Pin Installation Machine

[ndzied1]

I would say the escapement is not staying powered on long enough for the pin to get to the pin present switch. Then it times out and the pin can never get to the switch.

So Y6 stays energized because the only thing that can turn it off is a pin at the switch but the escapement has closed :-(.

You may need a longer time delay on the escapement for the first cycle or perhaps the pin is hanging mechanically and the vibration from Y2 being cycled loosens it so it can feed on the next cycle?

Hope this helps.

 

[pstephens]

Thanks, Norm. You could be right, and it would be a simple thing to just increase the T0 setting. But if that is the problem, I wonder why it only occurs at start up.

Paula

 

[elevmike]

Paula,

It looks as if X11 is the primary way of dropping Y6. Are you saying Y6 is on when X11 is on, just on the 1st run? If so you might to add a run to clear or zero out the input register just on the 1st scan, or look check the retneitive ranges.

 

[pstephens]

Hi Mike...

It looks as if X11 is the primary way of dropping Y6.

Yes. X11 is the photoswitch that "looks" through the nosepiece to detect the presence of the pin. So, Y6 will keep air blowing through the tube until the pin arrives in the nosepiece.

The way it *seems* to be functioning is like what Norm said -- that the escapement (Y5) is not allowing a pin to drop into the tube, for whatever reason. So Y6 just keeps blowing, waiting for a signal from X11 that a pin has arrived in the nosepiece.

When the system is powered up with a pin in the nosepiece (as would be the normal situation), everything works like it should. But if there is no pin already in the nosepiece when the system is turned on, then the escapement apparently will not release a pin. As I mentioned, manually cycling the driver Y2 (triggering the solenoid valve with the tip of a ballpoint pen) seems to correct the situation.

...you might add a rung to clear or zero out the input register just on the 1st scan, or look check the retentative ranges.

Could you explain further how to do this?

Paula

Edit: I'm thinking that a good way to troubleshoot this problem might be to temporarily increase the timer values in the pin-feed sequence so that I can better see what's happening, while it's happening.

 

[geniusintraining]

This might work, it would drop a one on the first scan, if it was not there...

Edit: branch was in the wrong place

 

[rsdoran]

Personally not sure why you need a timer for the escapement, just use pin NOT present to energize that valve, when pin IS present the valve will de-energize.

Then use PIN Present to activate timer for blow valve.

 

[Lancie1]

Paula,

Having studied your program, the problem is that:

After the program is active for some time without a START (X0 being pushed), Timer T0 times out and blocks Y5 in Rung 1; therefore no new pin can ESCAPE.

Solutions:

1. Add some input that tells the machine, "Hey, its time to wake up and Reset T0 so that a new pin can be put into place". If there is nothing to tell the machine to reset T0, then you could possibly use a timer with a long time setting to reset T0 after say 15 minutes, or 2 hours, or whatever you think appropriate.

2. You said: "Whenever the control system is powered up with no pin in the nosepiece, the pin-feed sequence does not work. The Pin Blow valve (Y6) is energized, and just stays on, as though the Escapement step is bypassed." The key here is "Whenever the control system is powered up" you need a contact in Rung 1 that is on when "control system is powered up" and off when it is not. Call it "Control Power ON". Then when the Control Power ON" contact goes off, Timer T0 gets reset to 0 and will be ready for the next first cycle.

Without some On/Off contact in Rung 1, once T0 times out, your Escapement is disabled by the T0 NC contact, and the Y6 Pin Blow turned on continuously, until X11 sees a pin present. Your program works by assuming X11 goes ON then back OFF in a repetitive cycle. When X11 just stays off (because the control power is off and machine not running), then it does not work correctly on the next first cycle. A work-around is to cycle the E-stop Off and then back On.

 

[Mickey]

Paula,

Having studied your program, the problem is that:

After the program is active for some time without a START (X0 being pushed), Timer T0 times out and blocks Y5 and Y6 in Rungs 1 and 2; therefore no new pin can be blown in.

Solutions:

1. Add some input that tells the machine, "Hey, its time to wake up and Reset T0 so that a new pin can be put into place". If there is nothing to tell the machine to reset T0, then you could possibly use a timer with a long time setting to reset T0 after say 15 minutes, or 2 hours, or whatever you think appropriate.

2. I haven't thought of another one yet.

Solution #2, see #1

 

[Lancie1]

Paula,

I now agree with Ron. Timer T0 is probably not needed and is causing your start-up problem. Try the attached version, using (X2 Part Present and Not X11 Pin Pressent) to trigger the Y5 Pin Escape solenoid.

 

[Lancie1]

Rethinking the Pin Problem....

Paula,
I went back and found your description of the pin feeder. You said:

The sequence to feed a pin is quite simple: A solenoid valve is triggered (approx. 1/2 sec. duration) to actuate the escapemant mechanism. This releases a single pin into a chamber. After a short delay (approx 1/2 sec.), another valve opens to blow the pin thru a tube into the driver's nosepiece. LS8(X11) senses when the pin is present in the driver, and shuts the valve off.

Because the solenoid feeds a single pin each time it is energized, then it cannot be left on continuously as Ron and I asumed above. When if fails for whatever reason, a sure fix is to add a REPEAT Timer T2. Set T2 for the time you want to wait to see if the first Pin feed works. If not, then T2 will cause the Pin Feed sequence to repeat until it sees a pin. Add T2 to Rung 1, and Insert the new Rung 3. That's all, the rest is your original program.

 

[pstephens]

Thanks very much for the replies!

It's been a while now, so I'm not completely certain. But I believe that when the bowl manufacturer was demonstrating the operation, he used a momentary signal on the escapement, and then triggered the blow valve. I know that some escapment mechanisms, like this one for example, require two discrete moves to release a part. On my customer's feeder they may have used a simple slide block, exactly as thick as the pin is long, to move the pin to the air stream. In which case the escapement cylinder might not have to be retracted to release a part. I would need to do some experimenting (or simply call them up) to find out for sure.

At any rate, you have given me some approaches to think about and try out when I go back next week. I am extremely grateful for your help!

Paula

 

[pstephens]

Having thought about this further, I don't think that the problem has anything to do with how the escapement is triggered. After all, the 4 main rungs of this program were lifted directly from the old program, which worked fine for the last 5-6 months. I think the problem has more to do with how the sequence is being initiated.

The goal with this modification is to reduce cycle time to the absolute practical minimum. Therefore, I don't want the pin-feed sequence to take place between the time the START button (X0) is pushed, and when the driver (Y2) begins to extend. I want to utilize that "dead time" when the operator is loading/unloading the part. Hence my thought that the pin-feed sequence would be "demand-initiated": Is the machine on? Yes. Is there a pin in the nosepiece? No. Then, send one NOW. Typically, this would happen when the driver has retracted just far enough, after having pressed a pin, for X11 to become true. Thus, the period of time from when X11 just goes true, to the time that the operator has removed the completed part, loaded a new one, and pressed the START button, is available for the pin-feed sequence.

Another requirement is that the program must be "patient". This gets to what some of you were saying about the timers timing out before another cycle happens to reset them. This operation is only one of several that the operator performs in this work cell. The time between cycles of the program might be as little as a couple of seconds, to as much as several minutes (or even longer, if they don't shut the machine down), so the viability of the program cannot hinge on a new cycle being initiated at some minimum, regular interval.

All of that said, I don't know exactly why the pin-feed sequence is failing at start up, when it works fine otherwise, but I think that it might be because of how the sequence is being initiated. I should probably not be using an E-STOP button contact (X1) to start the cycle. The way it's set up, the E-STOP interrupts power to the MCR (and the PLC outputs), and a RESET button must be pushed to pull the relay back in. HOWEVER, the PLC sees the X1 contact closure immediately. So, there is a delay between the time that X1 goes true, and the output commons are activated. This is what I think is causing the problem at startup, since the timer is triggered when X1 goes true, but the escapement output is not enabled until a few seconds later. It also explains why manually cycling the driver "fixes" the problem, as X11 is reset by the driver shank interrupting the beam. Only, this time the output commons are activated, so the pin-feed sequence happens the way it should.

I think I need to change how the pin-feed sequence is initiated, perhaps with a separate RESET button cantact?

Paula

 

[rsdoran]

I know I did some of this in the beginning but had to go back and review.

Personal thought, nothing more, you are doing something I would never do i.e. use a latching STOP button to power-on and start up. An ESTOP should never be used as a reset.

The best thing would be to use a START button with the (E)STOP button. In an earlier post I offered example 6 of my program, the main difference between it and what you have displayed is that rung 1 initiates a start cycle. I think if you had this START CYCLE rung then it will work as expected, of course you can remove the drill part as needed. This will require the (E)STOP to be pulled out and the START button to be pushed, all this should allow time for everything to work as expected.
http://www.plctalk.net/qanda/showpost.php?p=155301&postcount=61

I was going to modify my version to remove the drill sequence but the index part confused me, will have to go back and read more.

 

[Lancie1]

Paula, The problem is that you really need to know if a pin is in the dispenser BEFORE you trigger Escapement. That may be impossible, so you are left with the Photoeye that sees the pin after it is blown into place.

Just because it worked before means very little because you are dealing with a different situation now. Its like the old PLC saying: "Making it do what it is supposed to do is easy. Keeping it from doing all the other things is the hard part.

I think you are correct about the Start-up problem. If T0 times out before the MCR is energized, then the Pin Escapement WILL hang up. Your logic has a teeny-tiny flaw that only allows continuation of PART of the Pin Escapement Cycle, the Air Blow. Once T0 times out, it stays timed out until X11 closes (pin present). If there was no pin in the dispenser to dispense, then X11 never closes, so the pin dispenser is hung up.

You could solve the problem by hard-wiring a contact from the physical MCR Relay in series with X1 on Rung 1. Alterrnately, you can wire an MCR contact to a PLC Input, and use MCR to inhibit the t0 timer, which is the usual solution.

But I think my new timer T2 will solve your problem, simply and easily and without adding new contacts or switches. If there is no pin, it keeps repeating the escapement pulse until one shows up. It should only have to repeat 1 time for the FIRST pin, if there are pins in the bowl feeder. As long as pins show up on time, the Repeat timer will never be triggered. Actually this would be more fault-tolerant than your original version. If a pin temporaily hangs in the feeder, this new Timer T2 will allow the pin to eventually be dispensed. In your original, if it does not get dispensed in the first 0.4 seconds, for WHATEVER REASON, then it will be hung up until someone cycles the system.

That brings up another prolem: if the bowl is empty, it would keep repeating, trying to dispense a non-existent pin. It would keep going forever, just as your Pin Blow does now. You should add a Counter that counts the number of times the new T2 Repeat Timer bit is enabled, and after 5 times, triggers a blinking alarm light that says "Bowl is Empty". You probably have some spare lights on the old drill panel (the old Fault Lamp, Y10 Output, perhaps?), that could be relabeled and reused. You may not even have to change the wiring to the existing light.

 

[Lancie1]

I should have said above that if Y5 gets energized before the MCR is energized, then the Pin Escapement will hang up.

Your logic has a teeny-tiny flaw that only allows continuation of PART of the Pin Escapement Cycle, the Air Blow. Once T0 times out, it stays timed out until X11 closes (pin present). If there was no pin in the dispenser to dispense, then X11 never closes, so the pin dispenser is hung up.

You could also solve the problem by adding a timer that delys Y5 Output until the MCR relay has had time to energize.

But I think my new timer T2 is still the best solution