Understanding Propane Heater Behavior: Causes and Solutions

[DaveC426913]

I've asked a related question before, but this is new.

I have one of these propane heaters:

As months go by, it has an increasing habit of spazzing out.

It will frequently do this in a strong gust. But with even a little wind, and more often when it is set on low, it will spontaneously and startlingly begin screaming like a rocket engine. When it does so, the mantle immediately starts cooling.

It does this in one heater more than the other, so there's definitely some factor internal to the device, not just wind or fuel level.

It will occasionally spontaneously stop doing this, but sometimes I have to shut it off (by nudging it so it's anti-tilt safety mechanism activates, cutting off fuel.) I can always reactivate the fuel, and, if it's still hot enough it will auto-ignite and continue to heat up back to normal.

Obviously, something is happening to the position of ignition prior to the mantle, causing it to still burn (albeit very loudly), not to heat the mantle properly.

Let's pretend you've already suggested that I
a] take it apart and clean the nozzle parts, and/or
b] not mess with the mechanics of compressed explosive, flammable fuel delivery.

What is happening when it goes into this rocket engine mode? Is the "ignition front" moving back from next to the mantle to some other point farther inside the fuel feed?

 

[berkeman]

As months go by, it has an increasing habit of spazzing out.

It will frequently do this in a strong gust. But with even a little wind, and more often when it is set on low, it will spontaneously and startlingly begin screaming like a rocket engine. When it does so, the mantle immediately starts cooling.

Is it outside?

 

[Tom.G]

Is the "ignition front" moving back from next to the mantle to some other point farther inside the fuel feed?

Here are a few similar instances I've run across:

• Something similar happens with Oxy-Acetylene welding torches, even (especially?) with brand new tips. The usual cure is to run a cleaning burr thru the holes in the tip. (the burrs look like an assortment of wire feeler gauges but are actually round files)
• Another possibility is that the backside of a mantle is damaged or dirty causing a later flame front and sudden ignition (explosion?) of the incoming fuel.
• With lower pressure gas flames, a partially clogged nozzle or a too high inlet pressure will push the flame away from the nozzle with some audible turbulence

Maybe check the pressure regulator, or artificially reduce the pressure by partially closing the tank shutoff valve to see if that makes a difference.

Cheers,
Tom

 

[Asymptotic]

Which burner element is affected more ... the one closer to the hose connection, or farther from it?

Edit: Does it do this more often when propane tank level is low rather than when fitted to a full tank?
Is this a Dyna-Glo model TT30CDGP?

 

[ChemAir]

Two things I've not seen mentioned:
1) One thing I'd check (you probably already have) is the orientation of the air mix adjustment. On this type heater, it is usually right between the safety start solenoid and the burner. Usually a swivel collar with a couple of holes over a fine screen make this adjustment. Sometimes it either moves, or can get trash in it. This may need adjustment.
2) Propane tanks have excess flow protection that *sometimes* doesn't work very well, or maybe works too well. Normal procedure is--tank valve has to be off first with other valves off. Tank valve on, then light burner. If the tank valve is opened with a burner open (don't think you can do this with this type) the excess flow protection will restrict the max flow out of the tank, and it will be less than what the heater can use, and it will be erratic/unstable. I'd exchange the tank if the other suggestions don't help. I have seen some of these that just wouldn't work right, and the gas volume never would get to where it should be. It was always low.

Usually, performance deterioration over time is like @Tom G suggested, slow plugging of orifices by minute bits of debris, usually corrosion products or insect/spider parts. I wouldn't leave out the other suggestions, though.

 

[DaveC426913]

Thanks guys.Wow! I came to the right place.

Is it outside?

Yes. But under a deck roof, so protected from weather.

or artificially reduce the pressure by partially closing the tank shutoff valve to see if that makes a difference.

It has a primary flow valve off the tank with 3 settings. It is worse on a low setting than a high setting.

Which burner element is affected more ... the one closer to the hose connection, or farther from it?

