New project: Fire sensing laser cut out system - and maybe more later.

[Graham Facer]

So I have 2 posts tonight. One is for an issue or 2 I'm having (though I may wait until tomorrow as I have some ideas after researching here...) but the other is a project that may eventually interest a lot of the folks here (and some may have tried it and have advice).

Anyhow I have been looking for an arduino micro controller project to do but one that wasn't useless (like blinking morse code in LED's) and eventually I realized I should be able to wire up one or many flame sensors in my laser cabinet and then if they sense flame for a certain length of time then they trip the laser power switch. So the gantry would continue but the laser would go off. Oh and an alarm would sound.

If I really get ambitious I might eventually try and rig it up to a relay controlled CO2 extinguisher for the ultimate in protection but that is several steps up in cost and complexity.

For now I have a $10 arduino uno rev. 3 knock off from ebay and a sample program that will sense the flame and start a count. If the flame disappears, the count resets, if not then the count eventually reaches a trigger and trips the relay/siren/whatever else I add.

So I'm still waiting on the flame sensor module and then I can play with it testing. The big questions are 1)if the cutting path will trigger it 2) if I can/should mount it with the laser head or under/over the table etc.

Anyhow should be fun to play with (for me at least) and hopefully reduce the damage of another fire (actually had a little one yesterday but no damage - just a little mess - need to go get a CO2 extinguisher.

I have been working on this for a bit but hoping the main sensor arrives in a few days for proper test.

Cheers, GF

[Graham Facer]

The micro controller is a SainSmart Uno R3 from ebay. It is powered by 5v which I hope to get from the USB inputs on the laser file input thingy but easy enough to figure out if that fails for some reason.

The sensor is one from DX (I think I got the 5 channel/sensor one but maybe 2 or maybe both - too lazy to look right now) and then also a higher value relay 20amp from DX too.

[Keith Colson]

Hi Graham

Great idea, I have been thinking about doing this myself for a while now. I just wanted something that would squawk so I could run and check the machine as I am never more than 50 steps away.

I was planning to stick about 4 or 5 thermistors just under glass lid . I chose thermistors for a few reasons. They are really simple to interface to a micro. They work really well when covered in dirt as things in the laser cutter get dirty. They are cheap and simple so you can use many at a very low cost. They won't false trigger i.e.most flames go out quickly. I am interested to see your progress.

Cheers
Keith

[Dan Hintz]

Projects like this come up every so often but never go anywhere... there are just too many problems to solve in such an uncontrolled environment. If you attach the sensors to the carriage, the head has likely moved on from where the fire is, so it does no good. If you use temp sensors, they likely won't trip until the fire is raging because the exhaust is pulling the air out of the cabinet before it can get hot enough and temp sensors have a long dwell time. Pyro sensors would need a wide view angle to see all of the table (even with multiple sensors), and how do you prevent false positives from the typical flare-up that self-extinguishes?

[matthew knott]

Wonder if monitoring if monitoring the temperature of the gasses flowing into the fume extractor would tell you anything in ? I would think you could use a TIC but as Dan points out, seeing the whole bed is near to impossible without multiple cameras and you would still end up with a system you cant really trust and you SHOULD be keeing an eye on it. Most fire detection systems are really to prevent loss of life or serious damage, a small fire in a laser will destroy it in seconds. Guess if you are determined to live it unattended it will be better than nothing but i would also have a normal smoke detector above the machine so hopefully you don't burn the neighborhood down!!!

[Scott Shepherd]

I've been down this path once before. I decided to do the same thing you mentioned, arduino with a fire sensor. In the end, I scrapped it all because a lot of fires happen under the work, not on top, and the visual fire sensors wouldn't do anything for that, since it's under the work. I stopped working on the idea because I found that not all flare ups were created equal, I couldn't detect the fires under the work, and some flames are different colors and it was my understanding that the sensors used color to determine the fire, and it appeared that it might not detect fire from certain products. It quickly became an overwhelming task to get right, and in the end, my thought was that I wasn't doing anything other than trying to give myself the ability to walk away from the laser when cutting and I didn't think that was a good idea.

Hopefully you can figure out the things that overwhelmed me and caused me to stop.

[Doug Griffith]

I went down that road as well and false positives stopped me in my tracks. I've since thought to create a more simple device that checks if my blower/air assist are running and sound an alarm if they aren't. My cheap little HF blower has died a couple times mid-project from brushes burning out and my paasche air compressor has overhearted a few times and stopped running. Both scenarios turn small little flare-ups into larger ones real quick.

