Sensor - Bran /wheat bread factory

[ndzied1]

Eric wins! Good idea, sometime we tend to put too much thought into simple problems....

... Except for the possible stickey mess when the double dough sticks ot the switch wand and gets dragged over the next 10 baking pans. Don't know if this is what would happen or not but I would check it w/ a quick test. Maybe a teflon coated arm would help.

Also, if the pans go by too fast, the arm might not have time to fall down between sides. This would also be easy to check for.

If those issues prove not to be a problem, then yes, this solution is most elegant (and most excellent as Ted & Bill might say )

 

[Terry Woods]

"Time to Check the Donuts!"

So, you are a computer and today yer makin' dough! (The eatin' kind, not the spendin' kind.)

You're told that a limit switch will tell you if there is more than one ball of dough in the bucket.

So, the process begins. You send your apprentice out to read the inputs and report the status (what is the plural form of status? Stat-eye?) to you. He says, "It's OFF". Next time he says "It's ON". Then he continues to report that it is ON for the next several reports. Then he says "It's OFF". He continues to report that it is OFF for the next several reports. Then he reports that "It's ON".

So... Was there one ball of dough? Or were there two?

As it stands, the limit switch ain't telling me much. Of course the limit switch will come on when it detects the edge of the first bucket. Then what? When the limit switch goes off is it in the first bucket? Or is it in the second or third bucket?

You don't know!

You need more information! The single most important piece of information you need to know, even more important than the state of the limit switch, is... "What should I expect, and when should I expect it." You need to KNOW when the limit switch should be OFF. If it's not OFF when you expect it to be OFF, then you KNOW you have a problem.

You need to know when the bucket is centered under the limit switch. It might be the case that the bucket needs to be slightly after center - it depends on the length of the wand.

If the speed of the bucket rack is always constant then you can use a timer. In that case, the limit switch should be ON for no longer than x-seconds. If it's ON longer than that then you know you have a problem. You also have to ensure that the wand is not too long.

If, on the other hand, the speed of the bucket rack might vary, then you can't use a timer unless you want to do all of the math calculations (dynamically) that are required to determine the appropriate timer values. This can certainly be done without too much hassle but you'll need at least one additional sensor or some way to know the speed.

The easiest way is to use a sensor to indicate "The Bucket is Centered" or thereabouts. You need this information for any method that you finally decide on. Whether by sensor or speed calculation, you gotta know when it's time to check for the double-dough!

You indicated that there is a 15-mm gap between buckets. I suspect that this is the gap at the upper end of the bucket. I expect a larger gap at the lower end of the bucket, as shown in Eric's drawing (straight sides would make it harder to remove the bread, I should think).

You should be able to install a pair of photo-eyes (transmitter/receiver) or you could use a bifurcated fiber optic sensor which combines the transmitter & receiver in one unit, to look for the gap between the buckets.

Whether slanted or not, 15-mm is plenty of room for an eye-set to detect the gap (assuming proper alignment).

When that sensor "sees itself" then you know that the bucket is in such-n-such position and it's "Time to Check the Donuts!" (the status of the limit switch).

Actually, at this point, I would remove the limit switch and use your original diffuse photo-eye tuned to 90-mm.

 

[ianingram]

I too work in a bread plant

What Eric has drawn is 100% correct

We use a simple limit switch, because a "double-dough" will rise above the height of the baking tin, once the tins have been through the main prover (@ 42° C and 82% Relative Humidity).

Looking at what has been written here, Ramil may be trying to detect a double dough at a different stage of the process.


The simplest ideas always work the best!

 

[ScottC]

Yep, I agree, Eric wins!

 

[Terry Woods]

In his initial post, Ramil said this...

Now suppose two dows are present in one bucket and if that enters oven without our notice . then we have a big problem to remove it as it is backed and becomes big sized.

So, it's clear that he wants to detect the problem before the dough enters the oven.

If Ramil's situation includes what Ian says...

We use a simple limit switch, because a "double-dough" will rise above the height of the baking tin, once the tins have been through the main prover...

then, both Ramil and Ian can use a Tx/Rx Photo-Eye across the top of the bucket so as to maintain that "untouched by human hands (or other unnecessary things)" quality.

If you have a maintenance department where everyone "knows" their own way to fix this or that problem, and they have their own "right way" to tweak it... then you end up with a certain amount of chaos.

In this case, that could include the switch being moved here instead of there, or the wand being just a bit too long.