It started with the one nearest the flow. I can't even use that one now. It's now doing it on the other one.

Edit: Does it do this more often when propane tank level is low rather than when fitted to a full tank?

Worse when low.

Is this a Dyna-Glo model TT30CDGP?

Not sure, but it's virtually identical to the one in the pic.

Two things I've not seen mentioned:
1) One thing I'd check (you probably already have) is the orientation of the air mix adjustment. On this type heater, it is usually right between the safety start solenoid and the burner. Usually a swivel collar with a couple of holes over a fine screen make this adjustment. Sometimes it either moves, or can get trash in it. This may need adjustment.

I'l give it a clean. Presumably 'adjustment' means it might need to be opened a little, as opposed to closed.

2) Propane tanks have excess flow protection that *sometimes* doesn't work very well, or maybe works too well. Normal procedure is--tank valve has to be off first with other valves off. Tank valve on, then light burner. If the tank valve is opened with a burner open (don't think you can do this with this type) the excess flow protection will restrict the max flow out of the tank, and it will be less than what the heater can use, and it will be erratic/unstable. I'd exchange the tank if the other suggestions don't help. I have seen some of these that just wouldn't work right, and the gas volume never would get to where it should be. It was always low.

I've encountered this, yeah. If I don't tighten the connection enough, I get almost no flow. PRetty sure this is not the problem.

Usually, performance deterioration over time is like @Tom G suggested, slow plugging of orifices by minute bits of debris, usually corrosion products or insect/spider parts. I wouldn't leave out the other suggestions, though.

Thanks, I'll scare up some pipe cleaners.

 

[Tom.G]

Thanks, I'll scare up some pipe cleaners.

Don't be surprised if they don't do much. Any debris in the orifices will be well carbonized.

 

[Asymptotic]

Don't be surprised if they don't do much. Any debris in the orifices will be well carbonized.

The same type of cleaner used to maintain oxy-acetylene torch tips may prove useful. Pictured is a Hobart #770084, and it (or it's functional equivalent) ought to be available from local shops that deal in gas welding supplies. This model has 12 round cleaning files spanning a range from drill size 49 to 75 (0.073" to 0.021"; 1.854 mm to 0.533 mm).

Edit: Reading through the thread more closely, @Tom G had already mentioned them.

The usual cure is to run a cleaning burr thru the holes in the tip. (the burrs look like an assortment of wire feeler gauges but are actually round files).

 

[DaveC426913]

Thanks. I might just put the cost for that cleaner toward buying another heater. They're about 60 bucks.
But considering the season, I might just wait till autumn.

 

[Asymptotic]

Is this a Dyna-Glo model TT30CDGP?

Not sure, but it's virtually identical to the one in the pic.

Reason I'm asking ... there is at least one mention of a similar symptom in the Amazon critical reviews for this model, for instance,

Seemed cheap at first but it did work and seemed like a great product for about a month or so. Then one day out of no where the flame would “back wash” into the cone to area not designed to burn fuel while at the same time making an incredibly loud deep whoosh sound like an afterburner on a jet aircraft! At this point you must shut the unit down immediately as the back of the cone starts to melt and could very easily start a fire in the surroundings.

 

[Averagesupernova]

My experience with those types of propane heaters is that unless it is a Mr. Heater brand it is not worth bringing it home

 

[DaveC426913]

Huh. Good to know. Thanks Asym and Aver.

 

[Averagesupernova]

Mr. Heater does some rebranding. They are not manufacturing all of the stuff that has their name but I have found their stuff to be more reliable than others.
-
Thinking about this more, I have found similar symptoms to what you describe on a near empty tank. I don't know what the minimum inlet pressure is on the regulators that these heaters use but I suspect that they are made to be safe but that's about it.

 

[DaveC426913]

I don't know what the minimum inlet pressure is on the regulators that these heaters use but I suspect that they are made to be safe but that's about it.

I've seen it running right until it runs out of gas and the flame dies.

