Heat Sensing Car Idea: your thoughts?

A volt-meter car (which does exist) would do as well in sensing such bad track joints...

Sounds like a reasonable idea to me, but the heat sensor would have to react very quickly if you were traveling at any speed as all. I think I have a couple of temp humidity sensors for Arduino, maybe I will check them out and see how fast they react. 

I have one of those point and shoot thermometers from Harbor Freight, that might be an alternative? You can get them for less than $20 during their sales. I think I saw an add the other day with one for $17 or $18. Not real accurate (it was either 2% or 5%, forget which?), but good enough to sense differences between a rail that's hotter than the rest of them on one's layout.

Maybe GRJ or Stan (or others) will be along and have some more info on your idea here.

Edit: GN Man posted while I was typing, there is a thread here somewhere on the voltage car the GRJ built, I copied his plan (with his help) and now also have a voltage car. So that is doable with plans and info already available from GRJ.

Probably would not be a big seller, the problem could be an issue, but so rarely, that few are likely to buy one, instead of something else(just MY opinion of course).

 I don't know if it would alert someone to a potential problem enough earlier than they would find it otherwise, or not anyway.

If I remember correctly, Max was getting sparking at that point well before the fire issue happened, and SHOULD have been inspected before the fire ever happened.

Would it be Possible, and or a Good Idea to have on a layout? YES

Would it be a PROFITABLE product for a manufacturer to produce? Not so Certain

Doug

Live feed camera car with a thermal image add-on.... still not cheap, but easy.

Railrunnin posted:

...This would potentially prevent fires and point out areas where there are excess current draws.

Paul, if this car was designed properly and could react fast enough, it could be a good idea. However, one point of correction here. Such a car would not find areas "...where there are excessive current draws..."

What it WOULD find are areas where there is excessive RESISTANCE to current flow. It is the excessive resistance which generates the heat.

If someone made such a car that worked I would invest in it.

Thank you everyone for the comments, opinions and insight.

The heat gun is an interesting alternative, but this idea was also going to be used to justify one more car to a work train.

Paul

Funny you should post this Paul, I was thinking the same thing the other day. There are plenty of thermal sensors out there, it just needs to be quick read. All you are really looking for is any sudden deviation from ambient.

It could be mounted on just about any car. It would be nice to hook it to a LED bar readout.

Hi Elliot, How about an inferred heat gun mounted in a box car pointed down through it's base. All it would need is the inner workings of an off the shelf product. When the gun hits a high, or hot reading let it sound a high pitch audible and a red lap on the roof.

You could also have a volt meter hooked up that when ever it  sees a voltage drop a bell sound rings along with a blue light flashing on the roof.

There is not a lot new design work that is needed to do this. General off the shelf technologies should be able to be adapted to work.

Of course, the electronics needed to translate from heat and voltage detection to alarms will take a little work but I think it would not be all that tough.

It would be a lot of fun to play with .

gg1man posted:

Of course, the electronics needed to translate from heat and voltage detection to alarms will take a little work but I think it would not be all that tough.

I refer you to the last line of my signature.

gunrunnerjohn posted:

gg1man posted:

Of course, the electronics needed to translate from heat and voltage detection to alarms will take a little work but I think it would not be all that tough.

I refer you to the last line of my signature.

You did pop into my mind John. I'm at work tonight but as soon as I get a few minutes I will research some inferred components and a AC volt meter.

The volt meter could be connected between two roller pick ups, one on the leading truck and another on the trailing truck. As Rich stated above what we are looking for is a voltage drop. The only technical issue if figuring out how do we handle that info.

I bet I could find a couple of suitable devices  that may even provided us with a ground closure, or even an led indicator that can be amplified. If that's the case problem solved. I'll let you do the hard part, squeezing ten pounds of electronics into a 1 pound car.

Let me know if you come up with anything.

Railrunnin posted:

Thank you everyone for the comments, opinions and insight.

The heat gun is an interesting alternative, but this idea was also going to be used to justify one more car to a work train.

Paul

How about a voltage car to start with? You could possibly add 2 cars to the work train if the temp thing is doable. As I said GRJ has a thread here somewhere with info on the one he made, which I copied. That has been working very well for a couple of years now and his for longer than that.

I will try to fiddle with the temp sensors tomorrow, but the ones I have were pretty cheap and I am not expecting too much out of them. From some of the other posts above, maybe GG1man will come up with an infrared sensor somewhere?

As rtr12 says, a sensor that itself must be "warmed" to the target temperature could not respond in time.  So, as has been suggested, a sensor based on radiated emissions makes sense.  For 99 cents (free shipping) you can get a PIR sensor module as used for motion-detectors in automatic faucets, room occupancy detectors, paper towel dispensers, etc..  You apply 5V DC to the module and a single output signal goes on/off when motion is detected in the field of view. 

