Need Help Detecting Ground Beef

All
trying to detect ground beef in an auger/incline screw. have tried ultrasonic sensors but it seems i cannot get a stable signal, possibly due the the complicted angles of the SS hopper and the blender running.

was thinking a capacitive prox might work in the side of the hopper, has anyone ever tried this type of sensor on ground beef?

also had laser level sensor but the face has glass on it, i cannot knowingly put that over a hopper full of meat for fear it may someday fall in and not be picked up by the metaldetector.

been told radar would be the way to go but their output is analog and i need a discrete pnp output.

thanks in advance

Can it be achieved by monitoring the load on the drive?

edit: just seen your last bit about a descrete signal, although I believe you could buy something to give you a descrete signal at certain loads.

http://www.loadcontrols.com/ has a load monitoring switch with a discrete output.

Also, even thought its not discrete, will load cells under the auger mounts work?

REd Lion, API, Action Instruments, and a number of other sources make limit alarms that will trip on an analog signal above/below a certain setting.

If you are using a PLC you could also takethe analog into it and use comparisons to trip an alarm.

The temptation to say "install a small tiger and monitor the purrs" is overpowering - however I will refrain .

I like PeteW's idea , with a VFD , particularly with digital comms , and a bit of creative "adaptive" learning software , then this would work well - however , reading your post , is that what you want to achieve ? is it permissible to run the auger dry ? , and are you looking to sense product on the infeed or the outfeed hopper ?

not sure the load on the motor changes all that much, but i will look into it.

Some more background since its hard to answer if you cant see the aplication.

we dump from a large combo box ~2000lbs of ground beef into a large hopper with an agitator/breaker. large meat chunks are then broken up into smaller pieces where they fall into a 10" diameter incline screw. when the portioning machine downstream calls for meat this 10" screw starts. at this point we add a herbal preservative that has the consistancy of motor oil at .2% of the total weight of the meat. (98.8% meat .2% oil)

currently we add the oil when the 10" screw is running and as you would expect the screw will run out of meat and continue to dose oil onto nothing. this is where i would like to add a sensor to detect the indeed is meat there and permit the oil to dispense.
see attached .pdf sent to a local sensor company when i was testing the laser sensor.

thanks

I too think the best method would be using load cells and apply the mix based on material fed.

You might be able to use capacitive or photoelectric sensors, some photoelectric are programmable for range. If you know the screw speed then you may be able to calculate time between flutes and use this to compare the time sensor input is on/off. In other words the time sensor is on (or off depending on what type sensor is used) when meat is detected between the flutes, compare that time with the time between flutes when empty.

If possible, depending on situation, would use multiple sensors and compare along with something between hopper and screw to detect material being fed.

I have an odd suggestion.

Banner makes sensors that detect color differences. I have never worked with these sensors but I have seen them work on a box sorting application. Just a thought.

I would second Jimtech67's idea but

How can you see meat in the auger? I would use sound. If you fill a glass with water you can tell how full it is by sound. I bet I could put meat in a class and tell how full it is too. The trick would be to find the right frequencies that would provide the most accurate information.

Smithfield Foods?

If you can't sense the difference in the auger, I bet you could sense it on the large mixer, then just put a time delay on shut off for the dosing.

Does the dosing take place at the mixer and mix through the auger? Otherwise it may be easier to put some type of motion sensor on the other end and sense the meat dropping into whatever it drops into.

I've used banner ultrasonic on frozen meat in mixers, and just set the low level to just above the auger so the signal wouldn't get screwed up.

I've done the same in the past as some mention above, where the hopper is on load cells and you control the feed into the auger on loss of weight, you don't feed out until you finish the mix cycle.

Putting the auger on loadcells is also feasible.

Depends how much money you have really I suppose.

Sensitive capacitive level sensors will detect presence of meat with its high water content through air or plastic. Capacitive will not sense through the steel shell, though.

I've used Siemens CLS capacitive probes on field corn and gotten detection at 1" distance, through glass. The higher moisture in meat would trigger at a greater distance.
https://pia.khe.siemens.com/index.aspx?Nr=4922 or
https://pia.khe.siemens.com/index.aspx?Nr=4924

I'm not sure how the probe would react to the periodic presence of the auger blade, though.

It wasn't clear what sort of ultrasonics you tried. Ultrasonic level/distance? Or ultrasonic/acoustic steel-carries-ultrasonic-vibrations-whose-amplitude-varies-with-the-load?

The latter is used extensively to sense flow/no flow in pneumatic conveying, and might work for meat/no meat in an auger. The Siemens AS100 is an example (it works with a module that "alarms" with a discrete output)
https://pia.khe.siemens.com/index.aspx?Nr=5875

Dan

Keyence has some really good quality colour and high power photoelectric sensors. They are more expensive then other sensors but when you have a difficult application they usually have something that works the first time.

Get your suppliers involved to bring in their sensor kits and prove to you that they have a solution.

d

no good yet

Installed a P+F capacitive prox in the side of the auger shell. Prior to doing that I Did some bench testing that worked well, knocked a 1.5" hole in the side of the auger, welded a coupling on the auger, threaded the "teflon sensor well" into the coupling, threaded the sensor into the sensor well, powered up and turned the sesitivity to its lowest point and all be damn if i could get it to go low. Always in the "ON" state.


Spoke with P+F and have to application eng's stumped. they are sending a field guy.

here is the odd part, centering the sensor in the 1.5" hole and passing it through the hole into the auger it doesnt change state. when i put it in its well as soon as i get near the coupling the thing goes high. almost like the teflon well is acting as an amplifier.

any ideas?

All my experience is with capacitive level sensors, not capacitive prox switches (process, not discrete). I would expect a level sensor to trip when enclosed by a pipe coupling (although I've never done that), because the sensing is directed "to the sides, omnidirectionally" as well as "out the end".

My recollection of capacitive prox sensors is that their design is to read the 'face' or read-out-the-end of the sensor, so that you would expect a capacitive prox to work with its body enclosed by the coupling or the teflon well.

Have you mounted the prox in a coupling before, and had it work OK?

Not to be disrepectful, but are you sure the sensitivity adjustment was in the right direction? least sensitive, not most sensitive? (Maybe the Germans do it backwards?)

Dan