Using FLEX I/O without AB Processor

NevisGroup said:

You should be careful making such statements about a machine that you have never seen and a customer that you do not know.

Click to expand...

Getting a little snotty at people trying to help you go in the right direction aren't we.

Ken is right. The cost of your time is often what gets underestimated in these situations.

If you are dead set on avoiding the Logix CPU (and the cost of software to support it would be another compelling reason)...

Try a P2000 PLC which supports Ethernet I/P. You might spend 100 hours (which is only 12.5k dollars in my world) figuring out how to get it to work and would need to swap the C-net adapters with Ethernet I/P adapters, but I am pretty sure it is do-able.

The little P2000 is an incredible value and the software is free. It is not as good as Rockwell software in most regards, but never balks looking for a license file and has a bunch of other advantages.

It wouldn't surprise me if someone has already done the 100 hours of homework already and can save you some of that ca$h too.

https://www.youtube.com/watch?v=OLmhO_iTYhw

I just setup a P2000 to use Ethernet I/P to get data from some Opto22 Snapio hardware. The setup was super easy, it took longer to get the Snapio setup (also super easy) and configured for enetIP than to get the data streaming in the P2K.

the P2-550 (AutomationDirect Productivity) does EIP

My tone was definitely dismissive, and I apologize. It's been a very long week, and I am acutely aware of the costs of re-terminating and re-labeling wires. I am working on a proposal that uses network bridges and will require significant programming work, but will save me literally hundreds of man-hours (and Man #1 is me) underneath the machine frame covered in itchy abrasive cutter dust.

So let me emphasize the technical and practical aspects of my recommendation, which were implied and probably obscured by my attitude.

A 1794-AENT(R) module will snap into the place of a 1794-ACN(R)15 in seconds, and will present identical tags to the Logix controller program. Not just similar tags; they're be literally identical. You can copy-paste the I/O objects from the L55's I/O tree under the 1756-CNB into your project. As long as you leave the Adapter names the same, the project will verify right away.

If you have the original Logix program and can re-use the logic that runs those machine components, that means that you don't have to re-label or re-terminate a single wire. Any HMI work that was already done for this part of the machine can likewise be most easily re-used.

CompactLogix is fairly capable these days. You don't mention how many adapters you're running, but even the little 1769-L18 controller can run eight I/O adapters on EtherNet/IP. Move up to the 1769-L30 and you'll get sixteen, and the -L33 will run thirty-two.

Your customer can buy a less expensive edition of Studio 5000; the "mini edition" supports only CompactLogix and only relay ladder routines, but it's a fraction of the cost of the editions that support all languages and the full-size ControlLogix CPUs. Get your customer into a copy of that, and you should be eligible to download updates for your own Studio 5000 software.

In my opinion, this is an opportunity to spend a little extra money on hardware in order to greatly reduce your wiring and labeling and documentation labor. If the side effect is that you get up to date on the most popular control platform in Wisconsin, that's just a bonus.

NevisGroup said:

However we would prefer NOT to put an AB processor on the machine. (Mainly due to to cost).
I am thinking Modbus TCP.
I want to keep this as cost effective as I can.

Click to expand...

Your statement about cost saving is stink! I am willing to bet with you that if you can come up with more cost effective system than a drop-in compact system.

I'm 100% with Ken on this one. One of my clients has the mantra "spend on the hardware, save on the software".

Occasionally I find myself in the same basket, and occasionally I can't convince the customer that the upfront cost is worth the long term gain. I could count on one hand the number of times I haven't been able to say "I told you so" less than a year down the track. I mean, I try not to actually say it out loud, but you better believe I'm thinking it.

I certainly did not intend to start this kind of debate.
So let me clarify a few things.
I understand the difference between cost and price. I am not looking to be cheap.
First, the customer does not want AB. They don't have any other AB in their plant.
This is an 'out of budget' expense for them. This means that the money for this project is coming out of someones hide.
They are a VERY good customer. Easy 6 figures of billable time / year.
I will probably just give them the software effort. (If I spend more then 2 hours on it, I will buy everybody a beer)
The program will be strictly mapping inputs to outputs. Thats it.
It is quite feasible to rip it all out and just hard wire. ( I think that will be more expensive actually )
I will have my shop guy do most of the field wiring. ( I figured $500 in labor tops ... might even eat that too)
There is a pretty good chance that this whole thing will be mothballed / scrapped with in a year.

Believe me I have much more profitable things to do than try to pry nickles and pennies out of this machine.

Separately:
As to AB being expensive. I am caught in a bit of a viscous circle with them.
I don't use much AB because it is too expensive. It is too expensive because I don't use much of it.
For your basic blink the light / turn on the valve projects AB just can not compete with the little bricks out there ( CLICK )
For larger projects I have been very happy with Beckhoff.

I am sorry if my comments / replies may have come off as snotty or disrespectful. Certainly not my intention. I though I had asked a simple question. But in hind sight I had all of the above information. No-one else did.