PC based control system 1

[vanskat]

I am in the process of automating a small scale limestone processing plant and am looking for a cost effective PC based control system.

There a few I/O's (temperature, pressure, current etc) that I want to be able to monitor, control and historize. However, I have been asked to keep within a tight budget.

I have taken a look at "PANELMATE" which is functional but a bit of an overkill for the application. A section of my design includes the installation of VFD's that will be controlled by any of the parameters listed above.

I am looking for a PLC based system that can communicate serially with a PC with associated analysis software; something that is scalable and functional...

any suggestions?

 

[d23]

vanskat

I'm not sure if this is something like you are looking for. I don't want this forum to become a sales pitch so look at the site if you like what you see please contact them for prices. Compared to full SCADA this is low cost.

http://www.slb.com/Hub/index.cfm?id=id1140167&printable=true

Of course this is only one possible option available to you.

Good luck!

 

[SERVOCAM]

You can look at your Bigger name PLC Mfg's and their software. AB & ControlLogix, GE Fanuc & Cimplicity...there are also programs like Wonderware, Steaple Chase, ASAP. Try looking at Horner Electric and their OCS.

I'm a motion control guy, so maybe some of the PLC people here can help more. Cameron Anderson - Sales & Applications Engineer
Aerotech, Inc. -

http://www.aerotech.com

"Dedicated to the Science of Motion"

 

[EdDanzer]

Galil motion control has a modular controller that might work.

http://www.galilmc.com/products/eseries/ioc7007.html

 

[vanskat]

Servocam,

Did some research and found wonderware to be a very interesting product. I have spoken to a few users who are surprised by the functionality.
I am in the process of choosing a PLC at this point

 

[jsalls]

I would start off with Automation Direct. They sell some very reasonably priced PLC's, Operator interfaces, software, etc. Their products are made by major manufacturers at a fraction of the cost of most.

http://www2.automationdirect.com/store/Home/Home

Good luck!

 

[gcaudill]

How about writing your own control software and doing discrete I/O and analog I/O from the PC? You can write all of your MMI screens in a visual programming language. You can get PC cards, etc. to do field control. You could also use higher level communications protocols to download parameters to drives, etc. using Modbus, RS485, etc.

Just ideas for a shoestring budget.

 

[jraef]

I agree with gcaudill.

If you are going to use a PC with a Graphical User Interface anyway, why use a PLC? There are plenty of inexpensive and rugged I/O systems available now that can be controlled directly from the PC, such as Opto-22, Sixnet, Phoenix Contact just to name a few. Go to a search engine and enter SCADA, which is short for Supervisory Control and Data Aquisition. You will get hits on hundreds of suppliers. PLCs have their place as stand-alone controllers, but unless you want the redundancy, having a dedicated PC talking to a PLC is kind of a waste. Quando Omni Flunkus Moritati

 

[vanskat]

thanks for the feedback.

My initial thought was that a PLC would be a bit redundant. However, going the route of writing control software etc seems a bit time consuming and too much of a headache. My last experience with the above was not too memorable.

I guess I am looking for an easy, quick but cheap way out.
It shouldn't hurt to do a bit more research!!!!

 

[gcaudill]

Fairly painless is the PLC route. I do not want to advertise products so I will not mention names, but there are some dirt cheap PLC products on the market that are very capable.

However, adding SCADA and remote monitoring does notch up the complexity (and costs) a bit. In this application spending some money may ease the implementaion but will drive costs. It's about what is of most value to you.

 

[jraef]

If you do not want to write programs, that shouldn't be a problem. Some of the GUI software mentioned earlier, like Wonderware or Intellution can be programmed to communicate to your devices with simple drag-and-drop object oriented graphics symbols. It doesn't get any easier, but it will probably not be cheap (depending on what that means to you).

