Joe Tauser

Visilogic has a confusing way of displaying all coils of a certain address with the status of the coil. Your Reset coil shows it's on because the Set coil turned it on. If you call the Reset it will go off. You may want to write yourself an experimental program and test it while looking at it online to prove this to yourself. Joe T. PS - thank you for posting your code. So many people post problems on the forum and expect us to be mind readers. One of my pet peeves.

Use a multimeter to check continuity between the probe tip and either of thermocouple wires. If you get continuity, it's a grounded probe. This is how thermocouples are normally made, where the junction is actually part of the tip. It's perfectly fine until you start getting the problem you are experiencing. You can buy un-grounded thermocouples, but you have to specify that when you order them. Your problem is not going to be easy to solve. I'd start by pulling all the thermocouples out of the machine and raising the temperature of one with an isolated heat source, like a clothing iron. You'll obviously need a second device to measure the temperature of the heat souce while you try this. If you find they are accurate when they are out of the machine, then you have an electrical noise problem and the un-grounded units should solve it. Other suggestions from the forum are certainly welcome. We are going into hack land here. Joe T.

No, it is not normal for Socket 1 not to work on 20256. By default, this socket is initialized as a slave and you don't have to do anything. Did you happen to put a Socket Init in your program for Socket 1? If not, please elaborate on "breaking". Joe T.

Starting with Win 7 Visilogic is very picky about being Run as Administrator. You always have to do it. The easiest way is to change the property of each instance of Visilogic.exe in each folder it's installed in to Run As Administrator. Just adding the property to the whole folder doesn't cut it. A pain in the butt, yes. But it's the Microsoft world we live in. Joe T.

Any reason why you're using an M90 in a new application? This is an ancient model, and it's not suited for what you're trying to do. You could improve your speed with an Interrupt routine, but only to 2.5 ms per operation. There are several V130s that have hardware level stepper control. I would HIGHLY recommend switching platforms. What other I/O requirements do you have? Joe T.

You are correct on the decimal point. 18.5 is returned as 185. If you don't want the decimal you have to divide by 10 to get rid of it. If I understand correctly, the value in the PLC is larger than the actual temperature at the probe. I have had problems with electrical noise getting into thermocouple inputs and causing an offset. The IO-ATC8 is not electrically isolated. Are your thermocouple probes grounded or ungrounded? Joe T.

Does this product include a SIM card or is that up to the user to provide? Joe T.

Decimal point control is a weakness in Unitronics when it comes to floating point numbers. I hear you. Use the Font Manager in the graphic editor to load any font you have installed on your computer. Depending on your PLC model, you can go huge. I like to load a 24 and a 36 point in MS Sans Serif. It's the little button with the capital "A" on it or it's under the HMI menu. Joe T.

First, which module part number are you connecting the input to? I think you may have a bigger problem with the resolution of the A/D converter in the module. Our highest resolution is 14 bit, which means you get 2^14 or 16384 counts to play with. You've dialed the span down to 9.1 V to 1.0 V, so you'll get a smaller fraction, assuming you're using one of these inputs: 1.0 V = 1638 counts = X1 9.1 V = 14909 counts = X2 Total span = 13271 counts Your high end value is 1000 (Y2), so the best you can get if the low value is 0 (Y1) is 0.07 mb per count. In other words, trying to read more than one decimal of precision accurately is unobtainable. You can feed the output of the Linear block into an interpolating equation as Kikis points out, but everything past the first decimal will be meaningless. Joe T.

What model of PLC are you using? Joe T.

I dug deeper into their website - 59 Euros out of Bulgaria. I'd be curious if they have a US sales channel. Joe T.

I've never used one of those but they look really simple. How much are they?

If you're talking about something that automatically updates according to the filename - no, there isn't. I just put a text object on the screen and manually update it. Joe T.

It looks like those are AC input supplies. Joe T.

If your products have different thermal response properties you may have to set up a recipe table that copies different PID values to the block based on what you are running. It sounds like one size doesn't fit all. Joe T.

