OkiePC

If the JSR rung is true, the subroutine is scanned. If the JSR rung is false, the call is not scanned. Bernie covered question #2. In a SLC/Micrologix, LAD2 always runs and is the main routine. I highly recommend you make any routines controlling real outputs unconditional. I only use conditional routines for things like intensive math, sorting data, or things like that. Technically, you can put output logic in a conditional routine, but it can get hairy if you do it wrong, or make it excessively complicated.

Can you post the full error code description? I am familiar with only the M410 series Fanuc robots, but they are likely to be similar messages. There are lots of fuses all over the CPU, power supply, I/O, we had a fuse issue on an M410iW with RJ3 controller, and it took awhile to locate the little black fuse that looked like a pin jumper on the servo board, right under the EE connector header. Check LED status's too, you may have just popped a fuse.

Tri-Tronics has a nice sensor for this application. Very fast and reliable. http://www.ttco.com/industrial.aspx I have used the Markeye Pro, and Smart Eye Pro with great results, very trouble free. We used one of their older color sensors too, the color was "taught" and automatic setup worked quite well. WoW. This one is new: http://www.ttco.com/...rial/xp-10.aspx

The first one discusses a long list of Windows files that may not be correctly registered. You could ask your A/B rep to get the utility contained in that link. But the 2nd one has helped me with the same issue in 5.1 on my WinXP Pro 32 bit machine. It explains that you are running out of so called desktop heap. FTView takes lotsa desktop heap apparently, along with the increase in alcohol consumption and patience testing it requres, it needs double the default desktop heap. :banghead: They recommend doubling it to 1024. http://support.micro....com/?id=184802

I found these nice links showing lots of power distribution possibilities. http://173.247.254.1...istribution.pdf http://www.federalpacific.com/university/transbasics/chapter3.html

I haven't used DL either but I've perused some threads and looked at the demo software. I have noticed they have a new ethernet DL205 "DoMore" PLC with high speed counter cards...nice price, and fully free software with lots of upgrades to the older 205. http://www.automationdirect.com/adc/Overview/Catalog/Programmable_Controllers/Do-more_Series_PLCs_(Micro_Modular)/Do-more_H2_Series_PLCs

There's (*virtually) no such thing as too much resolution. Resolvers are superior in reliablility when exposed to high vibration and mechanical shock. They are also absolute and the decoder determines the resolution...8192 angle divisions per revolution are readily available. *Modern High Speed Counter Cards are very very quick but RTFM to find the frequency limit.

Using internal tags can be very useful. If you are careful with your naming strategy and tag structure, you can change them all from internal to the PLC source after the project is mostly developed. The built in emulator can communicate with a real PLC through your PC serial port in most cases. The built in emulator can do all sorts of cool stuff all by itself, even with PLC source data missing or done completely with internal tags. The built in emulator in Crimson 3.0 requires a 32 bit operating system to install and run at all. I use it for test apps at work where I have RSLinx Classic Gateway, so I have all sorts of quick display programs I run to monitor things. I can double click a cd3 icon, hit download and in a few seconds I can monitor it for up to 15 minutes just as if the real hardware were running. I have not tried to use a real G3 with an emulated PLC, but I suspect the emulated G3 will be better for you anyway, and the real hardware will run exactly the same when you get it properly connected and downloaded. I did have problems getting it to work within XP mode in Windows 7 Pro x64, but I have heard it works fine with a virtual machine (32 bit) running in VMWare through the hardware serial port to a real PLC. The same should be true for a software PLC and redirector software.

You're welcome Ken! Yeah, and once you get it working in Crimson, you can drag and drop the objects, programs, and tags between applications very easily, If there are name conflicts, no prob...just make a new folder with a unique name in the source (programs and tags can be folder structured very easily) and drag the stuff you want to copy into the folders, then you can deal with name conflict after the folders are dragged to the other instance you're working on. For all the things I have learned from you, it makes me feel good to know I gave a little back.

We have a few of these around going on two years with no problems: http://www.inteccontrols.com/pdfs/I-RH-TEMP.pdf

Yeah, I think that parameter is all you get with the lowly "Component Class" drives. Now that I have access to the manual, I did not find any other parameter to give control of the response time.

With that much weight, even on ballscrew driven elevators, I would have mechanical fail safe brakes attached rigidly to the screw itself, not on a reducer or belt/chain drive.

Look for something called software current limit. It is in there, and I am working from memory from a green acres ISP connection without a copy of the pdf. I think there may be an associated level and/or time adjustment parameter too.

http://www.youtube.com/watch?v=CYCQpZcuRvU I wish I knew whether they were using pressure sensing in the fingers or if it was just a position map. the movements are very geometric, but appear to be vision registered. I guess I need to "Google it up"...download the user manual ;) check for a local distributor... I need a two handed robot to stack and pack patties of frozen meat in column fashion into a bag lined box. I think this one with the right grippers might be fast enough: Wow, ABB, huh? Nice. Put some pressure sensitive fingers on two of them and a nice vision system...and that video is three years ago?

