Staging of boxes one by one

[vivrocking]

Hi all, I wanted to stage standard-size boxes on a staging belt one by one with a uniform gap between the boxes. The staging belt will have an encoder to track the belt movement. The box will enter the staging belt from another belt, the staging belt will start and run for a certain amount to make the box completely moved onto the staging belt, and then the belt stops. When the next box comes, the staging belt will start again and stop once the boxes get transferred to the belt. What do you guys think is the best way to write this logic?

 

[kalabdel]

Sensor?
You will need a minimum of one sensor and then you will need to expand into what happens when the box arrives.


Am example, box enters staging belt and is detected by photo eye:

If it's the first box on the belt so there's nothing ahead of it then the belt can keep going until it clears the eye by a certain number of counts and then stops waiting for next box.
Next box arrives, belt moves and so on.
Is that what you want to happen? Are boxes going to pile up at the belt before it? Are you controlling the other belts movement too, on what basis?
What happens to the boxes being staged, are we tracking them for ejection?

 

[drbitboy]

I assume this is not course homework. If it is this should get you a C-.

There are many ways to skin this cat; here is a clumsy one implementing more or less what @kalabdel suggested. The gap from the back edge of a lead box back to the prox (and to the back edge of the next box) is roughly proportional with the counter preset. There may be an offset if the conveyors run at different speeds, but as long as they are consistent the gap can be controlled empirically by adjusting the counter preset.

This uses only the pulses from the encoder and assumes no pulses are missed (each pulse cycle is at least two scan cycle times in length). Something more sophisticated could be done if the pulse frequency is higher than that with the encoder connected to the PLC's HSC and incrementing an ongoing number for the PLC to use.

​

 

[Jlandwerlen]

Sounds like a smart belt application. Search for that and you should get plenty of information. Rockwell has sample code, which may be useful, however, it's motion application.

 

[kalabdel]

Sounds like a smart belt application. Search for that and you should get plenty of information.

Why did you have to be so helpful

Now I have like twenty documents and 50 videos I want to read and watch. Why , why.........


Thanks, taht was very helpful. The right search term and bingo you got more information than you ever though exist. :site:

 

[parky]

What accuracy do you need, this is possible without an encoder a couple of sensors like DR has shown, however, If accuracy is really needed an encoder is fine & very simple you detect the edge of the box reset the encoder count wait till the encoder reaches a pre-set value, done, however, if boxes vary in size then again pretty simple, when a box leaves the photocell reset the count and then when the count reaches your pre-set level stop the belt. there may be other factors like what if the conveyor stops before the gap has been reached i.e. manual stop or buildback from futher down the line still pretty simple.
You have not said what PLC that would help

 

[drbitboy]

https://docplayer.net/43275151-Technical-application-notes.html


does that help?

 

[parky]

You doing the night shift Brian ?.

Here is an implementation of a box separation, assuming here that there is an inverter not a servo, and a PE at the infeed of the belt, not bothered with things like start/stop etc. that is obvious.
This works based on the principal that regardless of box size the gap is the same between them, On PE detecting the front edge of the box, the conveyor starts, when the box leaves the PE the HSC will be reset & start the new count for the gap required, note the slow set point just in case it might be required for more accuracy, so let's assume a gap of 100 counts is required, then at 70 the belt slows down, when it reaches 100 or above (just in case a pulse is missed during normal scan (assuming the count is only used within the normal scan program & not in some sort of interrupt)), the belt will stop & wait until the next box, so if the boxes are the same size or even vary the gap is the same, could reduce code but tried to make it a little simpler to follow (that ones for you Brian).

 

[arpus4KM]

you could go incredibly simple with no need for an encoder either.

Part comes from belt 1, hits photoeye 1

belt 2 takes off at the same speed as belt 1 until photoeye 1 is unmade.

belt 2 remains on for X time, then stops.



repeat..... all boxes regardless of length will have the same gap.

the only time you would need an encoder would be if you want to precisely position the boxes based on their sizes and the length of the belt... like staging boxes of different lengths at different locations on the belt until full.

 

[nickfury23]

you could go incredibly simple with no need for an encoder either.

 

[kalabdel]

This is what I searched for and there were so many hits of related and unrelated but still interesting that I thought it would probably take me a week to go through them.


ABB has one
Nidec has one
Siemens has a couple

Then all sort of blogs with various conveyor control


And there's youtube, I have never seen that ARB belting before.

 

[kalabdel]

you could go incredibly simple with no need for an encoder either.

After all these years and the bots still as stupid as ever, all they could they repeat.

 

[drbitboy]

@IanM8040: that's inspiring!

you could go even more incredibly simple with no need for a timer either; just calibrate the mean scan cycle time

​

 

[parky]

There are merits for most of the above replies, however, the OP did state that an encoder is on the cards.
Let's look at it in a bit more detail, although little information has been given so this is my take on it.
Using a standard timer:
Great, may work reasonably well, however, timers have a drawback (assuming standard pulse or delay off/on), if the belt stops the timer will either continue to time out or re-start (depending on the logic & timer type) so gap may not be consistent, are a little un-reliable regarding accurate timing, if the VFD speed is changed would have to be re-calibrated.
Using a counter or incrementing a variable based on a time interrupt or other:
Will work reasonably well, Can stop the count on belt stop but even a small delay i.e. ramp of VFD will cause errors, again speed ramp down or change of speed will have to be re-calibrated.
Encoder, simple star wheel & prox or actual encoder:
Consistent regardless of stops, ramps etc. generally more accurate as the pulses are consistent with movement of belt.

Finally, another possible problem is if the boxes bunch up on the infeed conveyor this may be due to operator loading too quickly or a box snagging on conveyor sides so the next box then bunches up releasing the two together, so there may not be a gap seen by the PE, simple solution is to ensure the metering belt runs at a slightly faster speed to separate the boxes.
So in my opinion, For accurate metering the latter two is the best option on what information supplied.

 

[parky]

Seems like people like reducing code, here is a two liner using an Encoder (minus any encoder settings).
On a falling edge of the PE i.e. Box leaves PE, the HSC is reset (note: this is a pulse as HSC may only reset on a pulse not on being covered).
The conveyor is started on the Box front edge covering the PE, this is latched in by Belt output, the less than count is also bypassed during PE being covered by box as this count will increment & box length would probably be longer than the number of counts, when box sees trialing edge of box the count is reset & the compare then takes over to stop the conveyor when the count has been reached, however, there are sone caveats, The box spacing on the infeed would have to be greater than the gap required or the infeed would need to be stopped or with a little more code, the speed of the belt would have to be higher to gap the boxes enough for the required spacing.