Lochinvar boiler “hoot”s when starting up cold

Users who are viewing this thread

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
I had a Lochinvar Knight WHN 85 gas modcon boiler installed in the fall of 2014. It's worked well, except for one foible: when starting up from cold, it will often "hoot".

The hooting is at about 300Hz, and sounds like a foghorn or a French horn. It lasts for 2-20 seconds, easily audible indoors but even louder out of the supply and exhaust pipes (which happen to point at the neighbors' house). It only happens when the boiler is completely cold, and when a heating zone first calls for heat (not when the DHW tank needs heat).

A few months after the install, the installer came in with a specialist who was able to replicate the problem. He adjusted the air/fuel mixture and claimed to have fixed it, but the hooting still happens.

Here's a video of one such performance:
It starts with the cold boiler, and I call for heat. The unit purges, ignites, and then throttles down for efficiency. At about 35 seconds into the video it gets down to its lowest setting of 20%, and starts hooting. It hoots for about 15 seconds and then stops, probably because the boiler had warmed up.

I've done some more exploration of the problem. The supply/exhaust pipes are about 12' long, which is about three times the wavelength of the hoot, so it doesn't seem to be the pipes that are resonating (as a test I added a section of pipe to the intake while the system was hooting, and the pitch didn't change). I momentarily covered the supply pipe, and that immediately stopped the hooting. I'm guessing that it's the flame in the heat exchanger which is oscillating.

How can I solve this? The original plumber has retired two states away, and wasn't up to speed on the technical details anyway. If I hire another plumber, he'll probably see the problem once, twiddle the mix, not see it again (probably because the boiler had warmed up), and declare it done.

An update of the boiler's firmware could fix this, if the manufacturer wanted to. Is there any way to find out whether there are updates available?

Thanks,
Dan
 
Last edited:

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
I'm not sure what the problem is, but that's a LOT of boiler, significantly oversized for the average home in Melrose. With a boiler that could modulate to less than 17,000 BTU/hr (the min-fire input of the WHN085) it could be set up so that cold-starts are extremely rare. (The typical 1500'- 3000' house in Melrose would have a 99% design load under 45,000 BTU/hr, many would be under 30,000 BTU/hr.) With a min-fire output more than half the design heat load, it'll never modulate during the shoulder seasons, and would have increasingly long off cycles to cool down. The WHBN055 would have plenty of margin for most houses, and can throttle back to 11,000 BTU/hr in, for long burn cycles even during the shoulder seasons.

It's probably only resonant within a narrow firing range. Just because a the resonant cavity is 3x the wavelength doesn't mean much. At exactly 3x it's simply the third harmonic of the fundamental vibrational mode of the resonant cavity. Listening to the video starting at around the 40 second mark I can hear another harmonic singing away maybe 6-12dB below the louder tone rising and falling in amplitude, about a musical third lower than the louder note.
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
I'm not sure what the problem is, but that's a LOT of boiler, significantly oversized for the average home in Melrose.
<snip, see below>
(The typical 1500'- 3000' house in Melrose would have a 99% design load under 45,000 BTU/hr, many would be under 30,000 BTU/hr.) With a min-fire output more than half the design heat load, it'll never modulate during the shoulder seasons, and would have increasingly long off cycles to cool down. The WHBN055 would have plenty of margin for most houses, and can throttle back to 11,000 BTU/hr in, for long burn cycles even during the shoulder seasons.

Verrrrry interesting. My house is a garrison colonial with basement, about 2000 square feet, with 2 1/2 baths and a whirlpool tub. The original 1980-or-so boiler was a fixed 110kBTU/hr, which was plenty, especially after I'd done a lot of insulation and sealing. My own calculations based on past gas use came up with a heat use at 5°F of about 31k. I got a formal heat loss calculation done, and that came out to about 56k.

When I got bids for replacing the boiler, one guy spec'd a 133kboiler; when I protested, he said it would be "unethical" of him to install anything smaller. Another specified 105k, and also wouldn't go lower. I asked the final one to go as small as he could, and although he was surprised at the request, he ended up specifying that 85k unit, along with a Lochinvar SIT40 indirect tank (given the whirlpool tub I asked him to bump that up to a SIT50).