 

[Asymptotic]

I've been wondering whether a subtle interaction between wind gusts, low burner temperature, and how the gas safety valve operates may not be implicated. Haven't found specifications on how this safety valve operates, but along the way found a resource chock full of related information that's making for an interesting read.

"TECHNICAL FEASIBILITY OF A CO SHUTDOWN SYSTEM FOR TANK-TOP HEATERS", United States Consumer Product Safety Commission
https://www.cpsc.gov/s3fs-public/tanktop.pdf

I had been envisioning a burner head similar to that of a stove burner placed behind the screen, but thus drawing suggests otherwise. Is this more like how your burner is?

Does the flame have a significant amount of blue to it as shown in the above pics?

 

[DaveC426913]

That is an exact representation of my burner.

No, it does not have any blue. It's all bright red, until it goes splah, then the glowing red goes away and I hardly see any flame. (even in the dark, so the blue should be noticeable.)

It does however, start melting the dish pretty quick.

 

[Asymptotic]

It does however, start melting the dish pretty quick.

Should I take it that the dish is aluminum?

Can you tell if the gas safety valve itself is making any noise (in addition to the rocket engine scream)?

 

[DaveC426913]

Should I take it that the dish is aluminum?

I lie. It doesn't start melting quick. It only did that the fist time, when I let the thing run for minutes, thinking it was doing no harm. After I saw the melting, I now always kick it off right away.

Also, now that I think of it, it's not the dish that started melting, it was the housing (the 'plenum' housing in your diagram).

Can you tell if the gas safety valve itself is making any noise (in addition to the rocket engine scream)?

I'm not gettin' my face near that.

 

[Tom.G]

Also, now that I think of it, it's not the dish that started melting, it was the housing (the 'plenum' housing in your diagram).

Ahh, that's a big clue!

That Burner Head Screen exists to keep the flame out of the plenum. Maybe you recall a demonstration in High School physics or chemistry class with a Bunsen burner and a metal screen. Rest the screen on the burner and ignite the gas. With the screen resting on the burner, there is a normal flame. Raise the screen and the flame rises with it, never returning to the burner orifice.

What is happening in your case is, for whatever reason, there is enough heat conductance to the plenum side of the screen to ignite the fuel in the plenum.

Some possible root casues:

• a hole in the screen
• screen mounting loose
• contamination conducting enough heat thru the screen for ignition in the plenum
• borderline design of the screen being too coarse

Hope this helps.

Cheers,
Tom

 

[Averagesupernova]

Yep. The fuel is supposed to burn on the mantle which in the pic is referred to as the burner head screen. I believe it is a ceramic perforated surface. Maybe the cheaper units are not ceramic?

 

[DaveC426913]

What is happening in your case is, for whatever reason, there is enough heat conductance to the plenum side of the screen to ignite the fuel in the plenum.

Some possible root casues:

• a hole in the screen
• screen mounting loose
• contamination conducting enough heat thru the screen for ignition in the plenum
• borderline design of the screen being too coarse

Erm. Maybe. You don't think the flame can "escape"* to inside, if flow is low, or a wind catches it?

*inscape? Is that a word?

Yep. The fuel is supposed to burn on the mantle which in the pic is referred to as the burner head screen. I believe it is a ceramic perforated surface. Maybe the cheaper units are not ceramic?

It's the same as in the pics. It's a woven wire mesh. Can they make and weave ceramic wire 1mm thick?

 

[Averagesupernova]

The front most part that glows red is woven. What about what is behind it? Isn't that ceramic?

 

[DaveC426913]

The front most part that glows red is woven. What about what is behind it? Isn't that ceramic?

Ah.

 

[DaveC426913]

Upon reflection, it might be doing this due to normal wear & tear. I put a lot of mileage on it. Like 2 tanks a month. It's my outdoor man cave ,where I smoke my stogeys all winter long.

And I suppose kicking it to shut it off is not helping the mantle stay in one piece.

 

[Tom.G]

Erm. Maybe. You don't think the flame can "escape"* to inside, if flow is low, or a wind catches it?

