maybe i am looking to hard but......
i have 16 inch center joist staple up applicaiton. (see other posts by me)
i need to insulate under this as it is over a crawl space.
i read for several hours regrding what makes sense to do. it seems:

1) tinfoil is addaquate to reflect the "radiant energy"
2) the radiant reflector (mentioned above) will get warm due to the convectional heat tranfer so it needs to be insulated

DOES THAT MEAN TINFOIL AND A GOOD R20 BAT IS THE BEST SOLUTION?

3) the bubble layer does squat. it is only there for convetional heat transfer and is a very low "R" so pointless

4) it seems they all have this multi layer approach??? why if no measurable "r" value comes with this layering of...

5) what is recomended from this site readers?


thanks

At lower temps the relative value of a reflective layer goes down a bit, and in many/most instances it adds significant labor costs, and many be counterproductive if installed sub-optimally. Reflective bubble pack adds another R1 or so, in addition to it's radiant reflectivity. Best value is to go with batts- high-density "cathedral ceiling" bats give somewhat better performance, but if I read your posts correctly, you're only running ~90-100F water(?) max, in which case you may even be able to skimp a bit on the insulation (many staple-ups see 150F-160F+ water on design-day, more 2x the delta-T you're likely to see over a well-sealed but unconditioned crawl space, more than you'd expect to see over a conditioned-heated basement.)

If the crawl space is leaky enough that it is expected to sometimes get well below freezing once there is insulation in the floor, go with high-density batts- (R30+ if you have the space- you probably do). Snug 'em up against the PEX & plates but don't over-compress them. But if the crawl space has insulated walls and coasts along above 50F, R15 is probably adequate.

Reflective radiant barriers don't affect R value much but they DO block radiated heat. R-value is only concerned with conducted heat transfer. Low density fiberglass batting is somewhat transparent to radiant heat, which is how the practice of applying radiant barriers under staple-ups began- it reduces the effective heat transfer considerably when used in conjunction with thinner, less dense batting. The emissivity of radiant barriers is quite low- they heat up because they can't/don't radiate away the tiny fraction of heat that they absorb. But they can conduct & convect heat away where there is air contact to the cold side of the barrier. Reflective bubble-pack DOES outperform tinfoil in this regard, but if you're insulating under it with batts anyway, the performance difference will be negligible. Where there's sufficient space, it's less labor-intensive to just go with thicker or denser batting to meet or exceed the performance of cheap-batts + radiant barrier.

exactly what i am hearing from other sources. amazing what they will try to sell on line......
i plan to run lower temperatures (GEO unit with max 127 but for efficiency reasons the lower the better) currently with no insulation my house is warm at 108 degree water.

with the Bat insulation i will realize
1) less flow required to maintian temperature
2) lower temperature water
and less running time on my unit


Now,

my application is staple up, plates installed, doubel runs per joist, goal of 100 degree water.


before i scrap the radiant barrier aand install R20 batt insulation without a radiant barrier....

any other comments??????

please!
i want to go ice fisshing for the weekend and my wife MUST stay warm without the woodstove
i can do the "bat" thing tonight.......i

With 100F water, R20 bats snugged up against the PEX & plates radiant barrier offer little to no benefit. If she's cold with that setup, she'd be cold with tinfoil & R20 batts too.

Right on Dana.

I INSULATED THE JOISTS WITH R20 LAST NIGHT...GOING FISHING SPECKLED TROUT!!!!! MAMMA HAPPY!

BUT,
Sorry for being stubborn but...

the tubes are hot. they radiate heat, the plates absorb this radiant heat and "transfer" this heat to a larger area of the floor as the plate warms up creating a higher BTU transfer to the floor. this makes sense.

but, the bat insulation snuggled up to the plate stops the radiant heat transfer down?????

why all the hype about a radiant barrier????
what am i missing?????

so, how then do i insulate the pipes that run to the different loops and zones?

a bat wrap or a radiant wrap???????

Because it sells well; You've got radiant heat, you MUST need a radiant barrier, right?? And, there is some merit to reflecting, not just stopping radiant energy. At least for some period of time. That is why I do not specify it: I am not sure how long it remains effective, or how effective it remains, and I would rather spend the money on more R-value.

You insulate pipes in cold space with foam pipe insulation.

If the runs are short you might not bother- the total exposed surface area is small, and the temperatures are low (compared to 180F water in some systems.) If you're committed to it, cheap ~6mm thick closed-cell foam pipe insulation adds about R1.5-R2. If the runs go outside or in sub-freezing crawlspace, R6-R8 fiberglass pipe-insulation is called for (and most already have a foil layer already integrated into the outer covering as radiant barrier.)

