'Exhausty' smell in house when boiler runs

Ken Johnson

'Exhausty' smell in house when boiler runs --

This is a long post, but I think the details will be needed since a professional has checked the obvious and finds no explanation.

My wife reports an 'exhausty' smell in the house when the boiler runs. We have had two long cold spells (for St. Louis) recently with temperatures not above freezing for more than a week. The smell is particularly noticeable in the morning when the boiler runs for a while bringing the temperature of the house back up after the overnight thermostat set-back. It is most noticeable on those very cold (for St. Louis) days such as we had during those two cold spells. The smell is most noticeable in the second-floor hallway. It is not noticeable in the basement. I previously made an effort to close openings between the basement and first floor, and the basement door is weatherstripped. On the other hand, I have made no effort to seal the basement, which is quite leaky to the outside (and needs to be to provide combustion air to the boiler). I am less aware of the smell. Now that it is warmer, the boiler is running less and the smell is weaker.

I have asked my wife to smell the radiators. They have the normal warm radiator smell, and that is not the exhaust smell.

I tend to trust her observations because the last two times she mentioned this kind of symptom, I investigated and found that the old water heater was spilling exhaust (candle went out) on one side of the hood. A professional from the company I call when I need a pro came out and tweaked the hood until it was no longer spilling (candle does not go out). Next time there was a smell complaint the water heater was spilling again. I could not adjust the hood to fix it, and now we have a high-efficiency water heater (uses less gas than the boiler pilot light, on average) with a fan-blown PVC exhaust.

The steam boiler is the only gas appliance that has been running in the house when the smell happens.

The house was originally built about 1905. Two stories, plus a walk-up (narrow stairs) attic that contains many fiberglass insulation batts laid on the pine floor. The boiler is a Burnham, installed in 1994, before we moved in. It exhausts through an 8" Zflex stainless steel flue liner installed after we purchased the house, about 20 years ago. When the boiler is running or is just on pilot, it was drawing well when the technician tested it. The technician also looked at the areas of the house where the chimney runs up through the walls, went up to the attic to inspect the chimney there, and then used the 1/2 height door at one end of the attic to reach the second-story roof, and inspected the top of the chimney. No problem found. At that point the technician had no other suggestions for investigation.

We have a CO detector with a digital display in the basement (where the boiler is) that has not gone off. Nor does it display a non-zero value. I moved the detector to the second-floor hall for a few days -- no display, no alarm. I tested the detector with a test kit that included a plastic bag and a little glass vile of CO. The detector showed a value in the 780 ppm range at peak, which was consistent with what was described in the test kit.

The boiler has been skimmed on a regular basis. When I do that, I always save a little cup of water from the start, and from the end. The first time I skimmed it, there was a noticeable film covering most of the surface of the start sample. Last fall, there were barely discernible scattered dots of oil in the start sample.

Ideas I have:

Get someone to make a camera inspection of the flue liner. The chimney has an odd bend in the attic -- perhaps the liner was damaged during the original install and has now failed at some point.

Make an extra effort to seal the upstairs hall from air infiltration from the chimney behind the wall on one side, and the crawl space over the first floor addition on the other.

Re-skim the boiler to see if anything is present now that was not present last fall.

That's all I've got. I would love to hear suggestions from others.

Ken

Neild5

if you can get your hands on an IR camera, taking pictures of the walls that butt up to the chimney. After the boiler has been off for a while leaks would show up as a cold spot and while it is running a warm spot. IMHOit sounds like a leaky liner.

Ken Johnson

Thank you for this practical suggestion. I learned that Home Depot has a program renting FLIR brand infrared cameras. Inconveniently, only one of the closest eight stores has one -- if anyone else tries this approach, check the Home Depot rental website, not the FLIR.com web page about the program. In about 10 days the current warm spell should be over, and I will try to get one then.

BobC

I would nose around the attic because I think @Neild5 is on to something.