[Graham Facer]

Ok great feedback. I can see the issues now. I try not to leave unattended but I'm a 2 man shop with a retail section so I am often pulled away. I usually try to check in at the start and then periodically. This is more a back up plan as if it just cuts the power to the laser - no harm no foul I'll just reset and restart.

I am leaning towards an under the table set up, false positives will, in theory, be done with a time setting. If the flame self extinguishes, the counter resets.

At any rate I'm treating this more as a hobby electronics project rather than a shop project so I'll carry on for now and post some feedback. Glad to hear others have tried, might save me some time in the end. Especially about the possible colour sensitivity of the sensors.

The tricky bit will be that I have the biggest issue with ABS (black) - both fires have been on that - and it also smokes like crazy when cutting. Anyhow, I might think about adding temp sensors to the underside of gantry as that is where the heat will damage the most. Anyhow, like I said, this is as much a project to learn the arduino as anything so I welcome the challenges - I'd thought it might be too easy but that has been dispelled!

Thanks, GF

[Matt McCoy]

Hey Graham: While I don't have anything to add to your cool sensor project, an angled mirror or small video camera helps to keep an eye on things from across the room.

[Scott Shepherd]

At any rate I'm treating this more as a hobby electronics project rather than a shop project so I'll carry on for now and post some feedback. G

That was the same thing I was doing It was educational to at least think about it and tinker with it some. One of my customers is heavy into electronics so I picked their brains about it and the possibilities. If it helps you in your situation then that's all the matters, plus you get to learn a lot from it all.

Just as a reference, Universal uses a temperature sensor inside the cabinet for their alarm system, so it's just monitoring the inside air temperature and setting the alarm from there.

[Dan Hintz]

I try not to leave unattended but I'm a 2 man shop with a retail section so I am often pulled away. I usually try to check in at the start and then periodically.

This is how everyone with a fire gets burned (no pun intended). They think "I'll just slip away for 30 seconds, what can happen in 30 seconds?" 20 seconds later their entire machine is engulfed in flames. No, that's not an exaggeration. Look at some of the threads here where people have stepped out of the room (some were in the same room but 10' away and not watching it)... entire shops have been destroyed.

Running jobs while dealing with customers is a sure way to eventually lose your entire shop. Don't say we didn't warn you.

[matthew knott]

Always thought something like this http://www.firetrace.co.uk/products/...ssure-systems/ might save the day in a nasty situation. Yes its going to make a mess and yes the laser may never work again but at least it should put the fire out before it lights the whole building up.

[Graham Facer]

This is how everyone with a fire gets burned (no pun intended). They think "I'll just slip away for 30 seconds, what can happen in 30 seconds?" 20 seconds later their entire machine is engulfed in flames. No, that's not an exaggeration. Look at some of the threads here where people have stepped out of the room (some were in the same room but 10' away and not watching it)... entire shops have been destroyed.

Running jobs while dealing with customers is a sure way to eventually lose your entire shop. Don't say we didn't warn you.

True enough. But that's how it goes. I can't sit and watch a computer controlled machine all day and make a living and even if I was in the same room it could still happen and quite frankly I don't have the patience anyhow. I'm am not saying that to be argumentative, that's just my feeling on the matter (your approach is way better).

I'd rather lose a 10k machine then waste 10k of time a year babysitting it. As for the shop, well I've had a pretty good fire it in once and certainly never felt the shop was in danger (except maybe from the smell). I can see that changing if I was cutting wood but most of what I cut has a slow spread when on fire. We will never eliminate the risks of fire when cutting with a laser, not even standing over it, we can only work on minimizing the results.

Anyhow on the project, one suggestion my brother had was an temp sensor externally mounted for ambient and one at the intake for the vent and then track the difference. That might work too in conjunction, but the more I think about it, the more adding in a CO2 flood to the cabinet if I can get the false positive things down to a manageable level.

Cheers, Graham

[Dave Sheldrake]

oxy.jpg

This started with a small fire in a laser that was unattended. It killed 3 (2 instantly and 1 died later) a 4th man now resides in a world of darkness having lost his sight and most of his facial skin.

I'm pretty laid back with the guys that work with me, they are friends as well as staff, if one of them left a machine unattended they would be unemployed the same day.

[Scott Shepherd]

I can see that changing if I was cutting wood but most of what I cut has a slow spread when on fire. We will never eliminate the risks of fire when cutting with a laser, not even standing over it, we can only work on minimizing the results.

Your website shows almost all of your products to be acrylic. If you are saying that that has a slow spread on fire, you might need to look for the post recently where thin acrylic was cut and caught fire. Acrylic is NASTY as far as fire. Been doing this about 8 years now and acrylic causes me GREAT concern. I'd leave wood long before leaving acrylic, and I cut a lot of acrylic.