Joe Tauser

Hangfire- You didn't mention where you are located so I Googled your IP and it came up Georgia, USA. I am in Missouri and use a company called Kore Telematics - www.koretelematics.com. I use the GSM modem offered by Unitronics and it has worked well. Joe T.

Can you upload your program for us to examine? Joe T.

Have you looked under Help->Examples? Otherwise, post back with the model of your PLC and I'll scratch one together for you. Joe T.

Look in the Help on SB314 and the Backup Security function block - perhaps this will achieve your goal. Joe T.

I have had a problem like this with grounded thermocouples on a floating system. Yes, everything was supposed to be grounded together but a volt or two of ground imbalance will wreak havoc on the IO-ATC8. The IO-ATC8 card is not isolated, and noise from the system can cause very odd behavior. The first thing to do is disconnect all the thermocouples from the card and replace each one with a short jumper (I use a cut off paperclip) across the thermocouple terminals. You should read ambient temperature on every channel. If you don't, there is something wrong with the hardware or the configuration. If you do read ambient, re-connect the thermocouples one at a time until the problem shows itself. I had to replace all my thermocouples with ungrounded ones to solve the problem. Joe T.

Need the model number of the Jazz you're working with and as much information as you can provide about the device in current loop. Joe T.

Damian - good description of the mystery maximum value. We still need to know a few things to be really helpful: 1. What is the encoder counts/rev? 2. What is a typical pre-scale value? 3. What is the implied precision of MI 2 and MI 3; you mention 1 to 240, but are these indeed supposed to represent integer values? If you don't use a reload type high speed input then you can use an ML as the accumulator. This application looks like a feed-to-length program and you probably won't lose a lot of precision if you reset the counter in ladder logic. You're not using the reload event MB 18 anywhere, and you enable the reload MB 19 with a real input. This tells me that you don't need the micro-second timing the reload function provides, and it's really not the right way to use this function. The "Enable" input on a reload high speed input does just that - if the counter is > the preset then it resets. I think you're treating it as a straight reset input, which it isn't. I'm going to go on the assumption that 1 to 240 represents inches with no decimal place. I'm also assuming that MI 2 is a slowdown preset and MI 3 is a cut preset. This is a typical method of feeding. Since the reload type high speed input really isn't needed, I've moved the input to a regular high speed input and used an ML. I've modified the shear logic to reset it when I7 comes on. Also - high speed inputs do not remember their value when power is cycled- they always go back to zero. I've got a standard bit of logic I put in to handle that. Joe T. J3920 Trim Line HMI Touchscreen - no home position JT.vlp

Whenever you ask a question like this you need to upload your program. None of us have The Great Karnak's turban. Joe T.

Dan, Brown-nosing will get you everywhere. I appreciate your comment! Joe T.

Boris- Upload the program for this one so we can have a look. Joe T.

If it's only one pulse you won't be able to measure the frequency - the HSC input averages the pulse counts over a period settable from 100 ms to 1s. As soon as your pulse is gone the frequency register will reset itself back to zero. A timer won't work well for this; the best resolution you can get is +/- 10 ms plus the scan time which is typically 3 -10 milliseconds (depending on your program size). Read the Help on the 1.25 ms interrupt routine and set it up to to use an immediate input to increment a counter as long as the pulse is high. The routine solves every 1.25 ms, so you move the count value to another MI when the input goes low. Then it's a matter of multiplying this count by 1.25 to get the pulse width in milliseconds. Joe T.

Probably a dumb question, but in UnitronicsLand "Shaft Encoder" means a quadrature encoder. Is this the kind you have? It would be helpful to list the encoder manufacturer and part number. Joe T.

The right way to do it is to add a sensor to the clamp closed position. A lawsuit will be way more grief than adding a sensor will cause you. Here's logic that will work if you insist on doing it your way: Joe T.

