I have a staple up radiant floor system, 7/8 in thin wall pex on 16" centers, thin aluminum plates 7" wide covering 95% of the pipe.  2" air space, then Radiant barrier and R-11 batt insulation below.  Slightly heated basement space below, never below 55 deg or so.

My question is this...

If I install additional 7/8 tubing, effectively doubling up the tubing, 8" on center,  and new plates to cover almost all of the exposed subfloor (instead of only half now) any rough idea what percentage increase in BTU heating capacity of floor system might be expected?

A little more detail:

3/4 plywood subfloor w/ 3/4" maple hardwood floor above.  Mostly exposed, say 20% floor has area rug.  Trying to get down to 120 deg max water temp.  Cold NH climate.

Thanks for any ideas...

Sub floor systems are the most challenging to engineer. I use the industry standards ignoring gimmicks like 7/8", thin wall, non-barrier, one loop per joist space, etc. etc.

Air space below an aluminum emission plate is unnecessary but R19 insulation is.

My own house has 3/8" PEX on 8" centers, 3/4" pine sub-floor, 3/4" maple finish and required the addition of Europanels to overcome the coldest days.

1/2" barrier tube on 8" centers is the standard for sub-floor radiant floors and much (scientific) information is available for their proper application and potential output.

For 7/8" tubing output (no advantage) you will have to back to the vender.

Thanks for the reply MA.  Do you mind a few questions so I can understand better?  If not, here goes...

Are the Europanels the Runtal wall panel type devices?  How much footage of these are you running? 

What is your max temp water in floor.  What Outside temp do you start needed the Europanels?

Do you know a site where I might find a BTU/hr comparison of under floor systems with different tube spacing?

Finally, put yourself in my situation, if you had only one run of pex per joist bay instead of two, how much do you think the heat output of the floor would be degraded?  You could say a % of BTUs, or something like "it would be like I had 10 deg colder supply water", or I'd need to start the panels when it was 10 deg warmer outside or it makes no difference once the floor is up to temp.

Thanks for any help!

 

I don't mind, and don't blame you for being misled. The bone I pick is with those who should know better.

The answer starts with a room by room heat loss. With this, you can match the radiation to the application. As for your heat loss, it is not for lack of flow that over-size piping fails, but rather the lack of even temperature distribution.

Not having enough information to design or redesign your system, I can only say that sub-floor systems in climates like yours and mine here in MN are tough to make work at low temperatures (mine operates at 180°F during design 'cold' conditions).

A Euro-panel is any of many steel wall-hung panels (Runtal, Myson, Thermotech, DiaNorm) that typically radiate AND convect (move air) giving you better response time and lowering your design water temperature. I use 1/2" barrier PEX to serve mine, but if you have a non-barrier tube, it would not be wise to tie a Europanel to the system. The oxygen/rust can eat through such a panel in one heating system (faster if you employ the silly "open" systems).

More good reasons for closed, barrier tube systems.

I hope this helps.

You can see some of these panels on my website if the moderate will allow it ;).

www.badgerboilerservice.com

OK, my system, at the current 16" spacing will provide more than enough BTUs at 180 deg water temp too! 

I'm trying to do it with 120 deg water.  It did work fine with the old boiler providing 150 deg water as designed.  It actually does pretty well at 120 deg and 16" spacing.  I'm now running it at 120 deg and lower, to simulate how it will perform when I switch to a geothermal water to water heat pump.

However, my main concern is to understand the likely effect of reduced tube spacing in this type of application - "stapled up" under subfloor and 3/4 maple floor on typical btu/hr/sq ft at a given supply temp, 120 deg in my case.  16" spacing vs. 8" spacing.

Would some manufacturers show such typical values at different tube spacings in tables?

It is more than likely that your output would be the same with 1/2" tube on 16" centers (perhaps requiring shorter loops) which is to say if you have good contact tube to panel and panel to floor over the majority of the joist space you are likely maxed out for performance. At 120F you will also gain little by insulating more.

Add rads or plan for backup (common with heatpumps) and domestic hot water will need backup also.

Good luck.

thanks for your input MA!

OK, My 2 cents,
Badgers right about heat loss, I mentioned the other day in another post about designs we see that are on 18 inch centers (16) with long 500' loop length, and for th most part these are the ones that fall of on cold days. 1 1/2 overall floor thickness is = to R 1.5+ more or less just for starters. When designing for cold climates like where you both live we always will specify heat transfer plates on an 8 inch pattern with 330' loop max. I have seen some debate on this site concerning the thin .019 or .042 gauge VS the thicker aluminum extrusion plates, not sure where the cost effective math will come in to play, another subject. The concern I have about 16-18 inch pattern is there is less BTU capacity per sq ft and uneven heat distribution. With a Geo thermal system it is reasonable to use an electric H20 tank an the buffer tank and set the stat on the tank below the delivered temp of the compressor, causing the tank to only go on when the heat pump can not keep up.
My advice in absence of proper heat loss, think 1/2 barrier pex, heat transfer plates, 8 inch pattern, lots of insulation ideally with foil face up.
Dan

Thanks Dan, I'm with you on most of that.

The question for me to answer is how can I estimate the difference in btu/sq ft expected with my existing 16" spacing vs. the 8" spacing I could go for by doubling the amount of tubes.

Take any product, my existing is 7/8" pex, but if you can point me to a table for a specific 1/2" or whatever pex where there is a table that shows expected BTU/sq ft with 8" spacing vs. 16" spacing, (all at 120 deg Supply temp) that would be a good start.

