The Tank › AO Smith (GS640YBRT) Pilot light problem
- October 28, 2012 at 2:03 pm #19110
I’m at my wits end with our water heater. It is a 7-year old AO Smith (aka State Select) natural gas heater. A few weeks ago, the pilot light went out for this first time in the 2.5 years we’ve owned this house. After lots of trial and error, I’ve discovered that the pilot and burner go out together in the middle of a burn cycle. This usually happens with the “WHOOMP” sound that I’ve heard other people describe.
I have read dozens of similar threads here and elsewhere. To date I have:
– Cleaned the flame arrestor plate (both sides) with compressed air, feather duster, and special vacuum attachment (purchased just for this occasion)
– Cleaned the black air filter at the base of the water heater
– Replaced thermocouple / pilot assembly
– Replaced thermostat/gas valve control
The parts were all replaced with stock AO Smith parts. The burner itself appears to be in very good shape.
No improvement. If I leave the thermostat on a lower setting, this doesn’t happen nearly as often (which is what steered towards a possible ECO/thermostat issue in the first place), but the replacement thermostat behaves the same way.
I now believe it may be an airflow issue (despite all the cleaning) because if I remove the orange gasket from the burner pipe there is an opening that allows more airflow and the water heater seems work properly. Obviously this isn’t optimal from a safety perspective.
Is there something I’ve overlooked? Or are there other options for safely improving the airflow?
-DaveOctober 29, 2012 at 2:28 am #19111Larry WeingartenParticipant
Hello: When this happens are you hearing any sizzling sounds beforehand? If a single drop of condensation lands on the thermocouple, it can cool it enough to shut the heater down. Also, does this happen after a large draw, like multiple showers? That would increase the likelihood of condensation.
I always look for the simple things first. 🙂
Yours, LarryOctober 29, 2012 at 7:50 am #19112
Thanks for the reply.
More often than not, it does seem to happen during a longer burn cycle. From my many observations the pilot goes out (along with the burner) during the burn cycle without, not after.
I’ve never seen any signs of condensation or heard a sizzle sound. There’s absolutely no rust on the burner or in the combustion chamber–I assume there would be some rust if this was a chronic problem.
Short of catching it in the act, are there any other tell-tale signs of condensation?
-DaveOctober 29, 2012 at 1:52 pm #19115Larry WeingartenParticipant
Hello: The biggest sign of condensation is drip marks at the bottom of the flue in the ceiling of the combustion chamber. Sometimes there are rust streaks running down from the center to the outer rim of the ceiling also.
Condensation is most likely to happen if there has been a big water use, making the tank start from cold; the ground water is cold, the relative humidity is high, or some combination of these.
Some commercial heaters come with little shields that live directly over the pilot area to divert any condensation “rain” from causing trouble, though I’ve seen some residential bowl shaped burners fill with water. Splash from condensation landing in the pool hits the thermocouple and shuts down the heater. 😛
I may be off on a tangent, but both pilot and main burner shutting off at once could be caused by condensation.
Yours, LarryNovember 1, 2012 at 7:53 pm #19123
Knowing this model I am placing my bet on the pilot assy temp limiter tripping. You can verify this by when it goes out you should not be able to relight the pilot for a few minutes. After the sensor cools down. The trick is to catch it going out. You bringing fresh air into the burner chamber by removing the gasket all most confirms this.
Things you can do.
1. Make sure the flame arrestor is clean. Both top and bottom. Feather duster works well and doesn’t damage the arrestor.
2. The radiation shield plate ( that big metal plate under the burner, isn’t warped blocking air flow. I take a long screw driver and pop the plate over the bump/bracket that keeps it in place during shipping. Make sure the plate is even all the way around. Shim it up if needed.
3. Do a draft test to make sure you are not blocked somewhere.
4. Pull the baffle out the top and inspect it.
One of these is your problem most likely.November 10, 2012 at 10:37 pm #19160
Thanks again for the replies. I’m hoping the flame arrestor is clean. I’ve done my best with the special vacuum attachments, dusters, etc. It’s very difficult to get to, and is almost impossible to visually inspect. Draft test was also favorable. Radiation shield plate also looks ok.
I’ve had the temperature set a little lower, and it has behaved for the last 2 weeks.
Today I drained it, removed the drain valve, and flushed it out. Lots and lots of scale came out. A lot went down the drain in the process, but I managed to save what probably amounts to 2 lbs of the stuff. Could this have acted as an insulator on the bottom of the tank and caused the burn chamber to get too hot?
I’ve since refilled the tank, and set the temperature back to my normal setting.
Side question: Where can I buy just the gasket for the combustion chamber door? The one that’s in there now got a little mangled. I think I’ve got it pretty well in place, but I’d just as soon replace it. I haven’t seen where they’re sold without being part of a burner assembly kit.November 11, 2012 at 10:11 pm #19162
Turning down the temp will decrease the burner on time and confirms that you need a new pilot assy. It comes with a new gasket. Also now that cold weather as set in this will help.November 12, 2012 at 9:58 am #19164
I’ve already replaced the pilot assembly as well as the thermostat/gas control valve. Problem still persists.
I’m fairly certain it’s airflow, but I don’t know what (safe) options are left.
Thanks.November 14, 2012 at 8:34 pm #19184
Something we discussed here is still wrong. You need to go back over your steps and check things again. Pull the pilot assy back out and look for discolorization of the thermocouple. Take a air compressor and blow out the flame arrestor from the inside to the outside. Raise the radiation shield 1″ all the way around. Then get back and let us know.November 23, 2012 at 4:38 pm #19215
You were right. Despite many attempts to clean the flame arrestor, I never could manipulate the compressed air can in far enough to hit the arrestor at point blank–it was always 2-3 inches away at best. Despite the numerous attempts at this, as well as vacuum attachments, feather dusters, etc., apparently none of it generated enough force to really clean the arrestor plate.
Anyway, I attached a flexible drinking straw as an extension to the tube that comes with the can of compressed air. With air blasting, I was able to move the straw over every square centimeter of the arrestor plate (inside the combustion chamber) with the end of the straw directly against the plate blasting from the inside out.
This must have done the trick since I’ve been running at full temperature for almost 9 days and no pilot light outages.
- You must be logged in to reply to this topic.