I’d like a consult on whether I need to explore a potential problem further. This is my own home, and I found what appears to be some duct leakage within our floor trusses. I don’t want to have to start cutting through the sheet rock from below if I don’t need to.
Note: You’ll see what might be an associated problem with a void of insulation in a large section of the floor truss band…perhaps accentuated due to the duct leakage which is close to this area.
Our hvac system is propane. Does this look to be a convective leak, or would a propane system’s trunk line (which is most likely where this is) produce this much heat conductively? Also note that there is evidence of increased heat both above & below the same problem area.
Thanks in advance for your insight & recommendations.
You’re right, and I failed to mention the results of this test. I was only able to get to 12 Pa with 1200 cfm flow, which TecBlast normalized to 1864 cfm @ 25 Pa (351 sq. in. of leakage).
Notes: I pressurized system to max 12 Pa through a large return, and sealed all registers & other returns with plastic & tape. One concern is that some air may have leaked through T&G of the pine flooring where some registered were located…though I sealed the ones that looked most likely to leak.
Thank you all for your responses. Jason, the IR pics were not taken during pressurization with the duct blaster, but during normal operation of the HVAC. No other duct blaster test was done other than the simple one mentioned.
Our HVAC guy, who admits not knowing about duct blaster, and maybe just a bit more about IR, said that the main trunk of this propane system would produce enough heat to cause what we were seeing in the thermograms. He reports that the air running through the R-6 insulated trunk & ducts is around 160-170 degrees. I don’t know if he’s just being defensive or if he has a point. It seems, however, that unless the trunk is actually in contact with the upper & lower surfaces that there should not be any conductive heat transfer…so it would have to be convective…meaning there is a leak. Please advise me whether this logic is sound…and thanks again for your responses.
Rick, do you have a digital picture of the duct work? If they are leaking air that bad and are installed up inside the floor joists I would recommend a section be taken down, insulated, re-installed and tested again to see if that corrects the problem. Dave is correct that it’s still within the conditioned space but my thought is it’s not performing as designed or intended. If the test area proves successful I would remove all the duct work, insulate and re-install.
Of course I’m looking at it in my perspective which in my area a leak that bad would have a huge effect on heating efficiency.
Keep us posted.
Unfortunately, I don’t have a digital pic…worse is that there’s already sheetrock installed over the floor trusses on the ceiling of the basement floor.
I do understand about this being within the conditioned space, though in reality until we fix the missing insulation in the joist/truss band it’s still connected to the outside. I’m also concerned about condensation running the A/C, as several of you have mentioned.
Does anyone know of a less than invasive diagnostic step…other than cutting through the sheetrock…that could facilitate an accurate diagnosis? Some kind of advanced duct blaster zonal test or something along this line. Please note that my experience with duct blaster is very limited…so there may be relatively simple tests that I’m not even aware of that may help.
I would consult with an HVAC guy first. But from my knowledge there are a few different tests than can be done. You could do a simple velocity test before and after the affected area. This is semi invasive, normally a few 5/8" holes are drilled and a few readings are taken across the duct. They would do that before and after the area in concern then take the difference to find the leakage.
Another would be to use a flow hood (balometer), which would be non invasive, but potentially less accurate. Depending how close you have ducts to either side of the area, it would be less accurate the further the ducts are. If there is a branch off either side, that test would be even less accurate.
Lastly, you could do a complete blower door test with a duct blaster used in conjunction. This is a true duct leakage to the outdoors. This test would not isolate the issue, just give you exact duct leakage throughout the entire system.
Even if it was in direct contact it does not explain the heat signatures flowing off to the sides in your images. I would be shocked if there was not a leak.
Personally I would do the flow hood test to confirm the leakage in the area. Then the velocity test to truly isolate it. If you plan on fixing this issue the invasive nature of the velocity test is a mute point.
(Jim Seffrin, Director of Infrared Training)
Based upon your imagery and descriptions of the subject areas I would like to offer the following.
During operation of the heating system, floor surfaces above heating system ductwork are likely to appear warmer than floor areas that do not contain ductwork. The same will be true for drywall ceiling surfaces immediately below the same ductwork.
In warmer months, it is expected that ductwork will cause these same areas to appear cool; however, they may not show as readily as the Delta T associated with an A/C system is lower than a Delta T associated with a heating system.
From my experience, I would expect that the pathway of the duct should produce a relatively uniform thermal pattern along its length without any pronounced hot/cold spots.
Discrete hot spots like the ones shown in your imagery are often caused by a duct leak(s) in the area of the hotspot. They may also be caused by missing/damaged insulation in the cavity or portions of the duct being in close proximity to the imaged surface.
Because the only way to positively identify the cause is a visual inspection, I would suggest that you visually inspect suspect areas by opening the drywall ceiling below. Since the thermal patterns may be the result of a duct leak, I would investigate the cause prior to the cooling season in order to prevent potential condensation problems that could occur.
I hope this helps. Please let us know what you find.
If you wanted to get really crafty, and it is only an idea on my part I am not sure how well it would work in a real world application. You could introduce smoke with one of these (http://www.aikencolon.com/Retrotec-AC107-AC-107-Air-Current-Tester_p_2348.html) to the system near the boroscope entry hole, downward from the air flow direction, and block any exiting ducts beyond that point. In theory the smoke should be forced out through leaks in that area, which you could see on the boroscope.
Rick, I’m sorry I couldn’t follow up with your question. I had to make a road trip today to visit an old Army buddy who just had a bunch of cancer removed.
Close the registers 90% (or more) throughout the system. If you have normal registers they will leak substantially even when completely closed.
Put your camera on a tripod and let it run.
Turn on the heat and get a good thermal anomaly visible across the floor.
Take scans at specific intervals (depending on how fast it heats up).
Switchover to air-conditioning and watch the thermal anomaly at the location of the hottest spot on the floor (continue to take scans at the same interval) . If you begin to get a cool thermal anomaly occurring in the midst of the hotter spots, you have a direct air leak.
If it is conducted heat loss, the temperature transition will be even and proportionate to what you saw in the heating mode as things warmed up. Do not leave your camera on automatic during this test. But do adjust your span for the upper levels of the anomaly so you can see what’s going on.
Aother thing you can do is close all of the registers 100%.
Make a small hole in the floor or ceiling (preferably the ceiling as you have hardwood floors). Or from the outside where your furnace vents outdoors.
Hook up your micro manometer from your blower door to a piece of tubing stuck in the hole you made.
Turn the circulating fan to the on position at the thermostat and measure the difference in pressure.
A quick but sincere thanks to all who’ve taken a few moments to respond to my post. It’s been very helpful & I have some direction for further assessment. Will be very interesting to troubleshoot this with these methods…I’ll document and post for the record and future reference. Thanks again fellas!