It almost seems they sacrifice accuracy for the sake of human contact way too sloppily up front. It seems totally uneccessary when tight torque controls and a little math could keep the precision and speed, just have "collision detection" which has been around for years in many forms at many levels of control from VFDs with shear pin faults to Fanuc palletizing robots with Payload registers and collision guard option. Slow it down when people are around, and the flat panel face is plain scarely looking...sorry, give me a Honda... http://www.youtube.com/watch?v=ltPy3nc02pM

I was excited to read about it, but reading the description and watching it move left me disappointed. Why is it not perfectly accurate and extremely fast? Or at least fast enough to do useful work. The most impressive robots I've seen lately are the Darpa llama and the other mules, along with those based on the same concept as the ABB Flexipicker.

Also, even though your examination of the shortcut on the runtime tab appears fine, go ahead and click on it in the pane, and hit apply again.

What's PA? The quartech modules I recall using between a PLC-5 and an array of indexing Ultra 200 drives was very open, scriptable, reliable as any other RIO link. It was actaully a DH+ to Ultra3000 interface din rail mount, very solid unit. It gave you all the tags in the Ultra and a place to write code to poke it in whatever legal PLC-5 addresses (or SLC) you wanted.

You could use Inductive Automation Ignition Vision Panel Edition for free. https://www.inductiveautomation.com/products/ignition/visionpanel/download

With the Panelview plus, I make a habit of copying the MER from external to internal storage. Then I remove the CF card. Then I load the application. When it asks if I want to replace the communication settings, I answer yes, even though for ethernet commns there are more steps. Once the app is loaded, I go back into the terminal settings, startup options, and select run current app on startup. Then I go into the comms settings to make sure that the comms are set correctly. If the IP settings need to be changed, I must do that manually (sometimes I do this first). Then I reboot the thing to make sure it does indeed do what I expect and to make sure I didn't miss a step. Once it is up and running, I stick the CF card back in there...I like to keep a CF card installed in all the PV+ units, but don't want to make the unit dependent on it for normal operation...I don't want to always load from external storage...I want to be able to "borrow" that CF card at any time, but try to keep it installed in the unit mainly for the purpose of midnight cloning when necessary. I keep more CF cards with multiple files on them in the office, but in each PV+ I like to keep only the latest and greatest file just for that installation. We have some OEM machines with varying levels of revisions, so this helps keep us from guessing the name since the OEM decided not to use any sort of recognizable naming conventions. I could rename all their files, but they lock their code and I'd rather keep that line between what we do and what they do very distinct.

That servicing selection bit is often just set manually (RSLogix500 right click, toggle bit) sometimes during troubleshooting, and saved that way, although it should be hard coded if used. The 10Mb half setting is probably leftover from a prior SLC with earlier hardware e-net daughtercard...both settings may be legacy settings since your new SLC supports full duplex/100Mb and at least twice as many connections.

For motor power cabling, use rigid conduit and follow the path with the diameter needed for less than 40% fill. If they are long sections, with motors spaced far apart, you want motor power wiring in conduit up to within 6' or less of the motor j-box. If your safety system is zoned, then distribute power panels by zone and this can greatly shorten the pipe runs and make them smaller diameter. Use electrician2.com calculators for motor branch circuits and conduit fills... If you are using armored VFD cable, I still like it inside pipe, but prefer parallel runs of 1" or smaller to 2"! I think there is an NEC rule that limits how many motor cables can share a pipe. Some of the (other) forum frequenters are electrical code experts, so I am sure they'll chime in. Paul

You could use a pop up page or roll your own with text object. This example may get you started, showing how to manipulate the text (scrolling list) in each text object label.

Could this be an electrical disturbance? Try a UPS yet? Any unusual loads starting up at that time that could affect voltage? We have one 1756-ENBT module that drops its link light once in a great while at random intervals, this is for some data collection and one hmi, and we can usually recover by RIUP with a nice pause between the R and the I. I think Mike finally replaced it with our spare after several months it happened on a weekend so he took advantage and we are going to reflash it. During the spring we definitely had network outages caused by electrical disturbances (thunderstorms, compressor motors failing, and even a huge HV transformer couldn't stand the 115F heat here last month). Several of these times, we had one or more 5/05 or switch or ethernet HMI that had to be "bumped" to get talking again. Even though they're powered by a UPS, the brownouts that last more than a few cycles but long enough to drop out good tend to be the most harmful to contollers in general. Why I remember that was a "cheat" we used to do with Atari Space INvaders. If you power cycled it real fast, but just right, you could have two missiles at a time. The other 90% of the time the game would freeze with a random garbled screen and sometimes a stuck noise channel... Paul

I.ve recently seen this twice, once I think was just the backplane, the latter case was the backplane and power supply (which measured good but wouldn't allow the SLC to boot). What I saw both times was no flicker and blank pause before the LED comes on, but it came on within a second or two after power was applied, and stayed solid...normal faults still have a LED time pattern that you get familiar with, and these two SLC didn't even get that far...