Yup. I've had it happen, the scarier one is with a welding torch. They get HOT!

*inscape? Is that a word?

If not, it should be.
(Oh, phooey. https://en.oxforddictionaries.com/definition/inscape says it is, but a completely different definition.)

 

[Asymptotic]

Was the location that was getting hot in line with the mixing tube?

Photo from the CPSC paper. In the 'left side' view heat discoloration is obscured by what appears to be flat black paint that had been sprayed on to facilitate IR measurements. Presumably, discoloration in the 'right side' view occurs with typical usage. I'm wondering, is the location that was approaching melt-thru approximately where the circled red arrow is drawn?

@Tom.G makes a good point in #19. I'm thinking along the same lines - the flame moves back into the mixing tube at which point the rocket noise commences and plenum overheats.

Typical temperature distributions (again, from the CPSC reference).

As the burner face cools off I'd expect the gas safety valve thermocouple (shown as a long cylinder with pigtail near the TC #1 position in figure 23) to also cool, close the gas valve, and extinguish the flame. It has me wondering why it doesn't happen this way - is hot gas from the displaced flame flowing over the thermocouple and keeping the gas valve open? Or is something else occurring?

 

[DaveC426913]

Looked like the in right photo. The top of that section started to deform.

 

[Sherwood Botsford]

We had Meker burners in a lab I worked at where the flame front would managed to get across the screen top mesh and burn in the barrel. I suspect you have the same problem here, that the heat from the burners is igniting the fuel in the supply line.

Things to check:
Compare the temps of the supply lines when it is operating normally and when it's making the racket. If parts of the supply lines get very hot, the hypothesis gains credence.

Can you run a single burner? If so does this happen with a single burner only.

Things to try:

Put some head shielding between the burners and the supply lines.
Put a wad of stainless steel chore-boy in the supply pipe just before the entrance to the heaters. This makes it very difficult for the flame front to advance back if indeed that's what's happening. Warning: This may cause propane to find new routes.

With our Meker burners we replace the top mesh with a finer mesh and the problem vanished.

 

[CWatters]

Check the air mixing tube.

I took a gas lantern that has a similar mixing tube camping once. After a few days it started to play up so I disassembled it. The mixing tube looked blocked and I couldn't see how gas was meant to go up it. Prodding with a stick I discovered a spider had gone in the air hole and sealed itself in. The web it had used looked exactly like the grey inside of the metal pipe.

Two years later, another camping trip and another spider did exactly the same thing.

 

[DaveC426913]

I've never understood what fascination gas ports hold for spiders.

I guess it's inevitable that spiders will find them as much as any other crevice.

 

[ChemAir]

We had a thermal oxidizer on the tail end of a sulfur recovery unit have intermittent "flame failure" trips on the burner management system. This was a pretty big deal. Turning off an incinerator while full of bad stuff wasn't something you wanted to do. Management pressure to fix this *right now* was high.

After a while, we found a spider using the air cooling bleed as an entrance to the flame detection port and it could get close enough to the detector for it to fail self-check. Insect screens, redundant flame sensors, etc. were installed to address this and other issues.

I hate to think how much money that one cost. After that, I tended to refer to any arthropods/parts in incinerator, furnace, or other burner instrumentation systems as "debris", which for some reason was more politically correct than "spiders" at that facility.

 

[anorlunda]

I hate to think how much money that one cost.

You would be shocked to hear how many people have died because insects block the fuel tank air vent in small aircraft.

 

[Asymptotic]

Have you gone poking around yet to see what may be blocking the mixing tube, and other flow paths?

I was thinking about a cheap way to fashion a steel cloth screen around the mixing tube air inlet. One option is to adapt stainless steel teapot mesh strainers similar to those by uxcell sold for $5 USD through Amazon like so.

Don't know how I'd join the two pieces together, but it's that time of year when fishing comes to mind, and sewing them together with stainless steel leader would probably do the job.