 

Probably the only way to accurately understand the difference is to place your building in to a proper heat loss calculator. A computer model that looks at insulation, windows, exterior walls, geographic location.
Design for 18 inch 7/8 pipe and design for 8 inch 1/2 pipe. you then will see delivered water temp recommendations, flow calculations leading to pump requirements for both designs.
There is no basic chart that I am aware of.
Dan

There is no table because no one installs, recommends or specifies 16" o.c. joist systems except a couple of companies out there. If they have charts I would love to see them.

If I had to throw a dart, I would expect a quite significant increase (20 to 30% or so). But not enough for a geothermal heat pump (and 120 is too high for a geothermal system to operate efficiently or with a long life either.

With no load calc and without doing serious math, I would say your options are what you have with supplemental heat or a do-over with double spacing and extruded plates... IF you are lucky. I strongly doubt you will get where you want to go with lightweight plates.

thanks for the input Rob, I was told that the rigid plates would produce about 4% more heat transfer than the lighter weight ones I'm using, so I understand it to be a cost/benefit question at that level.  Do you understand it to be similar?  Emission reduction should be similar with either type of plate.

I'm hoping your right re ballpark for improvement by going to 8" spacing.  I don't need to hit 100% heating at below 0 outside, just want to get closer!  Will probably use an AHU for supplemental heat at lower OA temps so 100% geo is still possible!

heavy gauge plates drop water temperatures from zero to 40 degrees or more, depending on the heat load and floor coverings you start with. Typically 20 to 30 degrees.

if you are using an AHU for supplemental though, you might be able to skip plates altogether, depending on how much you're willing to lean on the AHU, what it can put out, what your load is, etc. Plates would obviously bring more of the load to the floor.

So I would get the same heat I would get with my plates at 140+ using heavier plates at 120? That would solve all my problems if correct!

Might have to do a test row and check floor temp differences

I don't know. Often that is the case, for double runs of tubing in the joist bay.

For a single run, I'm not sure, and they don't make these plates in your tubing size.

A real heat load calculation would give a pretty good indication of what you could expect.

Ok, so I am curious where I can find the math that supports a 40 degree difference on using the Extrusion VS a 019 or .024 gauge RHT heat transfer plate that snap fits tightly around the pipe. That just seems like an extreme difference in performance.
How is that supported, is there a performance curve based on different entering H20 temp? Pipe length? Diameter?
I support the reality that 16 inch spacing is a pipe dream, but would like to see the data supporting the difference in Heat transfer plates.
Dan

Using industry accepted output charts by Uponor/wirsbo, 8" o.c. installation and downward insulation to "negligable" downward loss, standard 3/4" subfloor.

30 BTUs/sq ft under tile, 65 degree room temp, 160 AWT with lightweight plates and 120 AWT with heavy gauge plates.

The difference in performance is extreme. Though .024 plates are heavier than most lightweights in the .011 to .019 range and probably narrow up that difference a bit, it's still half the thickness of the thinnest heavy gauge plate, with inferior contact to boot do to reduced rigidity and compression in the tubing groove. Also as loads drop, the difference drops, but a 20 to 30 degree maximum water temperature requirement drop is quite routine.

Frankly, in underfloor applications these days my recommendation is either heavy plate or no plates, but lightweights are kind of a waste of money in a joist application. In sandwich, wall or ceiling applications they seem to make better sense.

Interesting...as we said earlier, if my 120 deg water will now act like 160 deg water used to - which was REAL GOOD! - with the heavy plates, it might be worth looking at this option -

Leave the first loop of 7/8 as is on 16 inch centers, then put in a second loop of the highest output tubing that will work with good heavy plates, again on 16" centers, giving me 1 tube of some type every 8"  

I made a quick troll through uponor site, wasn't able to find much on this topic.  Can you provide a link or post those tables you referenced?
thanks!

 

they are in Uponor's CDAM.

However, again, the water temperature drop is for a very different configuration than you have. And you don't know what your heat load is (there is no way you were doing 30 BTUs/sq ft average at 16" o.c., so a 40 degree drop is not realistic for you).

the idea that tubing is "high output" is pretty silly though. Pretty much, PEX is PEX. "Thin wall" or not makes very little difference.

You're frankenstein-ing your system with no idea of what your load conditions are. I suggest you back up, and do step one that should have been done before a foot of pipe was installed in this house. Then you can get "on the charts" and see what you should be able to expect from other configurations, and compare that to what you are seeing now.

I will reiterate that 120 is too hot for most geothermal systems to operate efficiently or with long life as well.

Finally, you would be very likely to have noticeable heat striping to just pair up heavy plates with light plates. If you're ok with that, great, but I wouldn't be happy with it.

Alright, a heat loss calculation has been done for my house. At the design temp of -7 F, the radiant floor to fully heat it's space is somewhere around 30K - 35K over about 1000 sq ft of floor. I clearly need more heat than I can get now at 120 degs. - I think the space will be able to use all the heat that can be produced from that temp water in my staple up system, if I have too much heat, we just lower the temp.

So, the issue I'm trying to resolve is how to maximize the amount of heat transfered to the floor wiith that 120 deg water.
I think these are the realistic options we have come up with:

1. Leave as is
2. Double up on tubing, use more thin plates
3. Double up on tubing, use heavier plates on new tubing if cost can be justified by increased heat transfer

Striping will not be a problem, as you have to realize it can't be any worse than what is there now, tubes on 16" centers with some dead space in between. If we got "Bad" striping because we got so much more heat from the new tubes/plates vs. the old ones, I would be Very Happy!!