# Thread: Very Technical Duct Insulation Question

1. Regular Guest
Join Date
Jun 2005
Posts
67
Post Likes
Carnak:
You may be right. I was going off a dew point table with air temp on one axis and % relative humidity on the other. Dew point where they cross.

here's the link, in case you want to look: http://www.lamtec.com/dew-point-calculator.htm
It shows both Fahrenheit and Centigrade.

Still hoping to see the formula posted.

2. naima.org has a calculator for pipe insulation that'll do the same thing.

3. Regular Guest
Join Date
Jun 2005
Posts
67
Post Likes
Carnak:
Maybe you are right about maximum dew points, but....

I was up in the attic this morning and took the following readings (all fahrenheit):
dry bulb 90.2
wet bulb 79.5
according to my calc this yields 76 deg dew point and 63.3 relative humidity.

This with attic fan off and commercial dehumidifier running constantly.

I sliced into the side of the fiberglass insulation wrap (supposedly R6 but not more than an inch thick) on a 16" round supply duct and the outside skin of the pipe was wet top to bottom. Apparently the moisture runs down the side and either drips out the bottom or causes max condensation on the bottom...not sure which. If the moisture was sealed in there, though, I would expect the condensation to be rising with the water level, which it does not....

Anyone come up with that formula to calculate how much insulation is neeed to prevent condensation yet? Thanks.

4. Professional Member
Join Date
Nov 2004
Posts
2,928
Post Likes
Originally posted by beachwalker
Carnak:
Anyone come up with that formula to calculate how much insulation is neeed to prevent condensation yet? Thanks.
Hey beachcomber,the answer to most of us is "whatever the code will allow"...but you want better ,and therefore I will do my best.
I am not a math wiz ,but with this formula:

heat leakage=area x U value x temperture difference

you should be able to bend it around,plug in different U value and find an answer.
(u-value is the inverse of r-value:
r-value 6 is 1/6 or 0.166.)

5. Regular Guest
Join Date
Jun 2005
Posts
67
Post Likes
Jacob:
Thanks very much, but I am unfamiliar with the term "heat leakage". I was expecting a formula with parameters like:

dew point or relative humidity
ambient temperature
duct temperature
some sort of a constant (maybe) for metal duct

seems to me there must be a relationship in there somewhere that can be solved for an R value. Can you help me out?

Ablib said:

Think of it as a slow water leak in your attic. Can you think of any drawbacks to having someting drip every day for months at a time?

In Texas based on what I have read about mold calamities, most of them have been traceable to an undetected water leak. Just one example. If you read the writings from Joe Lstiburek he will never recommend any practice which allows leakage inside a house. I would recommend paying high respect to anything from Lstiburek or his company, Building Sciences Corporation.

Regards -- P.Student

[Edited by perpetual_student on 07-05-2005 at 06:18 PM]

7. Professional Member
Join Date
Nov 2004
Posts
2,928
Post Likes
The way I would solve the problem is to simply fix the leaks in the insulaton or try one with a greater r value.
That doesnt mean my way is the best ,though.
I will get out the books and try to figure out what you are looking for.Its a good question.

"heat leakage" means heat flow.The heat from the air surrounding your ducts is "leaking" past the insulation and flowing to the metal,thereby causing condensation.

8. Well you have a multiple part problem beachwalker.

First off, you said these were variable speed, so they will run colder than 55F at times, assume 50F and take that as the outside duct wall temperature.

You are power venting the attic which I think is a bad idea because it just sucks humidity into the place.

So your heat transfer will be driven by the difference in the attic air temperature and the ductwall temperature assuming the foil back vapour barrier more or less reflects away most of the radiated heat.

So you can simplify your problem into two resistances in series.

Heat has to transfer from the air to the surface of the insulation , so lets just assume you have one mother of a power vent fan and treat this resistance like it was a blowing wind and give it a low and conservative value of 0.17 Btu/(hr ft^2 degF). Remember this is per square foot.

R convection = 0.17 Btu/(hr ft^2 deg F)

If this was a recatangular duct the rest of the problem would be much simpler. If you believe me skip everything that follows except the very last parargraph, else continue on.

Get the R-value of an insulated round duct, it is radial heat flow.

R insulation = ln (ri/rd)/(2 x pi x k x L)

ri would be the radius of the insulation and rd would be the radius of the duct. For example your 16 inch duct could be insulated with 2 inch duct wrap, 1.5 pounds per cubic foot density, that compressess by a maximum of 25% when properly installed. This formula works in the units of feet but since you are dividing you could still use ri=9.5 and rd=8, for your 16 inch diameter duct with a 2" insulation wrapped around at 1.5" thick.

You should be able to get the k values from a manufacturer, this is the thermal conductivity of the insulation.

2 x pi x L is part of the values used for the surface areas, however when they do the calculus to derive the formula, the 'r' term for radius has been manipulated out of this. The radial flow comes in to play as the surface area changes as the heat flows. Perhaps you can set L to being equal to a 1 foot of duct length.

When you use the convective R value of 0.17 remember to adjust it for the surface area of one foot of insulated duct. You would use the one foot length and the outside radius/diamter of the insulation to get this. 2 x pi x ri x L.