Motion means thermal movement in the field-of-view of the funny looking lens designed to effectively amplify small thermal variations that move such as a human body on a hot summer day with ambient temperatures also in the 90's.  It's a remarkably clever concept.  Adapting this concept to the extremely short target distance from a car's under-belly to the track would be an interesting exercise.  But think of how the nominal target distance for these passive IR sensors started with many meters for room burglar alarms but now they have them in automatic towel dispensers with target distances in the inches.

But Rich makes an interesting observation that the heat is coming from current flowing thru a resistance.  That makes me wonder if the problem can be re-stated as how to detect an abnormal voltage drop between two points.  So, for example, if you have a 2-roller car, why not measure the voltage difference between the two pickups.  This might be easier to do than measuring temperatures.  Yes, there is a lot of noise from dirty wheels/rollers/track/etc., and going over turnouts, cross-overs, etc. must be considered, but sometimes the break-through comes from asking the right question.

That's a great idea Stan. I would guess that the lens can be adjusted for sensitivity.  Your right it is resistance that we would be looking for, but that would create a voltage drop. Any difference in potential would cause that.

I got a feeling that this car would have to operate at slow to medium speeds for best results.

gg1man posted:

That's a great idea Stan. I would guess that the lens can be adjusted for sensitivity.  Your right it is resistance that we would be looking for, but that would create a voltage drop. Any difference in potential would cause that.

I got a feeling that this car would have to operate at slow to medium speeds for best results.

If used on a conventional layout, a train run at slow to medium speed may not be demanding enough voltage to get that joint hot enough to be detected, or be an issue.

On a command control layout with 16 or more volts to the track at all times, you would have enough voltage to heat the problem joint, but the question then is, will the high temp sensing car detect the problem in the likely short time that it is detectable before it becomes a more serious problem such as a fire, or detected by the smell of the melting track.

Doug

An infrared reader set to trigger an alarm at 120° would likely catch things before they got dangerous, or just to help find trouble, but closely monitored gauges at they transformers would tell you something was up too

Get ready to have uncouplers trigger it unintentionally too.

I'm sorry, but this all seems to be a solution in search of a problem.  One would have to have such a heat sensor car in constant operation and monitoring to be marginally effective.

Exactly how many model railroads have suffered a fire due to sparking or high current draws?  It seems more likely a fire would be caused by poor under layout wiring, deteriorating wiring to the transformers or even a loose terminal on a wall outlet.

A smoke detector and fire extinguisher are more effective.

Rusty

Honestly, I feel that the easiest, most effective solution to this problem is good old common sense.  

Todd Knoll posted:

Honestly, I feel that the easiest, most effective solution to this problem is good old common sense.  

Unfortunately, common sense isn't always that common.

This car would be an electrical problem diagnostic tool. Dealing with electricity, even at low voltage, can be above many people's "pay grade", especially young Max whose problem prompted this topic in the first place. I know that when I was 14, I didn't know about this stuff. Thinking back, I probably experienced this when I was 7 or 8, but had no idea what it was.

The most likely cause of this issue is the repeated assembly and disassembly of sectional track. That process can loosen the pins and elongate the openings, creating a high resistance joint. If you don't know to look for this stuff, you can run the risk of starting a fire, which is what happened to Max. This topic is really about finding a clever solution to identify a hidden problem.

   Replacing the hollow pins with solid ones would be enough solution for my worrys; Ive got a"thing" against using them. My feeling track for heat lets me know when it needs cleaning too, yuk!. But after some thought, a thermal camera could be the closest thing to a "catch all" for the topside of a layout as a whole. A car or loco with a sensor falls into the gadget category though.

You don't need a track cleaning car.

You don't need a voltage reading car.

And you don't need a searchlight car to see that one more good gadget car and a decent gadget train is born 

A cheap infrared gun with a peak hold, a hole in the floor and wall of a boxcar, move the display to the car's wall and point the sensors out the floor. (Keeping it following the center rail in curves means aim perpendicular to the rail and from close to the trucks; a tight area to aim through with moving obstacles, a 2 axle bobber would be easiest to do this for.

A car as described might be a good seller if marketed well and had the "right look" to it. People like clever devices and looks are very important as well in selling the product. K-line had their voltmeter car and I purchased one because I liked the looks of the car and not just because I thought it would be useful. I could've easily used a voltmeter to do the same job, but it was the appearance that sold me on the car.

I just ordered $5 worth of the PIR sensors that Stan posted above. Can't resist these things. We will see what happens. I do agree that the voltage car might be an easier project. You could possibly even have pick up rollers on the far ends of 2 cars for a larger distance between the pick up rollers. This might be a fun project, we will see what happens.

Also for anyone interested the Harbor Freight IR temperature guns are on sale for $17.99, saw an ad for them this morning.