Bob

EBEBRATT-Ed

It sure seems to come up as a chimney liner problem to me. Don't know what else to add. I suspect the smell is worse when coming out of night setback because the chimney liner has had a chance to cool down.........less draft

Ken Johnson

One more item ruled out, I think -- I skimmed the boiler. The start sample wasn't bad, and it did not match the smell. The end sample looked darned good.

Ken Johnson

Rented a FLIR E6 camera from a Home Depot 20 miles away. That seems to be a nice piece of equipment, and I thought I got good value for the money spent. A couple of things about it: It also has a visible light sensor, and takes a simultaneous visible light photo (configurable) when you save an IR photo. Makes it easy to to identify exactly what the IR photo is showing. Also, uses edge detection in the visible photo, and superimposes (faintly) those edges in the IR photo, which also helps interpretation.

A shot from the master bedroom side of the flue seems pretty suggestive.

The start of the purple flare (warmer than the black parts) is about 4.5 feet above the floor on the second floor. My next step is to find a chimney sweep who can do video camera inspections.

In the meantime, spent some time yesterday sealing the base of the wall and other openings in the 2nd floor laundry nook, which also adjoins the chimney. In the short term this may help with smell issues. IR camera also pointed out two switch plates near the chimney that were warmer than expected and may have leaks. I will get those soon.

Thanks for the useful discussion.

Ken Johnson

Sealing the base of the wall (remove quarter-round, insert latex spray foam, replace quarter-round) and all of the other wall openings (laundry connection, gas line for dryer, electrical boxes) all around the chimney area on the second floor seems to have helped reduce the smell. That's the good news.

A video inspection of the flue performed by a chimney sweep with a 'Chim-scan' (trademark, probably) was not much help. That device is a camera and light installed on the end of a series of flexible poles. In this case, pushed up the flue from the basement. The camera is pointed at 90 degrees from the poles, with a field of view of 90-120 degrees (my estimate). Because the poles have to make a bend entering the flue, and because there is some friction between the fingers that hold the camera in the middle of the flue and the flue itself, it can take significant effort to rotate the camera to see all sides of the flue. Also, it can resist turning, then suddenly turn. This arrangement makes it more difficult for me to be sure we really got a good look at everything. I am not surprised that there is not a large, obvious hole. If this problem was going to be easy it would be fixed by now. Anyhow, the sweep's report said 'no problem found', and his suggestion for our smell problem was that maybe there was a natural gas leak that we should look into. He offered no explanation for the thermal signature shown above.

It occurred to me that perhaps it would be possible to pressure test the flue liner. I would not try to blow it up like a bicycle tire (60 lbs), but it does seem to me that it ought to hold 1 or 2 pounds. Is that crazy?

Jamie Hall

A natural gas leak has a very distinctive smell. I don't think you would confuse it with an "exhaust" smell.

A metal chimney liner might hold one or two psi, though I wouldn't be at all surprised if it didn't but was still good. I have to admit that that warm flare in your picture does suggest a heat leak. However, if the liner isn't tightly cemented into the original flue, but just slipped or dropped in, it could be a place in the original chimney which does have a leak -- and what you are seeing could be air warmed by the warm liner escaping at the old leak, rather than a leak from the liner.

BobC

I live in a 100 year old house and there was a patch job on the chimney in the kitchen where a stove mounted gas log was vented. I've seen similar patches on second floors where wood burners were removed. If your house had that sort of patch it might be a place to look.

Bob

Ken Johnson

Mr. Hall,

100% agreement re distinctiveness of natural gas odor.

The liner is just dropped in, and only cemented at the top and bottom. Your comment about air in the flue cavity warmed by the exhaust within the liner made me wonder -- could old exhaust deposits on the brick flue walls, when warmed in that way, escape into the house and give us the smell? It seems likely that coal was burned in this house at some point, though probably more than 50 years ago.

Mr. BobC,

There are some patches, but if the condition of the mortar of the chimney in the attic (I was up there today) is any indication, a bad patch might not be required to have a leak.


Gentlemen, thanks for the comments.

Jamie Hall

Oh yes indeed -- you'd be surprised how long an odour from old exhaust deposits can linger. Particularly with limited circulation. Coal is particularly bad this way.