Parts Detection in High Temperatures

Possibly a laser beam and sensor. Or, if the parts are a different temperature from the belt, a thermal sensor.

https://www.elprocus.com/optical-sensors-types-basics-and-applications/

This is nice but do you have recommendations as far as which or what brand to use? I think a combination of laser and thermal imaging would greatly improve the process we are currently using.

I don't know what you are using now. What I had in mind are the sensor pairs sold as safety devices for garage doors. They are cheap enough that you could try one out.

https://www.northshorecommercialdoor.com/dimocrunbese.html#video-tab

I don't know if the far infrared (heat) from your parts would affect the sensors, but a piece of glass in front of each sensor would block the heat while letting the near infrared beam through,

I hope this helps.

Hi Craig,

Thanks for your reply.,You can try Ametek Land Instruments or several others. Go google Infra Red Temperature measurement and you will have very good comparison.

With a 9 foot wide opening for the beam to travel you might run into some alignment difficulties. So the transmitter and receiver should be firmly locked in place after adjustments have been made. The size of the objects leaving the oven should be larger than the transmitted light beam to assure a complete occultation of light. However, the smaller the light beam is the more critical will be the alignment of transmitter and receiver.

I would go with a thermal imaging system....precise temperature control is key to quality products...

https://www.lumasenseinc.com/EN/solutions/industrial-materials/steels-metals/making/annealing-furnace/annealing-furnaces.html

Williamson IR is used on hot rolling mills, to measure speed & temperature of ingots. They have the technology to filter intense IR energy to perhaps discriminate between the conveyor and your product. I've seen their equipment installed in aluminum hot rolling mills, to confirm ingot position and temperature, so their application engineers have developed the instruments to stand up in your environment.

These sensors have a wide enough field of vision so you can probably detect the parts from above, by temperature and/or alloy IR signature, rather than height

https://www.williamsonir.com/

Check out this link:

http://entherm.com/Williamson-twowire.pdf

The output is standard 4-20 Milliamp and may be used by a controller,such as speed,temp,etc.

These comments are a welcome sight first thing in the morning. I will investigate and evaluate. Thanks.

Load cells on the conveyor idlers would eliminate the wavelength and alignment problems. Is the idler arrangement such that this would be possible?

Yes I do. Use a reflective sensor across the belt.

Email me for more info.

mike@andersoncontrol.com

Hi, the best possible solution for this is the Thermal Infra-Red Temperature Sensor. There are many reputable suppliers in the market.

It is a fairly accurate device and does not have any physical sensors protruding into your oven unlike traditional bi-metal temperature sensors such as RTD or Thermocouples.

I am looking at two coincidental controls. A beam at the end of the heat zone will be a signal initially to let the operator know a part is ready to quench. A 10 second window exists to go from heating to quench for correct properties to be achieved.

The thermal imaging is desired for ISO standards so we can log each part's characteristics as each piece moves through our process. Exact placement and utilization of the IR has yet to be finalized. We are in development which is why I'm asking questions.

I appreciate your feedback.

Cool. Leave music on the transmit coil. When the volume fluctuates, the part is ready to quench. Let the operator choose the music and optimal loudness. With multiple operators, use headphones if that is not a safety issue. This leaves the two controls decoupled and you will not ruin one adjusting the other. It also gives the operator some boredom relief.

__________

thewildotter

Inverted "T" Humbucker TR Metal Detector Search Coil

I would use a TR (transmit receive) metal detector with a humbucker wound double receive coil in a plane parallel to, and above, your belt. I would use a rectangular coil shape for all three coils. The coil would be a long rectangle (48" long for this belt) with (say 4" diameter) rounded ends. A single loop for each of the three coils would probably work but some designs might use spirally wound coils depending on the characteristics of your part. The transmit coil would sit perpendicular to the humbucker coils and centered between them with the bottom at the level of the humbucker. The humbucker coil pair might be printed or wound on a temperature indifferent(aluminum oxide ?) board. The transmit coil should probably use identical form factor and materials to half of the humbucker and be solidly mounted to the humbucker. One could consider stretching six thin stainless steel cables across the top of the belt up in the air if convenient endpoint mounts can be implemented. Creating the TR coils with stretched cables would be a minimalist approach. The downside to stretched cables is that they may be too sensitive to vibration.

A metal detector signal higher than that due to the belt and surrounding stuff would indicate a part passing under the humbucker. A level threshold could be used to adjust sensitivity for actuating a timed relay for your quench. Just think of the 4" high, 8" wide, inverted "T" assembly crossing your belt as the TR metal detector search coil. The mechanism should be quite adaptable in terms of part geometry but will work better if all the parts are at least tall enough to get near the humbucker inverted "T" search coils and if the belt itself is reasonably uniform and moves smoothly.

If you use stretched cables implementing only one turn in each of the three coils, you may want to run the TR metal detector at a higher frequency than normal TR and it might be possible to implement it with an extremely cheap commodity FM radio transmitter such as an RF link between your IPOD and your car FM radio I have seen advertised. These devices are low power FM transmitters that can be received on a normal cheapo radio receiver placed close to your humbucker wire and otherwise isolated from your transmitter with an Altoids box containing the receiver and an insulated metal plate electrically connected to your humbucker conductor. The headphone jack output could be rectified and compared in the Altoids box to some desired threshold of signal intensity higher than that generated by the belt alone. Drive your transmitter with your favorite music to tune your setup then switch to a constant audio tone for normal use.

Call the rig your "Part Detector Theramin". It is very much like the volume control section generally used in Theramins hybridized with the humbucker pickup coil on an electric guitar.

Humbucker is electric guitar jargon which means that distant (or balanced) signals such as mains power 120 cycle hum null out in the coil but geometrically complex nearby (vibrating string) signals are not cancelled.

I do not know if anyone has already done this and offers it commercially. I would do it myself to gain complete flexibility but YMMV. The RF is kind of like a beam vertically into the belt and vertically back rather than across it. Part geometries and positioning are likely to be important. You may even be able to characterize and perhaps even identify dramatically different parts by their signal strengths and timing. I have been assuming conductive parts or at least parts which affect near field RF.

_________________
thewildotter

I hadn't thought of that.