The hype about radiant barrier tends to be overblown. It's effectiveness increases somewhat with the temperature of the water, but the large fraction of the heat loss in an exposed radiant staple up would still be conduction to the air inducting convection & air currents. Radiant barriers also stop the air currents, so they reduce this loss too, but high-density batts do a reasonable job of slowing air flow, and are fairly opaque to the radiated heat compared to low-density or thinner fiberglass batts. In spaces where you can't get R20 batts radiant barriers can be a significant performance enhancer, but doing it right is fairly labor-intensive, and if NOT done right it's nearly worthless. The fact that you need to leave an air gap gives rise to potential air movements through that cavity- it has to be fairly well air sealed to perform anything like the hype indicates.

[edited to add: I see Rob beat me to it. :-) ]

It is amazing the information out there how can they all be correct???????
data i could find:

thick plate pay back is not there - sorry NRT.Rob. do you have data or information other than your opinion regarding the results of using heavy plates?

radiant barrier: holly S@#!!!! so many different opinions....seems though the ones with data indicate with plates radiant barrier is a waste of time if good R20 bat insulation is used

just amazing.....like politics i guess

You're using Geo. Ask Geo guys. As I understand it, first of all, you have a maximum water temperature of Geothermal of.. practically speaking.. about 110 to 115. So if you are over that, it's not a question of "payback", it's a question of "does my system work".

If you are already under that, then a lot depends on what your actual water temperature requirements are. If you are already running 108 with no insulation, then heavy plates probably won't help you too much and you should find insulation a significant benefit regarding your water temps.

But I find it pretty hard to believe 108 degree water, with no downward insulation, was adequately heating your house with 16" on center light plates.

if it was, then you have a ridiculously low heat load, and geothermal was probably a waste of money in the first place, cause you couldn't much much over 7 or 8 BTUs/sq ft under design conditions.

thanks for responding

with the pad (16 by 24 two 200 foot loops with a mixing valve set at 85 degrees as per information i had read). (this area is two floors with nothing in second floor which is same level as rest of house)

rest of area (20 by 36 with only 75 % pexed,  three 300 foot runs, light weight plates, hardwood / tile combination floor and r20 bats under

temperature of water at 108 house was warm machine running at about 60% duty cycle

as i work at it and experimant i slowed the flow through the pad and the house got really cold!!!!!!!!!!!!! the pad is not well insulated underneith according to what i read. i did use the white backed bubble style radiant barrier as per recomendation but learning now i should have done more.....

the floor in the living room (where it is not pexed) is very cold...... the crawl space where this all lives is very warm currently

i am hoping to add the other 25 % pex tonight more insulation and plates

i do believe my heat load is low as the house is small and all re insulated new low e windows etc.

i did not have any heating other than wood. as i am a DIY the initial extra outlay of cash is being almost comepnsated for by a government grant ($7000).

the lake loop was not far from front door and the lake is deep...

thanks for taking the time to respond. i realize as  DIY i have been in this over my head but as i am a stuborn engineer i will plod on

 

What does "house was warm" mean, and what is your heat load (or, how did you insulate and where are you)?

Again, I find it pretty hard to believe that 108 degree water, in a joist, with no downward insulation, was heating much of anything. Under very good conditions the 7 to 8 BTUs/sq ft upward would be expected. With no downward insulation I would ballpark your output at half that into the room above.

If the pad were doing all of the heavy lifting though, I could see it, maybe that's what is happening, but that's asking a lot unless that pad has no finish floor, perhaps; if that were the case you could be doing quite a bit of output over there (15 to 20 BTUs/sq ft) at 85 degree water. It would also be noticeable chillier in the joist heated area, I would expect, as you moved further from the concrete area. and a warm crawlspace in this case is a sign of wasted heat for sure.

i think the pad was contributing to the "house was warm" it was not hot, just barily comfortable. to add to this the not done yet area is at the other end of house and you could feel it getting colder as you moved into that area. the pad has no finish at all on it other than a coat of paint. no carpets at all.
so, yes i think your correct as it does get coller the further from the pad....
i am in Sault Ste. Marie, Ontario Canada. (on the north east side of Lake Superior)
the crawl spce heat is obviously from my pipe work as i do not have anything other than the pexed plated pipes insulated. it is on my list for sure
i hope that the added 25 % pex with plates will allow me to reduce the flow through the pad as it is the den and not occupied all the time.
it was cold today with the pad flow reduced (obviously to much)

it is all starting to make some sense ( i guess)

I tend to believe people who actually measure stuff...