Then there is the wear and tear on the wand. If the wand is in an out of the way area then who would know that the wand has worn down to the point of being useless. Of course, you could use a stainless-steel wand - and leave a drag-mark across the loaf of those doughs that are just a bit too high (maybe they didn't lay exactly right in the bucket).

All-in-all, for any sensing problem, I believe that a "no contact" method is inherently better - on the sensor and the sensee(?).

And again, I fall back to my question about when should the input be OFF? And how do you know when it should be OFF? The limit switch does not know the difference between the edge of the bucket and the extra dough. Once the limit switch detects the edge of the bucket, if the switch does not go OFF (because of an extra dough) then the limit switch thinks it's still seeing the edge of the bucket... no?

Just a little bit more hardware gives a "positive" Go-NoGo.

Yeah, yeah, yeah.... I'll put a sock in it!

 

[Eric Nelson]

No Sock Required...

Terry, you keep beringing up the point about how to know WHEN to look at the switch, and I keep saying "Yeah, so?"...

Now that I went back and re-read Ramil's second post, I understand your point. Notice the "At this point, you check the limit switch" part of MY post. I foolishly assumed that he already had some sort of "interrogation" signal (For some reason, I prefer the term "gate" signal). IOW, a point in the cycle where the 'height sensor' would get queried. It seems that he does NOT!

Yes, you definately need a "Check Sensor NOW" switch. There might be ways around it, but just adding this sensor would be the simplest (and most reliable IMO). IOW, finding a sensor that will R-E-L-I-A-B-L-Y detect 'extra' dough, yet NOT see the bucket edges might be tough! Some here will remember I recently went through a similar exercise. See THIS THREAD (which, by the way, is installed and running great!)...

I think we can rule out the "look across the top of the bucket" method. Great idea, but Ramil specifically states "If two doughs are present it will exactly cover one bucket hieght". He didn't say "will exceed the bucket height"...

I think the 'best' solution will be the one that not only works, but also fits Ramil's budget! Give me an unlimited budget, and I can come up with some ingenious ways to test just how 'unlimited' it really is!...

beerchug

-Eric

P.S. Ramil will be the one to declare who "wins". If he just sticks a human there to take out any 'extra' dough, we ALL lose!...

 

[Ramil]

Thanks

Eric you are right. At present we have our technician sitting there permenantly to remove that extra dough.

We tried some methods and finally we had two cases where two doughs went into the oven without notice of our sensor or i had not tested the sensor method properly.

I had a difficult time answering our production people. I just want to avoid that again. So until we solve that problem "unmanned" we will have our technician there. Production people refused to put their person sit there.

Our guy does a lot of job there. Sitting there one day myself found that there are so many jobs to be done at the conveyor itself.

I will list you what job our technician does there.
1. to remove double dough
2. some dought might be flat, he has to make it round
3. reset our own sensor alarm(double dough)

money - no i donot have that much money/budget either. But could request our prod. guy to release their budget for us.

thanks agains

 

[Eric Nelson]

Re: Thanks

At present we have our technician sitting there permenantly to remove that extra dough

Look Terry, someone took your "BE THE PLC" mantra literally!...

beerchug

-Eric

 

[BillCassens]

Ramil:
Hang two sensors above the buckets, sensors should be spaced at half the diameter of the bucket. Wire them in series, adjust sensitivity to trigger when height is higher than one dough. If both trigger at same time there are two doughs in bucket. If only one triggers (bucket edge) nothing happens.

What a great idea! No timers, no position sensors, no programming, no contact with the product, only 1 input. Can't wait to try this at work and take FULL credit for it. Thanks ed

 

[shooter]

eric i like the idea

However i would use 6 sensors or for every bucket separate
Did you ever look at the entrance of a tunnel with these strange wireframes above to detect height of lorry?
on the end there is only one switch on a rail.

I would make this with six together on a small frame that moves down into the buckets. If one bucket is full the switch (normal microswitch not very expensive prox. ) will be activated.
A blind person feels it same way just puts his finger in and when touch it is full.
I would put this setup right in front of the oven to detect every thing unusual like to buckets on top, or dough too high.
so have a sensor for maximum height also on it.
if buckets are always same size put in movable frame above ovenentrance and lower it every time one shift goes in.

 

[jdbrandt]

think both ways before crossing

This problem, and many like it, has been solved before, and
a commercial solution is available (email me if interested..
this forum is non-commercial).

Granted, doubles into the oven are bad, but, they are even
worse between the de-panner and the stacker.

Worst case, they sould never get to the slicers if your
foreign-object detection is working (high density trip).