My project

[Pulsar2003 0]

Hello. I'm new to this forum. Here is my project: Getting data for five ovens. Why: Loosing production. And replacing older Nais FP-1 PLC. Here is are a few details: A complet cycle includes: a cooking time, cooling time and unmolding time(if that is how it's said in english)(to removed the cooked part and fill with plastic powder). Some times there will be some interference because an oven has two carts to that there will be a delay before the cycle can start. http://www.rotomachinery.com/content/view/...0/lang,english/ I'm starting with a first oven. So I have a computer in the supervisor's office with a 1784-KTXD card (So I will use DH+) a SLC500/04 cpu (SLC1) and a 1200e Panelview(PV1). So I'm gonna have them connected in that particular order. Next upgrade will be Supervisor's computer, SLC1, PV1, SLC2, PV2. The resistor will be moved. I thing I got that correct. I doubt I will have 1500' of wire. I still wonder if DH+ is better than DH-485. What I need to know, and that is if I go the the strict minimum of what informations I need, I want to know at what time an oven gets started Sunday night when the week begins and if there are significant pauses in between cycles throughout the entire week of production. Now if that can be done, I'd like to make statistics for every employees. They would need to log in a station (so we get the logging time) and we can build a proper file with their personal stats. And in the best of all worlds...each products have a specific cooking and cooling time, this means that I would like the have a file with all the receipts that I can transfer directly from the supervisor's computer to the proper station. I did check out the free software linked here. I only have one life!! I want to stop here and see what can an cannot be done. Thank you.

[MikeSmith 0]

Hi, This application is very similar to some FTMetrics jobs I've done, you may wish to have a look at that. Be careful with the amount of traffic going over the network though, it's quite easy to overload with any data logging. ~Mike

[Pulsar2003 0]

I really like AB's software. great products. But quite expensive. Can I do this with Visual Basic? Currently I have an other over (more recent) that uses a computer HMI to communicate with the PLC and it has some of the features mentioned in the earlier message. We do have a log for changes made to the recipes, alarms. Would it be better to use a 1761 NET-ENI converter from the DH+ to Ethernet? And I could add a communication card (Remote I/O) to go from PLC to HMI. Currently my RSlinx doesn't recognize the converter. Possibly RSLinx Lite. Thank you for the tips.

[BobLfoot 302]

My esteemed colleague Paulengineer may disagree but I would stay with DH+. DH-485 is limited to 19.2 Kbaud where as DH+ is 56.2 Kbaud minimum and depending on distance an clock out at 230 kbaud. Now as to your suggestion of a 1761-net-eni. Avoid this. Yes ethernet sounds nice but the 19.2 or 38.4 serial port between your SLC and ENI will be a bottle neck. Try this idea on for size. It can be executed in stages. 1. Stage 1 - Use a 1756-P72 Power Supply, 1756-A7 Rack, 1756-DHRIO Card and 1756-ENBT Card to create a Bridge or Gateway between Ethernet and DH+. 2. Stage 2 - Add a 1756-L63 CPU to the rack and repalce the SLC's with 1747-ASB modules using the RIO feature of theDHRIo card now and use the power of the ControlLogix to run the system. 3. Stage 3 - replace the SLC I/O Racks with ControlLogix or Flex I/O racks. BTW - unless process or other needs dictate so an L63 can probably repalce 2 SLC 5/04's if coded wisely.

[MikeSmith 0]

I think you could do it with VB or a programming language. FTMetrics only uses 2-3 tables and pretty simple logic, you should be able to implement the same sort of thing technically (no claims made by me to the legality of this) The quickest(/dirtiest) way would be using MS Excel, get the data from the PLC via RSLinx and DDE (really quick and easy) and then insert sets of these values to a table in MS SQL. Then use a bit of T-SQL to format, compress, present the data etc. Do your reports in SQL Reporting Services. I won't even pretend to guarantee the reliability of building on top of Excel but you should be able to get something like the above out in about a week or two if you know all the products involved pretty well.

[paulengr 44]

Yes, on the first one or two. Also, the NET-ENI product tends to arbitarily lock up every few months or so. An ex-AB engineer I know said that it is very finicky about power. I'd go third-party (Digi One) for this both for price and reliability reasons. Speed won't be any different but it will crash less. What you have to consider here has to do with infrastructure. DH+ is so proprietary that you can't for instance run it over existing Ethernet cabling along with everything else, so you will have to make special cable pulls everywhere you want it to go. The second problem has to do with the inability to multiplex it (yes, you can make bridges with 1756-chasses but they are bridges, not routers or switches). Think of your house. Typically fixtures have 1/2" water lines but the trunk system is 3/4". If the trunk was 1/2", you would have real problems running the sink and flushing the toilet at the same time. At 3/4", you can run roughly 3 "appliances" all at the same time. DH-485 and a NET ENI module are like 1/2" lines. DH+ is sort of like a 1" water line if you run the higher speeds. And Ethernet is sort of like a 6" water line. Once you get past about 6-8 devices on a single DH+ network, things get very slow. In comparison each "drop" is very limited with the NET ENI or DH-485 method. It is slow (almost as slow as 8+ devices on DH+) but because the Ethernet backbone is so "big" the slowness stays relatively uniform even as you scale up to dozens of systems. Plus, you can't talk to your PLC across the continent, bounce through wireless access points, access it from home, etc. The really insidious part about DH+ is that it comes "for free" on most SLC & PLC-5 systems. Ethernet costs extra, a lot of extra. On the PC/device side, DH+ and DH-485 are very expensive, enough to almost eat up the cost of buying an Ethernet-capable processor. Connecting PLC's with DH+ or DH-485 is relatively cheap. Connecting anything else gets expensive, and there's that pesky "small pipe" problem. So...it depends on your application, and future scalability. I agree for the most part. There is a time and a place to use DH+. But be careful of it's limitations. And realize that ultimately the best thing to do is to get away from proprietary protocols like this. The problem is that you don't want to be hip deep with over a dozen PLC's all talking DH+ to a bunch of old PC's (kept around because of the problem of finding hardware that supports DH+) and the attendant maintenance nightmare and the fact that you can't just troubleshoot easily from anywhere in the plant and that at that point, "modernizing" is a very, very expensive problem. Don't let your DH+ networks grow. Before they do, switch to something else. It is cheaper to invest in CompactLogix or ControlLogix PLC's, or even Micrologix PLC's (which aren't half bad in their own right) than it is to try to upgrade to SLC 5/05's or PLC-5's with Ethernet ports. Edited 1 Oct 2009 by paulengr

[Ken Moore 23]

Depending on the complexity of your project, you may not need a SLC processor, perhaps a ML1400 would do. You do not provide enough information to determine if this is the case. I have just finished two small applications that I used the ML1400 and Redlion G310 on, I really like the ML1400 and the built in ethernet. The ML1100 is okay, but too limited on expansion I/O. If you could use the ML1400, you would save lots of money and be able to network everything over ethernet.

[Pulsar2003 0]

Ok. Where do I start? I currently have many SLC 500 components. There will be a good deal of communication between the PV1200e and the PLC. There will eventually be 5 PLC one for each oven and the PV give infor and control to right and left. Now if there is too much trafic I wonder if I could use the RS-232 and the Net-Eni to link every PLC to the supervisor's PC and use the DH+ from the PLC to the HMI?

[Pulsar2003 0]

Ok. I've checked on AB's software. They have really great features. What is the down side is the coats. Any body can give an approximation of this software? Thanks!

[MikeSmith 0]

For the Rockwell products try http://www.rockwellautomation.com/distributor/uscanada.html Make sure they include the necessary MSSQL licences for the database. Oh, also, one downside of Metrics is that (when i last touched it) it didn't have the ability to change records after (and definitely not while) they had happened. If you need this, ask for a .net web application to do it, there's a few versions floating around Rockwell (one of which I wrote). Also, I'm thinking ahead a little bit here but keep it in the back of your mind, I would suggest not (from painful experience) to embed ActiveX's to either report or talk to the database in any HMI. the reason is that single threaded HMI/SCADA's hang while the database either replies or times out (the HMI freezes if the database is not accessible for the time-out period of the database). Edited 4 Oct 2009 by MikeSmith

[paulengr 44]

From experience, yes, RS Metrics is incredibly expensive, roughly equal to about 6 months of an engineer's time. By the time you finish developing an equivalent, you will have spent that much time or more. So it will save the company money by buying it. Or an equivalent software package (there are several similar packages out there...these are MES systems, avoid any which look like they are catering to accountants...they are rehashed attempts to make an ERP system work in the non-accounting world...round file all those).