase the radiation into the room thru the flooring. Am I being redundant?

josephny

HomerJSmith said:

Is that wood floor plank or engineered 12mm flooring?

As far as insulation in the joist bay under the pex, this is what I do. I use a polyurathane 1" solid insulation board with the foil side up against the pex which creates about a 1" space between the insulation and the sub-floor and then put fibre glass insulation under the insulation board. This spreads the heat pattern over the whole joist bay and not just where the plates are and it really prevents backloss.

You really have heat energy resistance thru the subfloor and flooring. Spreading the heat across the whole joist bay would increase the radiation into the room thru the flooring. Am I being redundant?

Floor is 3/4" plywood with 3/4" solid wood planks -- I don't know what kind of ply, but pretty sure the flooring is oak.

The 1" foil face up rigid insulation is a great idea.

I can't say I understand exactly what happens to the heat emitted by the pex/plates as it passes through the flooring, but my basic research shows the plywood-oak combination is about an R 1.5. Not that that fully explains (to me) what happens to the energy emitted by the pex/plates, but I understand that directing/reflecting as much as possible upward is best.

hot_rod

Most of the heat transfer is via conduction, the tube contact to the plates and the plate contact to the floor.
I would not worry as much about the foil, it will dust over in short order and not add much value. It needs to be clean and shiney to do any reflecting. I'd spend more money on r-value under the installation, at the very least 6" of batts.
The end of the joist bay, the rim joist is the biggest concern. I'd spray foam and put blocks of foam board there. You have both temperature and infiltration concerns at rim joists, a high heat loss detail.

HomerJSmith

Heat energy goes in a straight line to what ever is coldest. Your rim joist are certainly cold. You want the money you spend on heat to go into the room and not outside. hot_ rod's comment is right on.

I put 1-1/2" Rmax or Celotex against the rim joist and blocking.

josephny

hot_rod said:

Most of the heat transfer is via conduction, the tube contact to the plates and the plate contact to the floor.
I would not worry as much about the foil, it will dust over in short order and not add much value. It needs to be clean and shiney to do any reflecting. I'd spend more money on r-value under the installation, at the very least 6" of batts.
The end of the joist bay, the rim joist is the biggest concern. I'd spray foam and put blocks of foam board there. You have both temperature and infiltration concerns at rim joists, a high heat loss detail.

Thank you so much!

I have R-23 Rockwool in there now that I was going to put back under (lower than) the pex/plates, but I will put 2" rigid XPS sprayed at the edges inside the rim joists first -- thanks to you advice!

josephny

HomerJSmith said:

Heat energy goes in a straight line to what ever is coldest. Your rim joist are certainly cold. You want the money you spend on heat to go into the room and not outside. hot_ rod's comment is right on.

I put 1-1/2" Rmax or Celotex against the rim joist and blocking.

I had never heard of Rmax or Celotex, so your comment gave me the opportunity to research it -- Thank you!

Rmax is Polyiso, which I've been reading about, specifically as it differs from other forms of rigid insulation board (highest R value per inch, some variations come with reflective coating, among other differences). A quick look doesn't show what makes Celotex stand out.