Trying to make a customer understand

[todd99 0]

I have been a member of this forum for months, and have read many posts, but this is my first. I have received great information here and am hoping to get some input from some of you to explain something to a customer. The project I am doing is for a shop that is building a checking fixture for automotive bumpers. It is a fixture that holds the bumper while the operator installs the license plate lights. This shop specializes in machining, fixing things, basically a job shop. They do not have anyone with an electrical or controls background, and as far as programming or automation, forget it. What they asked for was a control package that verifies continuity through the two lamps to call the part good. If there isn’t continuity through both lamps the part is bad. I achieved this by sending 12VDC through the lamp to drive a relay and used contacts from the relay to send a 24VDC input to the PLC. They made an assembly that holds the mating socket on the end of a cylinder rod to stroke out and touch the lamp socket. In essence, sending 12VDC through the lamp base, through the filament, back through the base and turn on the relay. If both lamps are good, clamps open and the part can be removed. If one or both lamps are bad the sequence stops and it requires a keyed selector switch to be manipulated by a supervisor to open the clamps to remove the part. First pass it worked great. If a bad lamp was installed or no lamp was installed, it would not release the part. Indicator lights on the control panel showed which lamp was bad. Now their customers come in for a look; what they are asking for is to have a visual indication that the lamps are good. They want to have the license plate lamps light up for 5 seconds, while at the same time getting the input to the PLC and verify the continuity. I cannot make them understand that what we are trying to do are two different things that require two different wiring conventions. Lighting the lamp requires +12VDC and -12VDC. Receiving an input to verify continuity to turn on the input requires a +12VDC signal going out the to lamp through the filament and back to the PLC. In effect making the lamp a switch, if the switch is true you have voltage through it ,if its false you don’t. Is there are any suggestions you all have that could explain this problem any simpler or any way to achieve this that I haven’t thought about please let me know. Thanks for any and all help Todd99\

[gravitar 3]

I think I understand what you're doing, BUT.. there's a couple details I don't follow. First of all, are you sure the car supplies +12VDC and -12VDC to the lamps? (for a total of 24V)? Now I know I'm way behind the times but I thought cars were standardized on a +12VDC, chassis-grounded system. So when you say the customer is requesting a "visual indication"... You're applying 12V to the bulbs, right? isn't that lighting them up? If you mean the PLC has to see the bulbs lighting, that would be do-able but not with an off-the-shelf photoswitch I don't think. Call A-B or Banner (or whoever you use) and see if they have anything that responds to visible white light, if it turns out that they don't then you might need to rig something up with photocells. And as far as the relay.. are you just driving a 12V DC relay? If the relay coil is in series with the bulb filaments, you won't have a full 12V at the coil. Once the filaments heat up, you should have a significant voltage drop there. If the 12V coil works then that's fine, but just bear in mind that you might not have good repeatable results every time.

[chas183 0]

What if you lighted the two bulbs. Like the customer wants. Then using Two current transformers to measure in the milliamp range the amount of draw the lamps have. Now no two lamps are created equal. So you would have to find the sweet spot. The current transformers would hook up to an anolog input to measure the results. The voltage created by the current transformers could then be used to do a compare statment. Could use three rungs to compare the values. For example the lights are good if you draw 1 amp."Probably much less" If you draw more than 1 amp they are shorted. Less than 1 amp bad bulb or wiring. Open condition. Chas

[BobLfoot 301]

I understand what you have given the job shop and what his customer wants. Consider this. Use a 12VDC supply to light the bulb and use an Analog Input on the PLC to monitor current in the wire. This will give you a PLC input and a Lit Lamp. It will be a somewhat costly change order, but thats up to the job shop and the customer to work out.

[Snerkel 0]

You could make a simple mod that satisfies both requirements. You would need an additional output from the PLC such that once you get the OK signal from your relay you short circuit the relay coil (using a relay driven from your PLC) thus illuminating the lamp. Of course your existing system will not detect a short circuit as the relay would still switch, my suggestion would cause a problem if you did have a short circuit. A better system would be a series resistor in place of the relay with a resistance approx = 50% of the lamp resistance. Using an analogue voltage input on the PLC measure the voltage across the resistor a good circuit will give you about 4v across the resistor, a bad circuit will give you less than say 2v or greater than say 6v across the resistor. This circuit will also light the lamps at a reduced brightness, the advantage of a reduced voltage across the lamps is that they don't get as hot. You would need to make sure the resistor had sufficent power handling to take the full 12volt across it. If you wanted to light the lamps at full brightness then place relay contacts across the resistor so as soon as you detect a good part you can switch the resistance out of circuit.

[todd99 0]

Thanks for the suggestions. I never really thought about going the analog route. I think I can do this with minimal wiring and programming, if they are willing to pay for the changes. I expected to here back from them by now, but haven't heard anything. The automotive company was supposed to check with the manufacturer to see if this is indeed required. It's just frustrating when you are asked to do a job and it works out perfectly, faster and cheaper than what I had initially committed to (although I didn't mention the cheaper part). And then they want to make changes, that I would consider to be extreme, considering the the simplicity of this project. But as they say in The Godfather "this is the business we have chosen". I'm sure you all have the same stories. Thanks again for the help.

[DanW 41]

WIth what you have, the relay operates wihen the system (bulbs) is OK, correct? Use a DPDT relay (instead of a SPDT) and use the 2nd set of contacts to drive a 5 second timer that turns a light on. If the system doesn't work, the relay doesn't work, not light. If the system works, the relay trips, the PLC gets its input, and the light lights. Dan

[gravitar 3]

Not to knock any of the suggestions that have already been made, but I just don't see the addition of analog inputs being much of a value-add to the customer. If I'm following what was requested, the customer wants a "GO/NO-GO" response for the electrical continuity of the light circuit. Not a measurement of "luminous intensity" reaching a certain threshhold or anything :) I don't claim to have all the answers, but my hunch is discrete I/O will be sufficient to get the job done.

[todd99 0]

I am happy to report that the customer agreed to have this shipped without any changes. Perhaps it was not as important as they origiinally thought. Gravitar, what they originally asked for was a GO/NO-GO but after it was built they wanted a kind of a GO/NO/GO/GO AGAIN. They wanted to see the continuity through lamps and they wanted them to light. As with some of my projects wiht a new customer I have a tendency to get prematurely nervous. My bad..... Thanks again for the help....I look forward to posting here again....

[TConnolly 13]

There are dozens of ways to do this, and you don't need an expensive analog module either. Here is one possibility: CR1 is a solid state opto-isolating type relay. (such as http://solid-state-relay.crouzet-usa.com/s...e_relays.shtml) The turn on current is much lower than what the bulb will draw so it can drain through the bulb as long as there is continuity. The circuit shows a resistor but it very likely won't be needed. Then to turn on the light, you activate CR2. Note that while CR2 is on CR1 will switch off. As soon as CR2 is off, CR1 will switch back on. Just account for it in your program. This kind of an arrangement works because I have used precisely that circuit before - though the application was considerably different for a hydraulic system where we needed to know that the valve DIN connector was plugged in. This is also a perfect application for a sourcing type input module, the kind to which you connect NPN devices. If your PLC manufacturer makes one of these that will switch on at 12 V – I would bet that most 24V inputs of this type will work at 12V, then this will also work. However the current from the sourced input won't be enough to light the light so you will have to switch on a relay – and like the first example the input will switch off at that point because it has equal potential on both sides. As long as you account for it programatically it should be fine. . Edited 1 Sep 2006 by Alaric