The PLC name requirement is a security thing. You have to know the PLC's name if you're going to be logging in from some remote location- say a telephone modem in China. Joe T.

I usually connect 0V to ground because it makes troubleshooting easier - clip the negative of your meter to a DIN rail and you can poke around with the positive lead. +1 on what HigHTech says - make sure you have a really good earth ground or you'll have problems with your analog signals if you ground the DC power supply. Joe T.

12V PLCs are kind of hard to come by, especially one with any kind of capability. Plus, this is the Unitronics forum so it's kind of blasphemous to recommend competitor's products. I have used one of these 12-24V converters when I've done a vehicle application: https://www.amazon.com/SMAKN-Converter-Module-adaptor-Regulator/dp/B00VRAQZVK/ref=sr_1_1?ie=UTF8&qid=1474326384&sr=8-1&keywords=12+to+24+volt+converter They work quite well, and now you can use any of our products. Joe T.

Flex has the right approach - PID won't work here. Our PID is not capable of heat/cool. You're going to have to write your own control scheme - I've done this with a motorized variac that had a similar control model. I'd make something with timed pulses to run the valve one way or the other depending how far away from the setpoint you are. You can roll your own proportional control by varying the amount of time you output to the valve one way or the other based on this. Joe T.

Your process low and high limits are fine. Heating with steam injection can be a beast. What are the values of your tuning parameters? How fast is the loop updating? It should not be too fast. Autotune doesn't always work so well. Have you tried manually tuning the loop? Start with I and D = 0 and adjust P to get a small continuous oscillation about some value, and make a note of how long the oscillation lasts. This is your process response time. Once you've gotten this, double the P value. Don't worry if it's not at the setpoint yet. Start with a value of I equal to your process response time, and adjust it to get to your setpoint. You probably won't need a value for D if there are no regular upsets to your system. Joe T.

Are you running the program as Administrator? Joe T.

+1 on file missing. The Creators have gone home for the weekend - hopefully they look at this on Sunday. Joe T.

Yes, you can add a communication module to the Samba. The part number for the serial card is V100-17-RS4X. Joe T.

Steven- I'm sorry you can't exchange the controller. The Samba, like the Jazz, is a limited-function product that unfortunately sometimes gets purchased before all the variables are known because the price is so attractive. Then you hit The Wall, as in your case. I've done it more than once myself. There's really no alternative other than to learn from your mistake and use a different controller- you'll literally spend hours trying to make something work that doesn't fit, and eventually you have to look at the value of your time. Older/cheaper servo controls drive me crazy for this exact reason. Anytime I get involved with a servo I use something that has a serial port. It's actually quite easy to use the Protocol block to generate the couple of commands needed to make the servo do whatever you want. The manufacturer's setup software does the heavy lifting of initial configuration, and from an application standpoint you just have to feed it velocity, position, and run commands via serial strings. Joe T.

This looks like an old-school velocity control via input signals servo drive. I didn't see any communication options in the manual, so talking to the drive via serial connection is out. You didn't give the full part number so we don't know if it has the -T pulse control option. You didn't describe what you're trying to do - is forward and reverse required? Do you need precision in rotational position, or is speed control more important? If you're limited to controlling it with a +/- 10V signal for velocity, you've got the wrong Samba model. See if you can exchange it for a -TA22, and then get yourself an external analog converter module that converts 0-10V to +/- 10V. They are available. Let us know. Joe T. PS - @ Kikis - the 500 Hz spec bothered me so much I got an SM43-J-T20 out of stock and put a scope on it. The sourcing output is like the old V120; it will go to 1 KHz with no problem and turns into a triangle wave above 1.5 kHz. The interesting thing is if you run it into a PNP to NPN converter you can get as much as 5 kHz without a problem, as the transistor switches at a low voltage without regard to the input waveform, producing a nice clean output squarewave. This is an un-authorized spec, of course.