A year ago last January there was a windy night that cooled to -7°F. In the morning I got up, took a shower, and then checked the boiler: it had already replenished the hot water, and was running at 36%.

The boiler doesn't seem to be short-cycling, though: it's reporting 3800 run hours with 5000 cycles (300 run hours for DHW with 2500 cycles). So, I'm fairly happy with that aspect of things.

But, that's not what I'm concerned about. Quoting two sentences of your reply:

With a boiler that could modulate to less than 17,000 BTU/hr (the min-fire input of the WHN085) it could be set up so that cold-starts are extremely rare.
<snip again>
It's probably only resonant within a narrow firing range.

That makes sense to me: it only seems to happen when the boiler is dead cold, and also running at 20% (the minimum). The question is how to fix it. Some ideas (ignoring practicality):
  • Change the firmware or settings so that, after ignition, the boiler doesn't go below 30% burn until after the first minute
  • When a thermostat calls for heat and the boiler is cold, start the DHW and boiler circulators to take heat from the hot water tank in order to warm up the boiler. Then, once the boiler's up to 80°F, start the boiler (it wouldn't be too much work to whip up an external microcontroller board that handles this)
  • Find and change that resonance (again, it doesn't seem to be the piping, as when I changed the supply pipe length the pitch didn't change, although I may not have tried hard enough)
  • Bump up the minimum fire level, and ignore concerns about short-cycling and efficiency
  • Add a muffler to the supply and/or exhaust piping (hello, Midas?)
Any thoughts?


Thanks,
Dan
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
To collect more info, I pulled off the burner cover and video'd through the inspection port during a "hoot" event:
The burner wasn't completely cold, so the "hoot" only lasted two and a half seconds. But, during that time, you can see the flame around the burner turn bright orange. The oscillation is probably disrupting the nice blue flame; it might even be producing soot for all I can tell.


FYI,
Dan
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
And, here's another video where the burner was completely cold, so it hoots for over 20 seconds:
(Not sure what this'll prove, but I'm just hoping it will help someone figure out the answer...)
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
I've long since given up letting the HVAC contractors spec the sizing- 9 out of 10 are absolute hacks, and only 1 out of 20 seems capable of doing the load analysis correctly. The 56K load calculation estimate is extremely unlikely, and you can prove it with a fuel-use heat load analysis., which it sound as if you've already done in some fashion. When specifying a boiler for a friend in CT he got the same " I won't install anything smaller than ..." push-back from multiple contractors, which proved to be true- none of them got the business. ;) If your 31K calculation is reality (sounds about right, give or take) the WHN085 is sub-optimally oversized, given that the min-fire output is more than half that.

ASHRAE recommends 1.4x oversizing, which is sufficient to cover the load at the coldest temperatures in the past 25 years. So you can't expect to EVER need even 31K x 1.4= 43,400 BTU/hr of boiler output. Anybody's ~50K boiler would do it, even the Lochinvar CDN040 might have been a decent choice, depending on where the fuel use load analysis comes in. The NTI Trinity TX51 would be even better, since it can dial back to ~7K at min-fire and is well north of 50K on the high end for very reasonable recovery times after filling the big tub.

The orange flame during the hoot does indeed appear to be an overly fuel-rich burn for that couple of seconds. With a combustion analyzer you could demonstrate just how much it's off during that period, but whether that's fixable with the usual adjustments isn't clear.

5000 cycles over 3800 hours is definitely not short cycling. But is it operating in the condensing zone during most/all of those hours? It takes at least 80' of typical baseboard to balance with the ~16K min-fire output of the WHN085. Ideally you'd have at least that much on any one zone, which would allow you to drop the temps deep into condensing range without increasing the number of burn cycles.
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
The orange flame during the hoot does indeed appear to be an overly fuel-rich burn for that couple of seconds. With a combustion analyzer you could demonstrate just how much it's off during that period, but whether that's fixable with the usual adjustments isn't clear.

Yeah, and I'm worried that anyone I get into the house will see it happen once, then twiddle the mix, then never see it again (due to the burner being warmed up) and thus declare it "fixed".

Any ideas how to avoid this?
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
Make them guarantee the "fix".

I'm not sure where the closest Lochinvar distributor is located, but you might have better luck if you can get a reference for a Lochinvar factory certified tech for dealing with this more subtle (but not really subtle) issue, rather than some random mod-con installer with some Lochinvar experience.
 