You should eventually be able to come up with two resistances in series and be able to add them up to get a total resistance.

total resistance = R insulation + R convection

Once you have the combined resistances, the heat tansfer per foot of duct would be equal to the difference of the attic and duct wall temperatures divided by the total resistance.

Btu/hr = temperature difference / total resistance

You can then isolate the convective resitance and or the insulation resistance to come up with the surface temperature of the insualtion, and hopefully that temperature will exceed the dewpoint of the ambient air.

Btu/hr x individual resistance = temperature difference.

The Btu flow through the convective air resitance and the insualtion are the same. Working from the convective air side, perhaps the surface temperature of the insulation would have a temperature differential of 2 degrres less than the attic air temp, therefore the surface would be attic temp minus 2.

If the surface temp is too low, like below the dewpoint you are shooting for, then you need a thicker insulation.

OR

You can make sure that your ducts DO NOT LEAK, before you reinsulate them, and then have a 2 inch layer of ductwrap with an "INSTALLED" R Value of about 6.5 properly applied. Make sure the foil vapour barrier is intact so that humid air does not short circuit through the insulation and contact the cold ducts.

9. Professional Member
Join Date
Nov 2004
Posts
2,928
Post Likes
Originally posted by Carnak
If you believe me skip everything that follows except the very last parargraph, else continue on.

Thanks Carnak.
I believe...I believe...I believe!

10. Originally posted by ralphtheplumber
naima.org has a calculator for pipe insulation that'll do the same thing.
yes ralph they sure do, good suggestion

11. Regular Guest
Join Date
Jun 2005
Posts
67
Post Likes
Carnak:
That link was right on point. Thanks...

Factory engineer was here today and their solution was to increase the cfm's to 1500 and jumper y1-y2 in air handler, so effectively, the variable speed fan is bypassed... so it runs at 1500 cfm for stage 1 and stage 2 compressor. This lowered the differential to about 12 degrees across the coil in an attempt to raise the duct temp and stop the condensation.

According to their meter, relative humidity in the attic was 100 % this morning and temp was 101.

I anticipate this change will allow sensible heat (air temp)reduction, but latent (humidity control) will suffer: humidity control now shut off. After running all day, temp is at 74 deg setpoint, but tstat indicates rel humid at 71%, which is up from where it had been at 65%. When humid control was on, it was between 50 and 55 percent.

They also suggested foaming the roof and sealing the soffits, then bringing in controlled amount of outside air for ventilation. Even maybe dehumidifying the attic. This in hopes of lowering humidity and temperature in the attic.

Any thoughts on these ideas?

12. Originally posted by beachwalker
Carnak:
That link was right on point. Thanks...

Factory engineer was here today and their solution was to increase the cfm's to 1500 and jumper y1-y2 in air handler, so effectively, the variable speed fan is bypassed... so it runs at 1500 cfm for stage 1 and stage 2 compressor. This lowered the differential to about 12 degrees across the coil in an attempt to raise the duct temp and stop the condensation.

According to their meter, relative humidity in the attic was 100 % this morning and temp was 101.

I anticipate this change will allow sensible heat (air temp)reduction, but latent (humidity control) will suffer: humidity control now shut off. After running all day, temp is at 74 deg setpoint, but tstat indicates rel humid at 71%, which is up from where it had been at 65%. When humid control was on, it was between 50 and 55 percent.

They also suggested foaming the roof and sealing the soffits, then bringing in controlled amount of outside air for ventilation. Even maybe dehumidifying the attic. This in hopes of lowering humidity and temperature in the attic.

Any thoughts on these ideas?
101F and 100% no way. They need a new meter, a new battery or more time for it to stablalize.

The high speed will mean more sensible cooling, do you have a 3.5 or 4 ton system?

Are you running the air handler fan all the time, that will raise RH.

Still suggest you seal your duct work, and then when you re-insulate, ensure the vapour barrier is intact all the joints are taped.

Is the air handler sweating or just the ducts. If the temp and humidity readings were close to being accurate, that air handler and the auxiliary pan would be demonstrating the evaporation cycle. It would be raining into the pan.

Not up on all the lennox details, but I would be surprised if there was much more than 1/2 insulation in that cabinet, maybe 3/4. If that air handler is not sweating and your ducts are the problem is the duct issues already raised by me and everyone else replying to this thread.

13. Regular Guest
Join Date
Jun 2005
Posts
67
Post Likes
Carnak:

I tend to agree with you relative to their meter... my own dry & wet bulb readings with digital thermometer have yielded closer to 61% rel hum and 75 deg. dew point in the attic.

air handler is not sweating... only ducts
fan on auto, not constant
unit in attic is 36,000 btus

Not counting the flex, I have about 80 feet of 16" and 14" supply ducts. My next move is to cut off a section and put a new fiberglass wrap and see what happens. If that works, then put the cfms back to where they were and check for condensation again.

the foaming and sealing the attic is a foreign idea to me, but the factory guys claimed it is gaining popularity and solves the problem. It will be expensive to do, but as long as there is no potential bad things, like decreased shingle life, I will consider it.

Thanks for the perserverance.

Page 2 of 4 First 1234 Last

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•

## Related Forums

The place where Electrical professionals meet.