Defender works... and doesn't work.

Independence wrote:

...but depending on the ambient conditions, the exhaust can sometimes be subject to water condensation.

Let the drops fall where they may. I am hoping heat from the 24/7 operation will keep most condensation away, and I would hope these gadgets could handle a bit anyway.

...I need to run an entire tank...

I have gotten into the habit of replacing the tanks when they still have a few pounds of propane inside. This is practical because I buy propane by the gallon, so don't pay for any residual. Also, I don't want to have to purge any air from a refilled empty tank, and I don't want any funny business when the tank pressure starts to drop into the regulator.

What I really want is a weatherproof strain gauge platform for the tank to sit on so I can have continuous weight measurements. This would save me some bites.

How did you come up with 70C as the low end of the run state for your Defender?

In the initial investigation, logging the data, I noticed that the trap always reached operating temperature if it had first reached about 70°, so why not declare success sooner? (One of the design criteria was for the trap to use the same signalling as a stock trap.) Also, with the cover off, I noticed that the trap wouldn't reach the run state, which meant that replacing the igniter, etc., or some other adjustment required replacing the cover before testing, then waiting and waiting. I got tired of that. Also, when it got cold outside, the thermistor temperature might well go below 85°C, and I didn't want the trap to fault on that.

...over temperature limit was about 155 C...
My current limit setting of 140°C was set to protect the NodeMCU, which was near the combustion chamber, from overheating and then who knows what. This could happen with direct sunlight on a hot day and abnormal combustion.

I am writing a new blog post about recent mishaps with my Defender.

dev wrote:

Let the drops fall where they may. I am hoping heat from the 24/7 operation will keep most condensation away, and I would hope these gadgets could handle a bit anyway.

Good luck to you! I had two BME280 fail with moisture problems, and both were covered in conformal coating. When moisture gets in that little hole in that sensor, you're SOL. As mentioned, I switched to the Dallas Semi 18B20 device, but I will go back and see if a few weeks of drying out makes them work again.

I have gotten into the habit of replacing the tanks when they still have a few pounds of propane inside. This is practical because I buy propane by the gallon, so don't pay for any residual.

I do the same thing, but rather than weighing the tank, I would like to rely on the run time counter. In order to do that, I need to know what the full tank range is when the nozzle and regulator are operating normally. I got my schrader valve adapter and plan on doing some live pressure monitoring/characterzing on the units tomorrow.

I am writing a new blog post about recent mishaps with my Defender.

In your blog, you talked about splitting the combustion chamber to change the nozzle? Is that necessary on the Defender? On the Liberty and Independence, you merely loosen the lock screw and the nozzle/solenoid assembly just slides out of the casting.

In your blog, you talked about splitting the combustion chamber to change the nozzle? Is that necessary on the Defender? On the Liberty and Independence, you merely loosen the lock screw and the nozzle/solenoid assembly just slides out of the casting.

The Patriot also has a retaining bolt.

The Defender chamber must be opened to change the nozzle, a real nuisance. When opening the chamber, the electronics, thermistor assembly, and igniter should be removed first so they are not damaged. Of course, the Defender's nylon thermistor screw always breaks, and the nylon standoffs frequently break, and scraping and reapplying the gasket compound takes forever. My "no frills" exercise yesterday did none of that, but I was experienced and ultra careful, and I opened the case only enough to remove the nozzle, and no more.

Dev wrote:

The Defender chamber must be opened to change the nozzle, a real nuisance.

No kidding! I know to avoid them then. In the end, I guess the Independence is still the best unit overall for me, since it has the option for cordless operation, even though I currently don't use them that way. Plus it has a larger and easier to empty bag. Still, I can't help wanting to give the Patriot a go, but they never show up for free on Craigslist.

Independence Wrote:

No kidding! I know to avoid them then.

Actually, when they work, they are fierce defenders, more effective than the Liberty and, oddly, the Patriot. A couple of times, my catch basket (which is easy to remove and swap out with another clean one) was completely full, while the Patriot might have been <1/4 full (albeit twice as large).

Yes, nozzle replacement is a nuisance, but it happens only once every couple of years. And perhaps the catalytic converter might benefit from some periodic cleaning as well, although I have not tried that, unsure of how to do it and whether it would have any effect, and not wanting to ruin (e.g., "poison" or dislodge) the catalyst. Certainly there have been photos of deposits partially blocking the exhaust that required removal. I had some the first time I opened the case after many years, but not since.

Independence Wrote:

> If you have a 3D printer, you can print a replacement
>
> https://www.thingiverse.com/thing:3685540

Made it down to my buddy's place today, picked up my CO2 meter, and printed this cap.

I'm not a 3D printing guy. The slot in the cap where I assume the CO2+octenol are supposed
to fall out isn't open, which would seem to be confirmed by what I'm seeing in an STL viewer.

I know it seems like a stupid question, but should it be?

debugger wrote:

I know it seems like a stupid question, but should it be?

It's an entirely reasonable question, but the cap has no holes. I always thought it just leaked out the slightly loose fitting cap, but here's a picture of the attractant tube that sites inside the plume tube:



You can see that it has a series of holes all around it. Since it sits inside the plume tube, as the hot air passes by the attractant tube, it should draw the octenol out through the holes.

The lack of the cap should cause too much octenol to flow out, rather than not enough. I imagine the cap and the holes are designed to release the octenol optimally within their 21 day cycle.

Have you made any more observations about your unit? If you don't have a thermometer to measure the plume exhaust, just test it with your hand. It should feel slightly warm. And if you smell any rotten egg smell, that would indicate an abnormal combustion.

Ah, maybe we're getting somewhere now. Seeing a stock cap without the gripping bumps (that the printed one has) clarifies that the slot is for turning it with the usual coin and not an air passage. I'm going to get it on there and fire this thing back up again for a day or two.

I've been quiet and not messing with it pending getting the cap and my CO2 meter, and doing a bunch of cleanup to make room for the tank scale. Once I see whether the cap alters the catch I'll proceed as appropriate and advise.