Leon82

Member
Messages
169
Reaction score
3
Points
18
Location
Connecticut
How long is your vent. The internet has many threads like yours and not exclusive to lochinvar. The solutions that have been to analyze and adjust fuel mix, brace some of the venting and adding vent length

The manual call for 12 ft minimum
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
Interesting point. I just checked, and the installation manual calls for a minimum of "12 equivalent feet" for each vent. I just measured, and they're each about 11 feet long. However, "equivalent feet" includes extra length for any elbows. Once I add those in, the intake becomes 21 feet long, and the exhaust becomes 24 feet long.

When it comes to resonances like this, though, especially at low fire (and thus low air velocity), my guess is that the elbows really don't matter that much. So, it may be that the manual should have called for a minimum of 12 actual feet, ignoring any additional "equivalent lengths" caused by elbows or other restrictions.

I'll pick up some PVC pipe and duct-tape it to the two external vents and see if the furnace still "hoot"s.
 

NY_Rob

In the Trades
Messages
317
Reaction score
26
Points
28
Location
New York
Noticed this in the HTP UFT-80 (a similar Mod-Con boiler) install manual:
"NOTE: If the boiler makes a whistling sound (harmonics) at low fire, adjust the Gas Valve Offset Screw in a clockwise (positive) direction (approximately 1/8 turn). "
So, HTP has published advise for the condition the OP is experiencing with his Lochinvar... maybe worth a try- of course in conjunction with a combustion analyzer.
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
I did another test, adding about 2' of pipe to each of the intake and exhaust vents and trying another cold start. Lo and behold, the "hoot" was less than a second long:
Truth be told I hope this isn't the final solution, as adding piping outside the house will look awful, and adding piping inside the (very tiny) boiler area will be a pain in the tuchas.
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
Good suggestion. What I want is a minimum fire level for the first X minutes (e.g. don't go below 35% fire for the first two minutes). Unfortunately, the ramp delay sets a maximum fire level; the boiler is still free to go as low as it wants.

BTW, after a couple of days of tests, the extra 2' of pipe on intake and exhaust made some difference, but not a lot; the hooting was shorter, but still clearly present. It's quite possible that adding 10' of pipe to each would completely fix things, but there isn't room for that in my (again, very tiny) boiler area.
 

NY_Rob

In the Trades
Messages
317
Reaction score
26
Points
28
Location
New York
At this point- you should probably contact Lochinvar tech support for suggestions.
The must have come across this situation before?
The fix may be as simple at adjusting the mixture screw...
 

Leon82

Member
Messages
169
Reaction score
3
Points
18
Location
Connecticut
On a cold start I would think it would go to max fire unless the ramp is used. I haven't watched mine run in a while.

I would check the setting just to see
 

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
On a cold start I would think it would go to max fire unless the ramp is used. I would check the setting just to see.
BINGO!!!
Yes, indeed: ramp delay was enabled, with the first two minutes set to a max of 20%. I've bumped the initial maximum to 40%, and on startup the burner went to about 36% rather than its previous 20%. The burner was warm, so it wouldn't have hooted anyway, but I'll test it tonight when things are cold.

(Side note: the "RAMP DELAY" setting was actually "ENABLE", not "ENABLED"; the other choice was "DISABLE". I at first thought that "ENABLE" meant that ramp delay was currently disabled, and by choosing "ENABLE" I would enable it. But, testing both ways showed that there was a missing "D" in the display. Eh.)

There's still a diffuse orange glow around the burner, so I'll get someone in to adjust the mixture.


Thanks for the help, all!
 
Last edited:

Daniel Griscom

New Member
Messages
12
Reaction score
0
Points
1
Location
Melrose, MA
A postscript: here's my furnace doing a cold start, no "hoot"ing:


I set the Ramp Delay for 1 minute at 50%, 2 minutes at 20%, and then stepping up from there. I'd initially set it to run 30 seconds at 50%, but it turns out the ramp timing starts from the initial boiler startup, and it takes about 30 seconds for the startup process, so there was no net change. Setting it to 1 minute resulted in about 30 seconds of 50% fire before cutting to 20%, which works a treat.

Thanks again for the help.
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks