Printer fire hazard

I have been very worried about a printer fire for overnight printing. So I got a metal double door wall cabinet (came with feet too) to enclose the printer. Still you could have a fire potentially producing a lot of smoke which would be bad. I think I have a cheap solution. I got an old style CO2 bike tire inflator turned from aluminum. I am going to plug the end that attaches to the tire tube with Roses metal that melts at about 200F. It’ll poke through a hole in the top and the 20 gram CO2 cartridge will snuff out the fire without damaging the printer. A second woods metal fusible link will open the power circuit at 158F first so the heat source is disabled before the extinguisher.

This along with a smoke detector in the printer room should be as safe as possible without going completely crazy. About $50 total.

The fire will die out quickly in a closed cabinet.
It takes force to punch a CO2 cartridge. You will need a system inspired from a hand grenade firing pin to punch it (they are battle proven to work in adverse condition ). You can't punch it in advance and rely on a valve that is likely to leak or block in the long run.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

I think it's a great idea but I think you may find it requires more engineering than it appears at first glance. My first thought is that you need to be really sure that the thing is still... armed, which to me means either that the bulb is only punctured when the system activates or that you have some sort of pressure gauge to reassure you that the CO2 hasn't leaked away.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

Sounds like an interesting idea! But I agree with previous posts; if you do not have some way to either puncture the tube when fire starts or pre-puncture and monitor pressure it might just lead to a false sense of security. I too worry about fire and would be very interested in a solution like this. Please keep us updated!

Perhaps consider one of these fireball systems which burst when heated and hang it in the cabinet.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

You don't have to invent everything. I always try and look for things that already exist.

[www.ebay.com]

or

[www.ebay.com]

Both under $100 and a higher probability that they would work than something I cobbled without the testing needed to insure function.

How about a large plastic bag of sand suspended over the printer that would melt and spill in a fire?

How about a smoke detector in the enclosure to warn you?

A thermocouple that would sense heat and cause an action (maybe roll the unit outside?)

How about setting the printer on a spring loaded platform that is held cocked by a thin rope strung over it, that would melt in case of a fire and fling the unit out a window?

A halon bottle with a fusable plug to melt and blow out the fire? $$$$ but you can't put a price on safety!

So many options, my current focus is on keeping my printer from starting on fire in the first place.

The "stove" extinguisher contains a fire suppressant that will corrode the printer. It will likley stop the fire but any part of the printer that survived the flames will be destroyed by the fire suppressant. The other one seems like a better choice. An "clean" fire suppressant like co2 is preferable for this reason.

Att what temperature does a co2 cartridge rupture when heated? I remember Them saying something like "do not store in temperatures above 50 degrees celcius". But I guess that statement has some saftey margins. IF it does rupture I guess it would burst at the weakest point, likley where you normaly pierce it.

Anyone got a cartridge to spare and want to try it out?

Have you considered careful examination of the printer and trying to reduce the risk of a fire starting?

Putting out a fire after it starts is like closing the barn door after the horses have run out.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

We all do that, or should do that as a matter of course. However some things are only apparent in hindsight, if something completely unexpected occurs I'd rather have something that deals with the problem caused by an accident rather than relying on predicting every eventuality beforehand.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

How many 3d printers have started a fire?

Quite a few. Do some google searches om the subject. Sure, most have been cheap kits with substandard electronics and thermal runaway disabled. But the potential to start a fire is there in all printers.

the_digital_dentist; why not do both if the extra protection against unforseen risk can be had for a reasonable cost?

Anyway, in a well closed steel cabinet, the fire will die out.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

I wouldn't uses the Roses metal as its difficult and local.
I believe the tyre kits come with a connector to attach to the tyre. If you modify this and fit a plastic pipe instead of the tyre you can route the pipe all round the cabinet. The pipe can have a nice low melting point - say 100C. This will melt where it is hot and deliver the C02 at the melting point. I wouldn't think you will get 100C anywhere in the cabinet apart from contact with the bed or hot end. By the time you got the entire cabinet hot enough to melt the metal i would expect a major fire to be underway.
And the canister is outside the cabinet - you just feed the plastic hose through a hole.
Seal the blank end with a screw and some heat shrink tubing.
I guess you will get a shot of air from that traped in the tube but the C02 should clear that.
Not sure what volume of C02 the cannisters contain but it needs to be sized for the cabinet.

Quote
MCcarman
I wouldn't uses the Roses metal as its difficult and local.
I believe the tyre kits come with a connector to attach to the tyre. If you modify this and fit a plastic pipe instead of the tyre you can route the pipe all round the cabinet. The pipe can have a nice low melting point - say 100C. This will melt where it is hot and deliver the C02 at the melting point. I wouldn't think you will get 100C anywhere in the cabinet apart from contact with the bed or hot end. By the time you got the entire cabinet hot enough to melt the metal i would expect a major fire to be underway.
And the canister is outside the cabinet - you just feed the plastic hose through a hole.
Seal the blank end with a screw and some heat shrink tubing.
I guess you will get a shot of air from that traped in the tube but the C02 should clear that.
Not sure what volume of C02 the cannisters contain but it needs to be sized for the cabinet.

He said it was a 20g CO2 canister.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

I bought two cylinders so I can test fire it. After puncturing cartridge I will bubble test for leaks. Might have to use locktite on cylinder threads for long term. Simple weight check to be sure it hasn’t lost any. CO2 won’t damage printer. Yes the metal cabinet may snuff the fire but lives are at stake not just the printer. Yes there will be a smoke detector too.

You can also get much larger 88g co2 cartridges and if even that's not enough you might look at sodastream cylinders or welding gas, I believe both of those have adapters commercially available.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

It will will be interesting to see how this works

The sodastream ideal seems great in my mind. A used sodastream could donate the screw om valve/seal. Those things will have to hold the system pressurized and ready to use when operated as intended. Adding a termo-trigger (I like the hose-idea that was suggested) and pressure gauge would give a quite powerful automatic extinguisher.

Save the planet, no more CO2 in the atmosphere please.
I even stopped drinking soda !

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
MKSA
Save the planet, no more CO2 in the atmosphere please.
I even stopped drinking soda !

You know all those Fat people in the 7-11 with a trolley full of huge bottles of fizzy goodness,
I now realize they are not Fat, just full of CO2
Though Now the Climate has changed, and this kind of joke is victimizing CO2 Addicts.

This net sleeve stuff to hide wires burns really easy.

Edited 1 time(s). Last edit at 02/20/2018 09:29AM by MechaBits.

Quote
MKSA
Save the planet, no more CO2 in the atmosphere please.
I even stopped drinking soda !

In a best case scenario it will never get emitted to the atmosphere

You could always use oxygen instead to save the planet!

Meh. So much better a little CO2 than the CO2, CO, H2S, HCl and other noxious substances that an electronics and plastic fire will emit.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts

OK, your printer is inside a metal cabinet, but it may catch on fire? Before starting the occasional overnight print, throw a chunk of dry ice in a poorly sealed cooler into the bottom of the cabinet and tape up the door. This should allow the dry ice to off gas the co2, and the cooler will moderate the change in state. This should provide an inert environment.

You can never be too safe.

I do this with my toaster, coffee maker and crock pot whenever I can't watch them.

Maybe take it out of the metal enclosure and watch it?

Any chemists here?

I know there are some minerals that thermally decompose releasing CO2 (Carbonates?). Anybody know what temperature decomposition starts at and how much mineral you would need to fill a cabinet with CO2?

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

CaCO3 + H2SO4 -> CO2 + H2O +CaSO4

CaCO3 = chalk
H2SO4 = sulfuric acid, but any strong acid will work

But I take issue with @kengineer- I think you *can* be too safe. It makes sense to protect yourself against reasonable risk, but it is a waste of time, energy, and resources to protect yourself against unreasonable risks. And mechanisms to deal with one risk (fire) can introduce other risks (e.g. handling acids).

NB:

1. Flooding the cabinet with CO2 isn't going to protect you against any fire started outside the cabinet. Are you going to put your PSU inside the cabinet too? How are you then going to safely get your mains power into the cabinet? What about ventilation and cooling for the PSU and controller? How are you going to ensure that there is sufficient CO2 in the cabinet all the time to prevent a fire?
2. Opening the cabinet door will introduce oxygen, allowing an incipient fire to ignite. It will also allow your protective CO2 to escape.
3. You also need to protect against heat alone burning/charring/melting an opening in the cabinet, or conducting through the cabinet to start a fire on the outside of the cabinet.

Quote

Quite a few. Do some google searches om the subject. Sure, most have been cheap kits with substandard electronics and thermal runaway disabled. But the potential to start a fire is there in all printers.

the_digital_dentist; why not do both if the extra protection against unforseen risk can be had for a reasonable cost?

I Googled. I found a few actual cases.

• two printers caused actual fires that spread: [imgur.com]
• two printer fires that could have spread but didn't [blog.lessdebug.com] [forums.reprap.org]
• a number of printer could have caught fire but didn't
• clouds of propane from hairspray which had accumulated over several hours had been ignited by a spark (unkknown source) and ignited other *very* flammable materials incorrectly stored nearby. The person present was overcome by fumes and died. This was headlined as "Explosion caused by 3D printer and hairspray kills teenage boy". I can't believe that (a) enough propane from hairspray could accumulate in several hours of 3D printing to cause an explosion, nor (b) that a 3D printer would cause a spark.

[3dprinterchat.com] has an interesting analysis. Common problems are:

• Bad connections at the screw terminals on the RAMPS board
• Bad power supply quality
• Wrong cables (rigid rather than flexible)
• Thermistor escaping from hot-end
• Not using cable chains (cables getting tangled, fatigued)

Addressing these would eliminate most of the risk of fire from 3D printing.

Thermal runaway protection in software will protect against some of these. I don't know of any cheap kits that ship with thermal runaway disabled. As a data point, I bought the cheapest Delta kit I could find and it had thermal runaway enabled in the software. An obvious fix is to replace the supplied firmware with something that has thermal runaway enabled.

One fire above was caused by stepper wires tangling and breaking and shorting. A bit of practical electrical cabling installation knowledge is necessary when building a kit, I guess. Incidentally, my cheap kit came with yards of spiral wrap, which works well as "cable chain".

A smoke detector which controls supply of power to the printer seems to me to be a reasonable protection, especially if the printer and detector are inside a cabinet. (Obviously, the wiring of the smoke detector and its associated relay needs to be safe. And, I guess, whoever wired up the printer will also wire up the smoke detector and relay).

However, protecting against *unforeseen* risk is a crap-shoot; if no-one has foreseen the risk, then it's a matter of luck whether protections designed for other *foreseen* risks also cover the unforeseen risk.

Reasonable precautions, as said.

CO2 is heavier than air. Opening the top would not immediately allow the CO2 gas to escape, nor immediately introduce oxygen in the case of material already heated past flashpoint in air. Of course HOT CO2 will be lighter, and may escape from the top.

CO2 has its own risks, not the least of which is that it's not breathable, A 1-2% by volume percentage of CO2 in atmosphere will have immediate and potentially serious health risks. It doesn't need much more to be potentially fatal. A well ventilated area is probably called for, and I would strongly advise against this if the 3D printer is located in a basement. (See point above.)

Dry ice in your 3D printer may ruin the functionality of the 3D printer. Dry ice is really cold. 3D printers work best when the chamber isn't.

Many of the cheap power supplies in these things aren't of great quality. I've yet to see a CSA or UL stamp on one of the "LED strip" PSUs that come with most of the kits, or are sold by places that deal with kits. Mine sits on a nonflamable slab, and is powered from a GFCI outlet. I believe that this is reasonable protection.

Bad connections at the RAMPS board poses a risk. I use terminal ends, soldered on. I don't like the cheap, thin wire that goes to the hotends much, and going to the heatbeds, I like it even less. I use 14AWG automotive power wire in its place.

I used to just carefully place the wires. I had the print bed cut a wire once. Drag chains no longer seem like a waste of time anymore.

I haven't tested, but I'm pretty sure that I'd get a "Heating failed. Printer stopped" if the thermistor fell out of my hotend.

Overall, I'd say that I have most of my risk bases covered. I have a small fire extinguisher next to the 3D printer, and it happens to be close to my smoke detector. I haven't left it running while I'm out, but have left it going while I slept.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts