Help with 3 stage timed controller

Your control circuitry could be a 6x6 board (maybe 4x4) with 7400-series counters and logic on it, or it could be a postage-stamp sized board with a PIC, AVR, or other itty bitty microprocessor, with more board space used for the connectors to the thumb wheels than is used for the processor -- even through-hole.

How much help do you need? Do you want to just pay someone to build you a new box, do you want someone to scratch out a schematic on an envelope for you to implement, or do you want to get pointed in a general direction and take things from there?

If you have time this would be a great excuse to learn how to use a Basic Stamp, or a PIC or AVR. For a one-off that stays in the commercial temperature range (basically 0C to 40C) you could get an evaluation board from Microchip or Atmel, program your application, and hook up your relays. It would be how I would do this job for a client, were I doing it.

I wouldn't do it with discrete logic parts unless I clearly had a client whose skin crawled at the notion of using a micro, but who felt comfortable with 7400 series parts. It _is_ a fairly straightforward problem to solve with discrete logic, but somehow a micro+program always does better after the first change order than logic+logic+logic. The real challenge when you have a microprocessor in the mix is controlling your urge to build an Ethernet-connected, VGA-display-equipped mini gaming station that may, if you remember your original design intent, operate a machine.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

I can't see the OP's post, maybe he used this dreaded Google domain. Anyhow, if a PC type architecture controls the rest of the machine a LabJack may be an option.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

IMHO your best bet would be to buy a micro-PLC for around $100 and program that to do the job. I would not suggest fooling around with component or board level stuff in an industrial situation. The functional logic part of the design is 10% and making it work reliably in your situation is 90%, for an experienced engineer. It might be 99% for someone doing it the first time. Note that the first part does not get smaller the first time..

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

I have to concur here - been there, done that. In fact, it was the same exact setting - a vertical extruder. I'm pretty sure it was done with level switches - turn on the vac when the level of resin in the hopper gets low, and turn the vac off when it (the hopper) gets full.

Cheers! Rich

Wow, I just got in and noticed a number of replies. I am doing this with google.groups as I don't use a news reader for anything else, so thank you for your patience.

What I am looking for is basically a schematic so that I can build the working circuit. The old circuit is in a grey enamel painted industrial electrical box that sits on floor next to the vacuum pump, and I am going to gut that and replace the innards. The box is about

10" x 10" x 6" so I have plenty of room. I am not really interested in doing any programming. In fact this thing will seldom get messed with. Once I figure out how long it takes the hopper to fill and need a dump I may not be touching that time setting but twice a year - when I spend all day just grinding scrap. The delay will get adjusted slightly more often depending on how quickly scrap from running parts get thrown into the granulator. Sometimes I want it to repeat right off, sometimes I what it to pause just so it is not running constantly. I basically kick the chrome toggle switch to the on position with my foot at the beginning of the day, and kick it off when I go home.

I have looked at a number of conveyor systems and controls, and they are all far more complex than we need. This is not a material handling issue, and I am not feeding a critical piece of equipment like an injection molding machine. This is just much less messy and much less labor intensive than pulling out the drawer and dumping it in the box every time it fills up. So anything programmable is overkill, there will be no reworking, and I am the client - it fills my own needs. I would however like to do it as inexpensively as I can. Also the old box has LEDs for power, eject, and vacuum - but those are simple enough to figure out, and it is easier to tell it is working by the loud vacuum pump and air blowing than some LEDs.

Google sequential timer bg

Thank you, i had not noticed cheap PLCs had gotten much below $1000 already. With that info i would not even mess with a uC, which was my first reaction.

The Journey is the reward"

ww.trexon.com

ff.com

The issue here is that I really don't care to learn how to program a PLC. If I had time to learn ladder logic, I would do so and write out the flaws and headaches in the two thermoforming machines I have running PLCs. Both machines have a program that work about 65% of what I want, and both are different. If I programmed the PLCs to merge the best of the two programs they would be about 90% of what I want, and I already know the behavior of the rest I need. But alas I look at the programming supplied by the builder and it is about 50 pages of ladder logic.

I don't have the time nor the inclination to learn this stuff at the moment. If I already knew how to do it, it might be the best solution. but hardware alone does not make the best solution. My boss will let me fix it, but not if it takes the equipment offline for more than a hour - and he certainly will not pay for a PLC to replace what is currently being done (poorly and intermittently) by a dozen or so discrete components and no ICs.

I guess the real questions are: How much productivity per month is being lost? How much is your boss willing to pay for a reliable solution? Is there a reasonable match between the two?

The micro PLC and uC development may have low capital costs but have serious development costs, and much higher for the uC. It is possible to build a discrete logic solution based around a dedicated sequencer and meet your design requirements, also at a large development cost.

Where do you want to go?

=A0"The Journey is the reward"

//

speff.com

There are other considerations you are completely ignoring because you with to push the PLC route (possibly because you are not that great at discrete component development), and of course you just assume everything is in place to be able to use a PLC since it is your preferred choice.

1) You have to program a PLC - this requires a computer. I don't have one out on the factory floor and I don't own a laptop. 2) This function is dead simple, no inputs, three stages. The last stage is literally "do nothing for a while, then start over" and you want to throw a computer at it. 3) I have to do this on my own because the boss does not care if it gets fixed, meanwhile my loss of production is making me look bad. I can gut the box and put something better in it, and even get him to pay for a few IC and components - but having him purchase an industrial computer; I might as well tell him I could do his dental work. 4) Your "Real questions" are based on the assumptions that this is a large factory that runs a tight schedule. We are a two person custom plastics shop that was 5 people before recent layoffs. Everything is custom and we usually have runs of about 40-100 parts at a time. The boss does not care that babysitting this piece of equipment has me running back and forth all day, as long as I get the part run done. 5) I fail to see how this simple time, pulse, wait, repeat circuit can have such high development costs compared to purchasing an industrial computer, and learning the programming language or software to make it work. 6) You never asked what my build abilitys are. I can solder to NASA standards and have worked RADAR and implemented wireless ISPs, so I can build this neat enough to not have to worry about stray capacitance or induced noise from the florescent lighting.

I am sorry if I am coming of rude, but I am the thread starter and you seem more interested in pushing an agenda of "use a PLC for everything" than helping with my issue. I have you opinion now Joseph, and it is filed, I need hear no more about it.

This might help.

-- Joe Leikhim K4SAT "The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason? For if it prosper, none dare call it treason."

"Follow The Money" ;-P

There are tons of programmable timers if you simply "Google". Why not buy three of these timers and cascade them together with relay contacts and diodes as needed to meet your requirements?

-- Joe Leikhim K4SAT "The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason? For if it prosper, none dare call it treason."

"Follow The Money" ;-P

It sounds to me like the part that may most need replacement is your boss.

Is your resume current?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

No, a very cheap (one out on the factory floor and I don't own a laptop.

Just not true, serious PLCs may require serious compute power but minimal systems can usually be programmed from "the front panel"

That is why i suggested a simple state machine.

So, you refuse to answer my questions clearly. How much productivity is lost? Do your fellow employees have similar issues?

Quite the opposite, i assumed that this was a small business, not over a dozen or so people.

Two observations, you are still there and your pervious coworkers are not; your boss undoubtedly could not run the equipment and run the business at the same time. You have more leverage than you think.

It is up to you to decide how to partition your time and money. Your expressed objection to learning PLC language has been noted. On the other side of the coin you are asking for a professional design essentially for free.

You are not only coming off as way rude, you are attributing to me attitudes that are not interested in solving problems. Based on the cost requirements and your stated abilities, a $100 miniPLC _is_ a good solution. Check it out, Sperho first suggested it.

Many here could do any number of alternative solutions, how much are you / your boss willing to pay? Bear in mind that in the post you are responding to, just such was suggested.

I have posted a schematic for a controller that may do the job for you. It's simple enough to be built in a home shop, but is robust and reliable. It does not provide 01-99 second time adjustment, as that is quite a complication, but the analog time adjustments are fairly accurate and repeatable. You can see the schematic and parts list at: