Jump to content

kratmel

Members
  • Posts

    520
  • Joined

  • Last visited

  • Days Won

    20

Everything posted by kratmel

  1. Hi, i do not know your gear+sensor setup. But main problem with counting - speed and vibration. If you use only one sensor and standard or HS Input - vibration at low speed generate false pulse and counter do not count correctly (+ count error present). If standard input used - you can lost some pulse at high speed sensor activation (- count error present). HS input in single sensor mode can be used for high speed sensor activation but start stop condition also make some count error. Best way for mecanical setup - use quadrature principle of measurement (you can read about it) and A/B shaft encoder HS input. Encoder with A/B output or two PNP sensor on the same special quadrature gear can be used. This idea work with best result for low and high speed. Vibration do not affect on count result. By the way as i see in your setup - two sensor used for mesure the same amount of data. You can use only one sensor connected to two HS input chanel I0 + I2 at the same time and count/reset two HS counters separately. Only one problem - HS counter backup ladder code needed. You can found V120 example in help project HSC directory.
  2. Please try to go to info mode and switch PLC to Stop mode. Than try to load JT project.
  3. I recently made a program for a plastic extruder. The user makes settings in 5 windows that will adjust to one recipe. Within each screen, I used the comparison method for a single part of the recipe (part of data table line of the recipe), not for the whole line. So there was a way that shows the user that the data on the screen is different from that specified in the recipe. I did this by hiding the SAVE button when the data is the same within the window. When at least one of the data is different, it changes color and the SAVE button appears, which disappears if the user has already saved the data in the recipe and they are the same as on the screen. There is a separate window for viewing the whole recipe - but as practice has shown - users like to display individual parameters within the window. The main task of employees is to use only permitted recipes. Therefore, after some time, managers have a wish that the SAVE key not only saves the changed data in the recipe, but also asks for a password before saving. So the manager sees the data changed by the operator in the process - if they are correct in his opinion, he saves them to the current or new recipe. As for scrolling the entire line of the recipe, this is possible only if all the data is visible on the screen at the same time. If there is more than one screen then scrolling recipes requires switching windows and leads to errors.
  4. You are not alone..... In engine START-STOP automatic algorithm present battery voltage comparator, time from engine start and even driver seatbelt switch. In fact, I keep asking myself about the logic of everything that contains a program in the controller. However, I am surprised when instead of a simple and clear algorithm they use something strange and a very excessively advanced controller that does not cope with the task. In my opinion, the main thing for the driver is that the brake algorithm does not start to be updated remotely when the driver needs to brake ... ... It's good that they still haven't mechanically disconnected the driver's steering wheel from the wheels ... Although they are constantly trying to put this task on algorithms and electronics.
  5. Thanks to AlexUT for the screenshot hint. I want to say that the main task in the testing process is to link the tested object with the report. I solved this problem with the help of a sticker printed by a thermal printer. Because the serial number is not usually assigned until the device is packed, it is very difficult to distinguish between tested and untested units. Attaching a camera that will photograph a tag with a unique number on the product and take a screenshot is a good way. Screenshot of only a set of parameters, even with the test time - an unreliable solution. Therefore, the author of the topic must determine the sufficient reliability of the received report. I had a problem with wire testing when an employee didn't check every wire, but only checked one quality sample several times and printed many correct results - but not all wires were checked. This is the main problem to be solved by the developer of the testing program.
  6. I build machine with direct print resullt on termal printer. PLC generate report via printer standard lenguage. Work fine. In your case separate csv to PDF PC based application needed. In some standard testing station (automotive wire production) only text string go to server and server generated result (pdf) printed on network printer near machine. All testing result is fixed on server with time stamp.
  7. In my application - all signal wire is black. Only correct numbering solve problem with wrong connection.
  8. Yes! It is my standard process of upgrade. I newer change current version of PLC program. First of all i "Save as" current version to the new one #+1 and then make changes. After implement changes - type on paper what is changed and what new settings appear. Than last machine project go to user with #15 software realase. One more increment needed after onsite testing. After all is OK - operand is exported to file and saved with last used project version #. But data in table is not owerwriten when new project update is downloaded. Separate data table synchro needed.
  9. For a long time i try to solve this problem. Then my solution - use little different style of programming. Then all my current application has default settings option (copy from datatable to user settings). I use datatable - that it can be downloaded separately - used for change something default in process of project modification (save user settings to datatable as default option present on PLC via password). If user need to change something (not recipie) - it have printed table with default and user changed settings. Then if something goes wrong - user can load default option -and change only small amount of parameter to start machines. Maybe this idea need more time for programming but it help to restore and modify project whithout problems. User settings for working application (all operand) - can be exported to txt file (Visilogic menu option) and saved in current downloaded project directory named as PLC serial# and time of modification. I can compare the same format files with different operand settings. And load specific operands backup to PLC.
  10. In real field installation - RS232 with shielded cable can work at full speed for approx 20-50m. Best way - use isolated RS485 (port 2) card inside V130 and use isolated PC USB to RS485 converter. In this case on the lower speed (9600) and even with unshielded cable - distance can be approx 1200m. At high speed - shielded twisted wire (only two twisted conductor inside) cable recommended but distance approx 500m is possible or even more. Ideal solution - use optical RS232 extender but it is not cheap solution.
  11. As i see, after last forum changes i have sometimes the same on my PC. I logged on the forum but i can't see myself on forum current user. Before update i can see what topic active registered members read. Now it is present only on next step "See full list" with all Guests. And this Rank roket... like Ilon Musk Starship on all members icons in topics....
  12. And we all here only dream of such possibility.... The pump is cooled by a fan, and the fan should probably be cooled by a slightly smaller fan and this smaller fan also needs to be cooled by something ... 100F in US 33C in EU.
  13. Great invitation Aus! By the way... Is it possible that some VFD is installed near the Unistream installation? I found some problem with communication and panel start if some motor with VFD is operated in the same power net. In my case GND wire of 24V power supply (-) connecting to machine GROUND solve problem.
  14. standard solution https://robertoweninc.com/4-20ma-plc-thermistor-measurement/ direct with hook up table linearization needed http://www.tri-plc.com/appnotes/F-series/F16160BAThermistor.pdf But +10V must be prepeared via some add cirquit.
  15. I pulled a V700 out of my cave and tried to experiment with this strange case of linearization block behavior. My goal was to determine if the problem was dynamic or static. That is, whether the sequence of actions affects the result. The answer is no. The result of research for the data submitted by the author: the first transition to the wrong result 1402485 to 1402484 then return to the correct one 1402318 to 1402317 again going abroad 1402152 to 1402151 The result is repeated in both directions of calculation. The problem changes if you change the number 6000 to a smaller side. But as I understand it does not disappear but moves to another range of values. Replacing 600 and 6000 gives the same result only on the contrary - we fall into the negative values of the result. So my conclusion is: the problem is the result of a combination of values. In the case of microcontrollers with a limited calculation data bus and a problem with negative values, it is this combination of data that goes beyond the calculations. Therefore, for practice we have the solutions defined above, and from the point of view of the linearization block, working with large values gives the probability of exceeding the capabilities of our PLC. Working with the formula, we can predict overflow and bypass it, taking into account the order of calculations. It is impossible to do this for a closed standard block. Therefore, if you saw a problem in the operation of the standard linearization and solved the problem via formula - PLC generally corresponds to your task.
  16. I'll try to fantasize a little on this topic. You have an AC fan. There are two points in such devices. The first - the fan is guaranteed to start only after reaching a certain supply voltage. Second - by slowing down the fan you can go to a voltage lower than the starting voltage and the fan will still rotate. Therefore, I recommend determining the PWM at which the fan will start if you gradually add voltage. Conversely, determine the PWM when it stops when the voltage drops. I think that's all you need to know. By setting the adjustment within (minimum starting voltage + some margin) up to 100% PWM - you will have everything your fan can. However, in general, it is said that it is illogical to apply a voltage below 40% of the nominal voltage to the fan - because the fan no longer creates an air flow.
  17. If you use Jazz PRG adapter - please check all pins inside PLC connector. Sometimes one of them bended. Best way compare pins in working and not working Jazz. Pleaese note - power pins is longer than IO pins inside PLC. Sometimes someone insert something inside PRG slot.... but not PRG adapter....
  18. I think it look's good. Do you test it? Sorry, my answer is a little incorrect. I found in your program that you operate with PWM output frequency MI1. Duty cycle in your code = 0. Then output is not active. Please SET some constant freqency to MI1. And run PID with MI2 output (Duty cycle). Then maybe system start working.
  19. Hi, maybe i know what is wrong. Your application works in real time. Then sometimes linearization maybe do not end at the time when ML must be reloaded. In ladder code you send timer to ML and after that do linearization. Please try to change ladder net - first linearization - then copy timer to ML. P.S. This problem appear only in this two times at every working cycle or bug happend in different time point at every new cycle?
  20. Please read PID section in HELP. In your program no PID RUN fb present. Some add net needed in ladder.
  21. Great tip Aus! I try to write the same but you are first one! 🙃 Is it possible that used sensor is broken? INOR provide sensor break option I<=3.6mA on output.
  22. Hi, You described the problem well, but did not indicate which Sambs PLC model you have in your hands. I think it's a PLC SM35-J-R20 A common mistake when connecting an analog input is to confuse I9 (descrete input) with the 9 pin (AN0) of the controller connector. Also you can monitor analog input MI via online mode and see what value present.
  23. We have also made some progress in this direction - made new legs for the controller that resemble goose paws (3D printer used). Unfortunately, I do not have a photo of the present invention now, but they are distinguished only by the presence of one clamp that easily secures the leg in the right place. Joe's invention needs to be patented! 🤣
  24. Hi, i try to post some tips about standard VFD, Encoder and PLC: 1) It is not possible to build closed loop control (servo control) on standard VFD+encoder + standard PLC; 2) Standard PLC with HSC connected to encoder can offer positioning machine driven via VFD. 3) For positioning two speed - FAST and SLOW setpoint must be proggrammed on VFD (or provided by PLC). Then user must program ladder net with FAST to SLOW changeover based on distance to setpoint. 4) If positioning used for CUT to LENGTH application - simple encoder register reset is used for -0- setting. 5) If you need to set position of standard linear axis - home position evaluation is the main problem. 6) It is hard to build (or not possible )positioning based on Modbus RTU command (to slow for send stop command). VFD logic input and analog setpoint input must be used. Then please post your machine setup and application type. Some type of VFD+encoder+PLC based application posted in forum topics. You must define what you need to do with this VFD+Encoder+PLC.
  25. In att. i post some sample with non standard modbus via protocol. You can see TX - send and RX - scan idea. samba_protocol sample.vlp
×
×
  • Create New...