Circuit to detect a pipe wall

[fikseren]

We need suggestions on how to detect the outlet of a 1-inch pipe. A unit will travel in the pipe and detect the wall and we need to detect the pipe end in less than 4 millisecond. The sensor must be electronic and we are thinking of an IR-sensor. The sensor must be insensitive to daylight. Do anybody have a proposal.

 

[nbucska]

Need more explanation. Why detect the wall, if the sensor is inside the pipe? What is the material? Is it a drawn or welded pipe? How is the sensor moved? ( cable, air pressure, rod, etc.). How much accuracy is required?
Any other info you may be able to tell?

<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032

 

[fikseren]

This is a small pellet that is supposed to ignite when leaving a pipe, which can be made of steel and/or GRP (plastic). As long as the pellet is within the pipe nothing should happened. Moving force is compressed air. Sensing a steel pipe is fairly easy but both steel and plastic on the same run is more difficult. Mechanical sensing with a micro switch cannot be used, as the unit must be totally sealed. The only requirement is the ability to sense the wall, to be or not to be in the pipe. An additional necessity is speed. Detection in less than 4 millisecond.

 

[nbucska]

Light detector?

<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032

 

[logbook]

I assume the pipe is empty since you have omitted this key "detail".

The trouble with IR is that the surface will affect the reflectivity. Is it old or new pipe? Old dirty pipe may absorb most of the light and make the item think it is in open air. Unless you can guarantee the nature of the surfaces, you would have to do extensive testing to ensure the IR would work. Make sure you modulate the IR at some rate (say 50kHz on/off rate) as this will make amplification and detection easier.

If the surfaces are clean then IR is a good bet. If the surfaces are a bit arbitrary then I would consider ultra sound up at 40kHz.

Micro-switches should not be ruled out. The easiest answer is a magnet on a plunger mechanism. If the magnet is allowed to move away from the case then a reed relay inside can be de-activated. This is by far the easiest sealed system as the magnetic field passes through the sealed interface into your object. You "just" need a spring loaded magnet holder assembly which lets the magnet move away when the pipe wall disappears.

 

[MacGyverS2000]

How about surrounding the pellet with a series of "ball bearing on top of plunger" deals, rather than just plungers... less chance of it getting hung up.

I'm really curious as to the application... igniting pellet? Sounds like a potential weapon.

 

[sreid]

If the launch is in daylight, how about a photo sensor that detects day light. The assumption is that the tube is dark and sensing day light means you have exited the tube.

The sensor could be activated only after an accelerometer see X g's for Y amount of time.

 

[fikseren]

The ignition pellet can be launched during day or night. The pipe length can be up to 2000 meter and continuously purged with dry air. The internal surface is clean but the wall may be sooty by burned hydrocarbon. Travelling speed will be in the range of 10 to 100 m/s. Using a ball bearing on top of plunger is a good idea. It is not a weapon. The pellet should put fire to a flare at the end of the guide pipe. As the pellet will increase and decrease speed during the journey, an accelerometer could not be used I think.

 

[ScottyUK]

What about sensing pressure? The pellet must experience a positive pressure in the pipe while it is travelling through the pipe and which dissipates as soon as it leaves the pipe. Not sure if the 4ms detection requirement could be met.





----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[fikseren]

If we loose pressure during trevelling in guide pipe, we will have an ignition in the pipe. No good.

 

[ScottyUK]

If this is an application which could give rise to a hazardous condition, then regardless of the technique you eventually use consider redundant sensors or monitor two conditions independently.



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[jraef]

I take it the pellet needs to be self powered by a battery or capacitor, i.e. no trailing wire for a power supply? For instance, an inductive proximity sensor could be looking for the presence of a ferrous metal, so when it gets to the end and the metal is gone, it changes state and ignites. Might require too much power though.

&quot;Venditori de oleum-vipera non vigere excordis populi&quot;

 

[Higgler]

If you have a small amplifier chip and add printed antennas to either end it will oscillate in the tube and stop oscillating outside the tube. A circuit to detect lack of oscillation could trigger the explosion.
When you place it in the tube you could sense the oscillation by external equipment to verify you're in a safe mode.
I agree two modes of safety should be used, or three. A simple timer to not allow the event until it has travelled X seconds from turn on would be good too.
kch

 

[nbucska]

Use external sensor and ignition with laser or HV spark.

<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032

 

[sreid]

How about some kind of proximity sensor. The idea is to sense whether you are near or not near a metalic object: inside the pipe or not insiude the pipe.

 

[curious125]

Could the initial "ignition" be from an external source?
If you mounted a coil around the the end of the pipe outlet you could create a magnetic field that the object passes thru.
Inside the projectile you could use a set of contacts either opened or closed by the field as it passes through it. This would allow you to carry only enough power required to set off the charge.
This way the pipe material and inside surface finish are not an issue (or less of one).

 

[notnats]

How about an ultrasonic device. When it no longer receives an echo then it is out of the pipe.

Jeff

 

[fikseren]

Thank you for all feedback, overwhelming. I should have given more information at the beginning. We cannot use proximity sensor, as the guide pipe is plastic and partial steel. The pipe end is close to the flare and can reach 800 centigrade. To hot for any instrument installation. We have tried IR-sensors but it is difficult to sense different materials and surfaces. May be our technology is too simple. We have suppose that an acoustic sensor may work, but fear that it will be too expensive. We may be wrong. As for now we are back to more or less a micro switch with a leaver sensing the pipe wall.

 

[ScottyUK]

Why can't you use a conventional igniter using an electric arc located at the flare stack? This would seem to overcome most of your problems, plus you don't have to have a pressurised system firing explosive projectiles into a flammable gas stream. Does this sound slightly dangerous?






----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[jraef]

"... Does this sound slightly dangerous?"
Absolutely, but also a lot more fun, reminiscent of dropping an M-80 firecracker into your neighbor's garbage can!

&quot;Venditori de oleum-vipera non vigere excordis populi&quot;