Industrial Metal Detection Ideas?

"Christopher Ott" wrote in message news: snipped-for-privacy@giganews.com...

Hi.

recyclers. Here is a link to a photo of a typical

companies. They run debris through a rotor/grating

other applications (compacting roofing shingles for

landscaping suppliers who turn yard debris and slash piles

Axes, sledgehammers, rebar, railroad spikes,

the rotor and blow out bearings. This causes several

rotor. This is complicated by the fact that the hopper

Non-metallic drag belts are far too fragile and are not

into the hopper does tend to confuse traditional

and drag chain made the output virtually worthless.

safely pass through the rotor and be picked out with

Sometimes (like with a railroad spike in a tie) the high

load can be reversed and dumped out the back to be

simply senses the hit milliseconds before the

before. Perhaps in a different application which I can

ideas? I'm open to any input...

I once developed a metal detection scheme that I believe could be adapted to your requirement. It would need some post-processing of its output to handle the varying aspects of the metal legitimately in the equipment. But it can readily be made to ignore constantly present metal and respond to changes with good sensitivity and low drift. The changes could, I believe, be processed together with chain movement information to filter out the changing component due to the chain links moving through the detection field.

The basic idea is pretty simple, amounting to a way of detecting minute changes in the portion of an AC magnetic field reflected from targets, where the excitation field is produced by a fixed arrangement of coils.

Please feel free to contact me if interested in a detailed discussion of this. I'm in the Seattle metro area phonebook.

Hello Christopher,

It seems that you would need a directional RF field that is applied on one side and detected on the other. The steel hopper should not pose a problem if the antennas are mounted in a flat and sturdy housing that is sleek enough not to obstruct the passing materials, or recessed. The drag chain could most likely be detected and ignored via a pattern recognition algorithm. As an added benefit you might even be able to detect damage (irregularity) to the chain before it is beyond repair.

The loader bucket is a problem but for this scheme to work you would have to restrict loading to the far end anyway. Else there wouldn't be enough time to stop the diesel or for the spin-down if using a clutch. So you could mount the RF scheme just past that point. If you don't, chance are someone drops a bucket tooth or sledge hammer piece just beyond the sensors and it goes in undetected.

Distinguishing between a soda can and a bucket tooth will be the real challenge. If you can't live with such false alarms this would probably require some ultrasound.

Regards, Joerg

Look at the mining industry which has exactly the same problem.

I have seen a setup in a old tin mine which actually went the other way and used large electromagnets to trap ferrous junk (on the poles) before the rock crushers.

Could you run the incoming feed over a pair of pole pieces which would both attract any ferrous scrap and (by means of a small AC current riding on the DC supply) detect the change in inductance when the ferrous scrap reduces the air gap?

Just a thought.

Regards, Dan.

Dump incoming material into a tank of hydric acid first and agitate it. The hydric acid attacks the steel and causes it to separate from the wood. The wood can then be retrieved and put into the recycler. Unfortunately, I understand that for environmental reasons, hydric acid is difficult to come by in Arizona.

Matt Roberds

Make the whole machine out of Inconel?

Good Luck! Rich

Hydric acid is extremely dangerous, especially when in concentrated form. It is safe to drink if diluted with ethyl hydroxide.

Aah, the dreaded dihydrogen monoxide.

Rene

I read in sci.electronics.design that Rich Grise wrote (in ) about 'Industrial Metal Detection Ideas?', on Sun, 1 May 2005:

You were given the answer in the post about 'hydric acid'.

This is the solution I'm currently looking at. An array of low power RF transmitters on one side pulsing in sequence, and a receiver array on the other side looking for shadows. The challenge is synching the TX and RX to a very narrow time frame to ignore any signal reflections.

Unfortunately this does still allow tin cans to fake out the system. How would you use ultrasound to distinguish them?

Chris

This would require a large swimming pool of acid be kept on site at thousands of yard waste companies worldwide. We're talking about huge volumes of debris being processed daily. One machine can fill a large semi trailer in under an hour.

It would be quite a sight however, seeing hydric acid drenched mulch melt a whole truck...

Think harder Homer...

Chris

Given the current cost of steel, it might actually be cheaper!

Chris

Is it possible to safely x-ray material outside of an enclosure?

Chris

In message , Christopher Ott writes

(snip)

How about x-raying it. Video the fluorescent screen and use a threshold detector on the video signal to pick out the solid objects.

Maybe if you looked up "hydric acid":

Cheers! Rich

Oops.

Regardless of my failure as a chemist, separating large debris piles in water is not practical. This would add a considerable amount of time to the process and the added weight of the processed mulch would reduce the amount which could be trucked around. It would have to be dried first adding even more complexity. We can already find metal by pre-processing the material through a separate detector. What I'm looking for is methods of detecting in-process on the existing machine.

Hydric Acid.... (smart ass)

Chris

They seem to do it at the airport.

Cheers! Rich

I read in sci.electronics.design that Christopher Ott wrote (in ) about 'Industrial Metal Detection Ideas?', on Mon, 2 May 2005:

No, YOU think harder. 'Hydric acid' is one of the humorous chemical names for WATER.

Separation by flotation is a very well-known technique for the sort of operation you are involved in, from coal mining to gold mining.

Hello Christopher,

Time domain will be really tough. I'd start with reflectivity measurements, very high frequencies etc. But you would have to stay in an ISM band, else the goons show up some day ;-)

I am not very familiar with ultrasound through air. My home turf is medical ultrasound. There we can echo through thin walls whereas hard calcifications are tough to penetrate. But we are able to see, for example, two layers of calcification, something that can indicate a dangerous condition in the medical world (if one flies off it can cause an obstruction and that can lead to a stroke).

I'd research this in the NDT field, they deal with that all the time. In the soda can case it would boil down to detecting the cavity. IOW, if the beam passes metal and there is no detectable cavity then the probability that it is a large and dangerous chunk of metal is high.

You probably also need some magnetic field measurements. It's not only the occasional soda can but also a hard fresh piece of manzanita that could fool the system or a full water bottle that got dropped in there.

Magnetic fields can't really be realized with loops in this case because of the loader bucket. You need a more defined field such as the one between separated transformer core halves.

Another thought to keep in mind is materials. Teeth from a bucket will usually be of a good quality steel whereas soda and beer cans are mostly aluminum. Unless someone throws away a Campbell's soup can but that's rather rare these days except maybe around clandestine camping areas.

Regards, Joerg

Hello Christopher,

OSHA will want to have their say here. Yes, airports do it but these are systems that are enclosed and have gone through a whole lot of agency appovals (one reason why they are so expensive).

It is also done in the canning industry and other places but not in areas where people are around all day. If I understand your system correctly it might be necessary to reach in at times to dislodge a jam etc. Then there is the guy riding the bucket tractor.

Regards, Joerg