Lets change it around a little then. Why do you want the PC? If you want cheap and simple, I then agree with gcaudill because micro PLCs have become very very cheap. If you want an Operator Interface Terminal (OIT) without the need for import/export of data to a higher level program, then check out companies like Total Control Products (TCP), Spectrum Controls, Exor, Maple Systems or a host of others that have low cost color or B&W screens with built-in PLC communication interfaces, some even with touch screens.
Quando Omni Flunkus Moritati

 

[vanskat]

I guess I'm between a rock and a hard place on this one.

Normally, for a large industrial type installation cost is not an issue. However, this is more of a back yard type plant (business venture) with a total of only a few dozen I/O's.

Being more of a favor for a friend, I do not want to unnecessarilly incur high costs before the venture becomes lucrative. I am still struggling to decide what is actually feasible for the application. I confess that I am trying to accomplish alot for the small scale operation but I wanted to provide some amount of flexibility and functionality.

I guess bottom line I need to consider fundamental needs and develop a solution that satisfies that need.

The primary reason for choosing a PC based system is location. This plant is remotely located and as such remote access would be a nice touch.

There are so many products out there to choose from, so it's a bit difficult to find one that gives the biggest bang for the buck.

 

[jraef]

My suggestion is to go with a PLC system and an inexpensive OIT. If your VFDs are from one of the biggies like AB, Toshiba, Siemens etc., their PLCs may be able to communicate directly to the drives like an I/O rack. That would help with simplicity. If not, choose one with an RS 485 Modbus comm port, since most VFDs today can communicate via Modbus. If you have older VFDs that can't communicate, it's all moot anyway. Put off the PC for a later upgrade since history and analysis are pointless if they go under ;-)

What VFDs have you got now?
How many analog I/O?
What kind of expandability do you need?

Keep in mind that free advice from anonymous sources is a dangerous way of engineering a system. This is just a signpost to help you along the way. Quando Omni Flunkus Moritati

 

[vanskat]

Jraef
Completely understand your comment about anonymous resources. That can be very dangerous...

My problem is that I am not very used to working with a tight budget. I have implemented many systems and my philosophy was that if you want functionality it should be paid for...

Honestly, seeing the different postings has allowed me to think on different levels..and of alternatives

Your last post is what I had settled on last night. I am going to buy a couple of Control Techniques VFD's (unidrive to be exact) since I have installed several dozen of them!
Currently, I have say 24 analog I/O with the possibility of going up to 64 in the near future.

 

[wired1]

If you didn't have to "historize" you could get by with micro plc and a panelview. Packages like Wonderware are good at historizing, trending, alarming, etc. But that's a whole other dimension that makes the system a lot more expensive. Maybe if you get off the plc kick, one the pc-based controls would be better. I come from a plc background so I can't help much on that.

 

[peebee]

Just making sure you haven't overlooked any other options:

Do you happen to have a building automation system or security system in place that could take care of your needs?

Do you have any existing Modbus RS485 data gathering devices?

Also, Ethernet-based data gathering devices are becoming more and more prevalent every day. Probably would lose the cost battle, but they might be worth investigating.

 

[vanskat]

Peebee

took all that into account..but have none of the above

got a budget figure today and have plotted some sort of path for implementation. task now is product hunting!

 

[peebee]

Well then, as a wise man once said, "if you want functionality it should be paid for..."

 

[skills]

If you want great performance with ease of development for a price that is cost effective, I'd say give the Lookout scada system a long hard look. Lookoutdirect is a great value also, if it's still offered as a bundle. Lookout has 1st day productivity, which is not the case with Wonderware, and doesn't load the computer as much because its core is more efficent.

 

[skills]

I forgot to say where info on Lookout is available. It is a National Instruments product

http://www.ni.com

and

http://www.automationdirect.com

has lookout-direct for $1295. This is a low cost variant high a limited tag or I/O count. With this and a small plc costing maybe $350-$400, you will have easy development and scalability along with internet access and the world's fastest database, 'the Citadel engine' to store your trending data.