This experiment clearly demonstrated the benefit  of plates (extruded heavy-duty type that Rob likes) vs. none:

http://www.ngml.ksu.edu/research/project_archive/project_archive_publications/ashrae/1036_final_report.pdf

It rougly doubles the heat flux into the room at a fixed water temperature, meaning it can handle roughly twice the heat load.  Alternatively, it means that you can run lower the water temperature relative to a no-plate staple up to deliver the same amount of heat into the room.  With all boilers (but particularly condensing boilers) the combustion efficiency goes up as the temperature of the water returning to the boiler is lower, so delivering a higher heat flux to the room has a net benefit to the system efficiency, independent of whether the boiler's output temp is controlled to run lower at lower outdoor temps (as is commonly done these days).

There's some discussion & testing about air gap between the tubing & insulation too.  (More air space lowers the effective heat flux into the room slightly.)  They didn't test radiant barrier either (the XPS used in the experiment is relatively opaque to infra-red radiation, unlike fiberglass) But they acknowledged (without data) that snugged-up fiberglass would likely perform more like the XPS with the greater air gap due to induced convection within the insulation. 

With denser & deeper fiberglass the impact of the performance loss becomes less consequential and the benefits of radiant barrier become lower.  With standard density batts, R19 without radiant barrier performs more like R12 when the hot side more 75 degrees above the cold side. Most of the study of radiant barriers has been in attics with relatively high inner roof-deck temps (150F+) but this (now classic) experiment was with standard R19 batting, at temps & delta-Ts not dissimilar to low-temp radiant floors:

http://www.fsec.ucf.edu/en/publications/pdf/FSEC-DN-6-86.pdf

It also demonstrated that there is about an "apparent R value" of between R5-R6 to the radiant barrier even when placed on the cold side (room interior side) of the insulation.

It would be useful if someone in the radiant floor biz did similar experiments to determine the actual value.  It may very WELL be worth stapling cheap R.B. to the joists below the fiberglass insulation before adding the gypsum! (And far easier than installing it on the warm side with the requisite air-gap between the R.B. and the hot tubing, if somewhat less effective there.)

I didn't find it again in a quick google search, but there was an experiment out there showing R13 fiberglass with R.B. on the hot side performing similar to unfaced R19 batts in a radiant floor.  This is consistent with other attic test data that show even the "apparent" R-value of R38 standard-density unfaced fiberglass loses somewhat more than half it's rated R-value in ceilings when presented with 160-180F interor attic roof decks.  Other tests have shown that fiberglass in cold-side-up installations loses R-value with greater delta-T more rapidly than in warm side up configurations like radiant floors, but that in either configuration it loses R value with delta T (presumably from both convective & radiant losses.)  Fiberglass not the greatest stuff from a performance point of view, but it's relatively cheap per unit R-value, so where there's space to make it thicker, it's still often the best value.  (And warm side up enclosed configurations are where it performs the best.)

Florida State, University of Texas, oak Ridge Nat't Labs, and Lawrence Berkeley Nat'l Labs all have volumes of study on insulation types, configurations and comparative performance data.  You can spend years looking it all up, if so inclined. :-)

seems consistant with what the latest data is indicating.
meaning
plates increase BTU transfer therefore lower water temperature
insulating below the plates is needed and once plates are installed air gap only hurts
radiant barrier is R5 so if i wanted another R5 i would do the radiant barrier

i do not like the idea of radiant barrier under the insulation. not sure why just seems odd

NRT.Rob....I had reduced the flow thru my pad to the point where the house got cold. i increased it a bit and the house warmed up. while doing this i did nothing to my joist flow or temperature.
now, i hope that with 25% more joist pex i can do the same and not have the house go cold

here is a question:
i cuirrently have the three loops for the joist open 100% on a 007 taco pump. each loop is about 300 feet. flow rate? i have no control currently just starting and stopping the pump.
i read about zone control valves. wh not just turn the pump on and off?

is too much flow a bad thing?

flow control valves are useful when attempting to balance output up or down between loops in the same zone.. say I have a bedroom and a bathroom on a zone, I might crank flow to the bedroom down to keep it cooler and up to the bathroom to keep it warmer.

also if you have loops of different lengths you might need balancing to avoid hogging flow in shorter loops.

zone control valves (on/off) are only useful if you actually want separate zones (thermostats) and you would use them instead of multiple pumps to save energy, for one.

exactly as i thought...thanks
i need to purchase a temperature probe gun to check but by hand the loops seem very close in temperature.
when i add my lasy loop it will be a bit shorter so might need to throttle it a bit. i did put isolation valves on each loop.
can a 007 taco run 24/7 - at 108 degrees what is the life ?

perhaps longer than you ;)