Sign in to follow this  
Followers 0
Davids

Some programming Question

20 posts in this topic

Hi, today i need to do a pin tester machine. my machine concept is like this. Maximum pin check is 80 pins and user can key in any number below 80 to do the pin tester process. for example: if user key in 30 in the display, that mean the machine will do 30 pin check. once pressed start, the tester will check 30 pins. if one of the pin had error then the machine will continue do the checking. the screen will display all the pins error message. if 5 pins is fail in the test, then the screen will display which pin is fail. here is my question: what is the best method i can do for my machine? i had try to think it myself before post this topic. hope someone can give me some guide. by the way, i'm using Fx1n CPU. very thank alot.

Share this post


Link to post
Share on other sites
Can you be more specific what you are looking for ? It looks more like a questions about how to the display values in a terminal than about PLC programming. Edited by Gambit

Share this post


Link to post
Share on other sites
Since the FX1N supports a max of 128 points are you looking for some way to multiplex your outputs and use say 10 inputs and 8 outputs to be able to test 80 points?

Share this post


Link to post
Share on other sites
There are many ways you can do this - 'shift bit' 'counter' 'increment data register' - but we need more information ; like, how you define a pass and a fail - are they tesed all at once or one per cycle - do you need total failure or which test failed.

Share this post


Link to post
Share on other sites
thank for all feedback here... for the connectivity test, if the pin is ok then input (X) will ON and indicate that is passed. if the pin is NG then input (X) will OFF and indicate the pin is NG. my application is all the pin are tested all at once and when the test is over then the display will indicate which pin is failed in test. beside that, user can key in any number below 80 to start a test. for example, if user key in 50 then the tester will do the test until pin 50 only. hope this time i can give a clear picture for you all. very thank alot here.

Share this post


Link to post
Share on other sites
Are you just testing conductivity of each pin? You could do that without a PLC by simply wire your tester to a series of diodes. Plug the connector in, and if the diode isn't lit, the pin is bad. Alternatively, you could wire the connector as a short circuit for the diodes, so that the diode is lit if the pin is bad. You could even put in some kind of selector switch for different connector sizes. The switch would activate/deactivate groups of LEDs. If you're serious about using the PLC, you'll either need a multiplexer or a lot of expansion inputs. You also haven't mentioned what kind of screen you'll be using. It would likewise be helpful to know how the connector is plugged into the tester.

Share this post


Link to post
Share on other sites
thank for the reply. my customer would like to use PLC for this project as user can change any number of pins to test the pin. that vary from 1 pins to 80 pins. if the faulty pins occur, it will display in touch screen. this tester is use for pin conductivity test only.

Share this post


Link to post
Share on other sites
Your description is way way too vauge If you cant describe it properly, how are we to help you program it properly. Start with the process from the begining - describe everything the operator will do including how the pins are held, (solonoid clamps etc) how you are test conductivity (micro switch - 24V through pin to input etc) A lot of things dont make sense to me - To be able to program a process, you have to know and understand every detail of the process. write a long and descriptive process report - and then you will get some answers

Share this post


Link to post
Share on other sites
Please answer my questions or I can't help you! Also, as Goody has mentioned, a more complete description of the desired end product is needed. Edited by JRoss

Share this post


Link to post
Share on other sites
sorry for giving insufficient information. the whole process is like this... 1. the operator will put in the chip into the tester. 2. operator needed to key in how many pin does they want to check. 3. after the setting is complete, then operator press start to start checking 4. when in checking process, the tester will check all the conductivity of each pin. (conductivity test by 0VDC through pin and back to PLC input) 5. if all the pin pass all the pin test, then the chip is OK. if one or more of pins are checked fail then the chip will be reject and a 7-segment display will show which pin is reject. 7-segment display will display one pin fault each time. operator needed to press reset button to check for another fault pin allocate. hope this time i can give all the relate information. THANK...

Share this post


Link to post
Share on other sites
Well, I'm not sure how the chip would be connected, but let's assume that it's a similar problem to testing the "pin to core" soldering connections on an 80 core cable with 80 pin D plugs at either end, and that these cores are not connected in a "straight through" way ( 1-1, 2-2 etc), but are crossed over so that 1 might be connected to 47, 2 to 35 etc. If that was the case, then I'd use a PLC with 80 outputs and 80 inputs. I'd solder an 80 pin D plug to 80 cores and connect them to the 80 outputs of the PLC. This D plug would connect to one end of the cable to be tested. Then I'd solder another 80 pin D plug and connect the cores to the 80 inputs on the PLC. This D plug would connect to the other end of the cable to be tested. Then I'd write a programme that energised each output in turn and looked for the correct input to turn on. If it did this correctly, then I'd set an OK flag for that output / input pair. I would also check the status of the other 79 inputs to make sure that none of these had turned on (faulty soldering) If OK, then turn off output, turn on next output and repeat for all 80 outputs. If not OK after a fixed time, then set a Fault Flag and move to the next output. Use the HMI to show the 80 test results -Fault or OK. It would take quite a while to set up, but if you've got lots to test, it would be worth it.

Share this post


Link to post
Share on other sites
what is considered good and what bad connection? what is test condition? for example you can't check if ac power bar is ok using low signals. you may need to pass particular current and maybe measure voltage drop. do yuo need to check only continuity or also between pins shorts? how about short to ground? what is the test time limit? multiplexing will complicate design a bit but it can save on I/O and wiring.

Share this post


Link to post
Share on other sites
I agree that it's not perfect in terms of voltage drop, etc, the thinking was that it would be capable of checking continutity and pin to pin connections, which, if I was soldering this cable by hand, would almost certainly be a problem. (eyesight, age...) Let's say that it was a "non straight through cable", ie all the pins on one end were connected to "other number" pins on the other end, and you wanted to do a continuity test to check that the pin to pin connections were correct, in this case it would be a long and probably error strewn task in testing the pin to pin connections with the a hand held multi meter. So with 80 outputs and 80 inputs (81 if checking short to ground on the cable hoods), I would write a programme something like: Start Command. Turn on first output. Check for status of expected input and start timer (say 200 ms) If first output and expected input are both on then core is wired correctly. To check whether there are any cross soldered mistakes, I'd use the SUM command, which reports the decimal number of bits that are in the 1 state and stores the result in a data register. So, SUM K4X0 would tell me how many inputs are on in the range X0-X15. Repeat this for all the input numbers you are interested in (81), which would mean 6 data registers. Add these values together and the value should be 1. If it's more than that, then you have a cross soldered connection. So if the correct output/input pair are on and the SUM value is 1, then set an OK flag, move to next output and repeat. This would tell me pretty much everything about the cable: For each output / input pair, the logic would be: Output + Input = OK, Sum = 1 ............all OK. Output + Wrong Input, Sum =1........... then wired to wrong terminal. Output + No Input, Sum =0 ............... then faulty solder terminal. Output + Input = OK , Sum > 1 ...........then solder bridge somewhere. I estimate that a 200 ms time would be plenty to enable each test be be carried out, with good IO pairings moving on immediately, which would mean a maximum 16 secs cable test time for a completely wrong cable, and probably much less for a good one. At the end of the time, use the HMI to display the results, highlighting the good and bad pins. Can't quite see how you could multiplex this, as you need to energise each pin in turn to do the full diagnosis , which either means lots of outputs or some external (rotary switch?) method of doing this.

Share this post


Link to post
Share on other sites
Dare I suggest 81 outputs and 81 inputs. Need to check hood to hood continuity as well as any pin shorted to ground. This means 162 i/o points more than the 128 this FX supports.

Share this post


Link to post
Share on other sites
It's a good point .... and, I guess that as the IO range exceeds the capability of the FX0N, then it's going to present a problem. Even doing it as two "half" cables wouldn't work, as cross soldered connections would not show up if they were soldered to pins that weren't connected on this part of the test. Looks like an old A series or something similar, in that case. PS, when I said X0-X15, that's obviously wrong,as it would be X0 - X17 on an FX PLC type, or X0 - XF on an A series.

Share this post


Link to post
Share on other sites
How about this Colin 81 Inputs {Hood plus Pins}, 41 outputs run to the common terminal of "ice cube relays" Omron G2A-432 type and 1 output to cycle relays. This would consume 123 of 128 i/o points and detect cross wiring since all inputs are monitored.

Share this post


Link to post
Share on other sites
That many outputs will not be required. There is no HMI involved, only a seven-segment display. Assuming two digits, that will require 8 outputs (2 BCD digits at 4 outputs each). You may also want an output or two for status lights during the test. You will need more inputs that that, though. Again, since you have no HMI, you'll need some kind of input to indicate number of pins. Probably a two-digit thumbwheel, which also uses BCD, so 8 inputs there. 80 inputs for the max number of pins. An input for the "reset" button. Maybe a couple others that I don't know about? So, 89-91 inputs and 8-10 outputs. Max I/O of 97-101, well under the 128 limit. If the test is simple conductivity do this: Take the thumbwheel inputs and convert the two-digit BCD to binary. Use that to "activate" the inputs you'll be using. On each "active" input, check each to see if you have the proper state. If not, save the number of that pin in a stack. Repeat this process for each "active" input. When the test is done, take the first pin number from the stack and convert it to two-digit BCD and output it to the seven-segment display. Each time the "reset" button is pressed, pull the next number from the stack, convert it, and display it. If the test is more complicated, as it very well could be on a chip, use some of the methods suggested above. At any rate, you need to get a programming manual and check out the BCD instructions. For a stack, you could use index registers. Also something to read about in the manual. Any specific questions, ask away! Why don't you take a stab at putting together a program? Post it here for us to review and comment.

Share this post


Link to post
Share on other sites
What's an "ice cube relay" Bob? Have googled the part number without any success.

Share this post


Link to post
Share on other sites
Looks like my part number memory G2A-432-AN is sutck in the 1980's. I can't find that omron number now but the MY series is close to the same.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0