Thank you for your input. One last question concernign timing. To make sure the safety timer U1 always starts when the output signal goes high, I should be able to move the trigger input to U1 to eye2 along with the pull up resistor. then add an inverter from eye 1 to feed pin 4 (reset) on U1. Would this make sure U1 is always reset and ready to start timing from zero when eye 2 is active?

Thanks again, you have been alot of help. any suggestions on where to purchase the cd series chips for reasonable price and delivery. No one around me carries them.

 

An alternative to the CD4xxx series is the 74HCxx series. Those may be more readily available in your area.

 

I'm confused as it seems you're changing your original requirements.

Can the timer be triggered by eye 2 instead of eye 1? Yes.

Can eye 1 reset the timer? Yes, but I'm having a hard time picturing what it is you're trying to do now.

Can you provide a new timing diagram outlining your new requirements? I can better help once I "see" your goal.

 

I see the confusion, In my original timing chart I have the safety timer going high with eye 1, my bad it should have been going high on eye 2. And the both the safety timer and the skip timer must start off in the high state at the begining of the cycle, to assure proper timing everytime the eyes detect an object.

I also see that in my timing chart I did not reset the 2 timer when the output signal went low.

Again, I appologize for the confusion concerning the timing chart. The design you provided should provide everything I need to accomplish with the few modifications to reset the 2 timers.

I thank you very much for your design and input.

Thanks, dave

 

I assume in your design th CD4000 logic chip is an NAND gate, correct

 

I assume in your design th CD4000 logic chip is an NAND gate, correct

Nope:
http://www.datasheetcatalog.org/datasheets/150/108513_DS.pdf

Of course, I didn't realize this IC isn't commonly available (if at all). So I'd suggest a CD4025. Unfortunately, it does not have a NOT gate built in like the CD4000, so you can either use transistors with some resistors to create an inverter or, more simply, go with a CMOS NOT gate such as CD4049 or CD4069.

I'm not sure what you mean by reset two timers. My understanding, which may be incorrect, was that you have one (safety) timer and one (skip) clock. Or are there two signals regarding the skip - one is a clock signal and one is a timer? You showed only one in the timing diagram, so I assumed you simply had a single skip signal that went high and needed a 555 in astable mode to create a clock signal.

Is the skip clock generated somewhere else and you need to control when it controls the output signal? Or is it a simple on/off (high/low) signal that you then need to create a clock signal from? The latter is what the last circuit I posted assumes. There is no timer though - the skip clock will stay on as long as the skip enable is high and the output is high.

I don't see by your timing diagram how the safety timer and skip timer are related other than the skip timer (clock signal) only comes on when the output is high. Is the skip signal supposed to come on for three cycles then shut off?

Again, if you could post a corrected timing diagram, we can help. I just can't "see" what your requirements are now. If you could elaborate on the function of the skip timer, that would help too.

I don't know where you are located, but you can purchase ICs from:

U.S. - Digikey, Mouser, Jameco
Europe - Farnell, Maplin
Australia - Jaycar
Asia - Futurelec

 

So I'd suggest a CD4025. Unfortunately, it does not have a NOT gate built in like the CD4000, so you can either use transistors with some resistors to create an inverter or, more simply, go with a CMOS NOT gate such as CD4049 or CD4069.

Could you use a spare NOR gate in the CD4025 as an inverter?

 

Excellent thought. Yes, if all three inputs from one of the spare gates are tied together and connected to the 555 output, the gate output will be inverted and act as a NOT gate. Good call Tracecom!

 

OK, til I get a chance to redo the timing chart tonight. I will try again, I know the chart is a thousand words....You guys are doing a great job of understanding whats going on.

First: Normal operation of this circuiti:

An object to traveling past the electronic eyes and is always travling in the same direction. Object is first detected by eye-1 and then eye-2. When both eyes are detecting the object the output signal needs to go high, at the time the output signal goes high the safety timer should start. As the object continues to travel and clears eye-1, the output should go low and the safety timer should reset and be ready for the next object. In the event that the object stops in front of both eyes and keeps the output signal high, the safety timer should timeout and drive the output signal low preventing damage, safety timer should reset and be ready for next object.

So under normal operation the output signal is high whenever both eyes are detecting an object.

Skip Timer purpose is to pulse the output signal under normal operation whenever the output signal is high. The skip timer needs to have adjustable high and low times within the cycle. And this timer should also start on a high side of the signal as displayed in the timing chart. How many cycles is inmaterial. Number of cycles will depend on the timing setting on the skip timer and the amount of time that the eyes are detecting an object.

Timers:

Safety timer purpose: is an adjustable time between .5 - 5 seconds that will turn of the output signal in the case that the object stops in front of both eyes longer than the safety timer is set for.

Skip Timer: will pulse the output signal instead of a solid high output.

This circuit will be used to apply adhesive onto an object as the object travels along a conveyor. The eyes will detect when to start and stop the spray which is controled by a 24volt solenoid that I intend to drive with a transistor output from the circuit. The adhesive can be applied in a continous strip or if the skip timer is turned on the adhesive will be applied in a skip pattern like a sewing machine.

I hope I have done a better job of explaining the purpose and goal of the circuit at hand. I can still provide a timing chart if you need one.

Thanks again to everyone for your input.

 

Ah, okay, that helps immensely. Tentatively, the eyes will be connected to a 2-input NOR gate and the output will daisy-chained to the safety timer input. I'll have to give some thought on how to trigger and reset the safety timer and skip clock. I've got some thoughts, but I need to flesh them out.

What are the max and min time durations need for both the high and low settings for the skip timer?

 

I am glad I was finally able to communicate to functionality of the project. The times will be approx as follows:

safety timer: .5 - 5 seconds monostable timer

Skip timer: this needs to be optional in the circuit and time would normally be in the tenths of a second for both the high and low times within the cycle.

I am intending on using 20 turn trim pots for the adjustments of these timers to give me better control of the time.

 

both the high and low times will need to be adjusted separately. control of the spray time and the no spray time is the reason for adjustment of both the high and low times, I think this would be an astable timer.

 

kip Timer purpose is to pulse the output signal under normal operation whenever the output signal is high. The skip timer needs to have adjustable high and low times within the cycle. And this timer should also start on a high side of the signal as displayed in the timing chart. How many cycles is inmaterial. Number of cycles will depend on the timing setting on the skip timer and the amount of time that the eyes are detecting an object.

Why should the skip timer start out HI. Does it really matter? Let's consider the other end of the object. If the skip time is LO and just about to go HI but the object clears the first eye first, the the final region of where the adhesive should go will not get any. If, on the other hand, the skip timer goes HI just before the first eye goes LO, then you will have a tiny burst of adhesive at the final edge of the adhesive area. If both of these are acceptable in the final region, is there a reason they they are not acceptable in the initial region? If not, then the circuitry can be made simpler by letting the skip timmer be a free running oscillator whose output is OR'ed with an enable signal before being AND'ed with the other outputs.

 

Based on the frequency requirements and precision you need to adjust the skip timer, this is a project much better suited for a microcontroller (uC). I'm not sure you'll get the frequency response you want from a 555 astable, but someone else here may know better or have another recommendation. And while pots are great, how will you determine how much to adjust the high and low outputs at tenths of a second? Are you planning to eyeball it? If you went with a uC and a display, you could use the display to accurately set your times. While I wouldn't consider programming difficult, this will take a bit of effort and plenty of trial and error.

Alternately, you could employ a PLC solution which is made for 12/24VDC and factory environments. They employ ladder logic which, as I've been led to understand, is much easier to program than a uC.

 

You are correct, I already have an adhesive system that is controled by a mitso PLC with the touch screen display that allows the user to precisely control the adhesive application onto to object. We control the on/off times normally in the range of .07 - .2 seconds as the object is normally in front of the eyes for approx. .5-1 sec.

The purpose of this project is to develope a low cost system. Not every customer wants or needs to expense of the the PLC model. And I do understand that the control will be less then using a PLC, but you know the saying -- YOu get what you pay for....

As for which cycle the skip timer starts on; The skip timer is normally setup so that adhesive is applied for the first third of the box, then no glue for the middle third of the box and then apply glue onto the last third of the box. This reduces approx 33% of the adhesive used without compromizing strength of the adhesive bond.

Even with a PLC control, the process of setting up the spray/skip pattern is trail and error til you enter the correct times into the PLC.

So I imagine that it would be the same with this simple circuit, where as the box are going by, you would start adjusting the spray/skip trim pots that control the times until you achieve the third/third/third spray/skip/spray pattern. This is also the reason I thought a 20-25 turn trim pot would give me a wider area of adjustment.

Again I understand that this is a better application for a PLC, but I really need to develope this low cost version.

I really thought Elec_mech last design was very close if the trigger point for the safety timer was fed from eye2 and eye1 controled the rest for the safety timer. Wouldn't this solve the safety timer interface with the eyes.

Can the skip timer be ANDed with the normal output, and somehow control the start of the cycle?

Good questions, I appreciate the ideas and hope a solution can come out of this.

Thanks, dave

 

I think you really should consider a microcontroller. The bill of materials will probably end up being lest costly than the present approach and now all of your timers are just variables in a program that is looking at just two inputs from the eyes. If this is in a product you are selling, then the cost of developing the program, which is a pretty simple one, can be amortized over the number of units sold.

Whether you go with the micro or not, I would also suggest using a coarse/fine adjustment process instead of a single 20-turn pot. For instance, use one control that can vary over the entire range in one turn and then use another control that can vary over 5% of the range in one turn. You now have the same amount of control as a single 20 turn pot. Which would you rather use? This also makes it easy to calibrate the dials so that you can have a close starting point.

 

I've tidied up your first circuit,with the edge trigger for the 555 try the schematic as shown here....

 

The logic is easy. You'd use a 2-input NOR gate with pull-up resistors on each input. The output is fed to a 555 monostable timer. The timer turns on and enables the sprayer through a transistor. In series with the transistor is a second one controlled by an astable 555. The concern I have is employing the tight control you want on the astable. This could be verified through trial and error I suppose.

Some questions:

1) Will these be adjusted once for a particular product/customer and never changed again or do you expect you or your customer will adjust them multiple times for use on various runs requiring different times?

2) How many of these do you potentially envision producing a year? 5, 500, 5,000?

If you'll be adjusting the timer/skip clock multiple times or if you expect to make more than like 10 a year, then I would strongly suggest the use of a microcontroller. You can get the precision of the PLC, a smaller parts count and circuit board, and eliminate manually setting each unit through trial and error. You figure out the times once for a given product, then program the same value into the rest of the units. This will save you a lot of time, labor, and money.

If you have a set number of times, you can program them all into the microcontroller and add an LED and button to allow the end-user to select the one they want for a given run.

 

Thanks for the responses and thoughts. The setup is usually done once at setup time of the equipment at installation time. So once the times are set, normally you do not touch them again.

When you say microcontroller, what exactlly are you talking about. Like an EPROM programed with the logic of the circuit? If so I am way out of my ball park at this time for that. I would need to hire one of you guys to do that for me.

As for how many units? I would guess less tan 10 a year for this type unit. We normally sell the PLC controlled unit.

Just for interest, I uploaded 2 old pictures of my proto type of the PLC controled system with a touch screen operator interface. The unit are currently basically the same only green in color and only slightly different on the inside.

As soon as I receive a few more required ng.parts, I will build the circuit and continue testing.

Thanks again, and you may hear from me again when I do continue my circuit testi

 

UPDATE:

I received my parts today and built the circuit that elec_mech gave me, with a few changes.

I have the circuit is working as needed. Now I need to move it from the breadboard to a circuit board. I think this could be worse than designing and testing the circuit.

I would to thank everyone that gave input to this project, attached you will see my final design. At least I think I transcribed it correctly from the breadboard to the schematic.