First, the legal aspect - if this is for a press then OSHA requires a dedicated two-hand anti-tiedown relay to run the buttons into. Now to answer your question - can't you just put two normally open contacts in series to run the timer? If either one is released the timer will reset. If you're trying to verify both were released before the next cycle then it gets a little more complicated - is this what you're trying to do? Joe T.

One master can talk to several slaves but it can only talk to one at a time. The Function in progress bit in the Modbus config can not be on when you call an RHR block. You need to manage who's "turn" it is. Post your code so we can see what you're trying to do. Joe T.

You can't make the HSC input selective like that, but you can utilize the 1.25 ms interrupt subroutine to get close. If you use an immediate input inside the subroutine, you can make sure it stays on for more than six calls (7.5 ms) and then count it. Joe T.

+2 I've had to restart a download more than once because I turned my attention to something else for just a minute. Joe T.

As usual, I read the other post first and replied before reading this one. Otto - do you need the ability for the operator to change the value in MI 151? As Simon pointed out, SI 31 is in its own format: HHMM. This is not a standard conversion from anything but rather a way to display the time in an easily human-readable way. If you need to set this time from within your program then you need to have separate MIs for hours and minutes. Multiply the hour MI by 100 and add it to the minute MI - the result will be in the format needed to compare to SI 31. Joe T.

An integer register has an inherent HEX value; all you have to do is Store SI 31 to an MI and you'll get the same "value", but if you view it in HEX mode the numbers won't mean anything. What is the end result you're looking for? Joe T.

The Select Operand box has to be somewhere. Do you have multiple monitors? Joe T.

Are you using AC or DC excitation? Do you have another load cell to try? You're to the point where you need to look at the actual electrical signals to determine whether the problem is in the loadcell or the LC-1. See if you can get your hands on a 5 1/2 digit meter to read the excitation voltage and the signal voltage at the zero point. You'll need something with at least 0.01 mV resolution. Then read it again after the offset shows up. Without actual electrical input readings to the LC-1 you're just guessing on this kind of problem. Joe T.

TIm- Yeah, I forgot to mention the Connect/Close requirement. Which I should have remembered, because the first time I played with Modbus IP I didn't realize that the Connect is required and I treated it like regular Modbus. It drove me nuts untill I looked at the V280 Modbus IP Master example. I mentioned in another post that the Modbus IP Help needs to be updated to include this "minor" requirement. Joe T.

You can absolutely use a single socket for many slaves. In the Modbus IP Config block you define the IP address and ID of each slave. In your Modbus read or write block you then use an MI instead of a constant to pick the index of the slave you want to communicate with. Or you can have multiple Modbus read/write blocks; just make sure you only call one at a time and check the overall Modbus IP status bit (defined in the Config block) to verify that it's not busy doing something else. Nir is setting his up to have simulataneous conversations. You need a separate socket for each if you want to do that. You'll have to look at your application and decide what kind of response speed you need and determine if the "slower" response time of multiplexing a single socket is acceptable. Joe T.

I went on a quest for one of these a few years ago- Google "panel mounted serial printer". There are quite a few out there and interfacing with them isn't bad. They all seem to run on 5 VDC which is kind of a pain in the butt, but you may be able to find a 24V version. Look for something that uses standard rolls you can buy at an office supply store. You can also Google "24VDC to 5V converter" - there are also plenty of those available. Joe T.

Glad to hear it's behaving. I would put 30 days of non-faulting operation as a 99% success of solving the problem. I never say 100%. Overall, this post has been a very good overall conversation as to the intermittent mystery problems and solutions that are often encountered with a real installation. If you've solved the problem, then it wasn't large frequency drives or three phase motors but seemingly trivial cooling fans and solenoids. The devil really is in the details. For the record, the "Spark Quencher" is an old-school RC snubber. We put them on every inductive device as a matter of practice and I buy them 100 at a time. You don't want to run out. I'm sure I'll be referencing this post in the future. Joe T.