I have an application where I am looking to replace a heap of old relay logic controls with a shiny new PLC that can integrate with the rest of the controls system.

The control cabinets are subject to rather nasty vibrations from the surrounding machinery, and there is a small amount of dirt/oily contamination that accumulates in these panels over time.

I am currently looking at using the A-B MicroLogix 1400 as the PLC, a selection I am very comfortable with from a cost/reliability standpoint, but I am trying to determine the best method to reliably control the process while making it easy to troubleshoot and replace faulty parts.

The PLC will be controlling two motor starters and some solenoids for pneumatic valves. The relays in the current panel are massive A-B Bulletin 700 10A industrial relays with 4-8 poles, but with the PLC we only need simple on-off control of the solenoids and starters since the actual logic is of course executed inside the PLC.

The MicroLogix 1400 has 12 internal relays, with contacts rated at 2.5A continuous current. The solenoids are in the 8-12W region @ 120VAC.

For troubleshooting purposes, I think it would be better to drive external relays to control the loads, since it gives a modular piece to replace if something goes bad rather than the entire PLC brick, but then I will need to get the TRIAC/transistor I/O module, and have the choice between using SSR or more electromechanical relays. I am not enamored of the idea of using the internal relay contacts to drive an external relay, since now I would have 2 mechanical components to fail instead of just 1. At that rate I am better off just using the internal relays.

I guess my question here is, am I better off forgoing the extra complexity and just controlling the loads from the relays internal to the PLC? The valves will sit in one position for several months at a time, does anyone have experience with the reliability of the A-B internal relays as far as contacts welding or oxidizing?

I am tempted to drive SSRs, but the loads will switch only every several weeks, and an 8W solenoid draws barely enough steady-state current to keep most SSRs I have looked at switched. I am also concerned about the possibility of an SSR failing closed, but I like the idea of eliminating as many moving parts as possible, since the vibration is a real concern here.

So now I ask the other professionals on control.com for their opinions. I can give other information if needed, just lemme know.

Thanks in advance for the input!

 

I assume that all your output loads are 120VAC. Given your concerns about reliability, I would suggest using a PLC that has 24VDC transistor outputs and using those to drive external cube style relays. The transistor outputs will probably never fail, and the cube relays can be easily (and cheaply) replaced.

Transistor outputs tend to be much more reliable than relays or triacs. The difference is particularly noticeable on newer PLCs, where the output relays seem to fail much more often than in older PLC designs.

Larger external relays such as cube or octal base relays tend to be pretty reliable provided you oversize them for the load.

As for vibration, I haven't noticed any particular problems with PLCs in that regards. That is a matter of degree however, so it's hard to say. I've never seen a case where that has been a problem though, even on machines with vibratory feeders. If it hasn't been a problem so far with the existing controls, it's not likely it will be a problem with a PLC.

As for oil and dirt, that can be a problem if there is too much of it as it can get into everything and gum up the works. Also, if there is oil getting into things, it will often be mixed with water which can cause corrosion and short circuits. Sometimes certain types of oils will also react with the insulation or plastics. The solution in this case is to find out how it is getting into the enclosure and to stop that from happening.

 

I'm right with you on the driving relays with relays issue. But I've seen it done in many places. The key here, I think, is the very low switching rate. If cost is no object, the transistor/triac outputs with replaceable relays might be the way to go, with suppressors on the outputs. That would be the gold standard way. But with only very infrequent switching, it would be hard to justify the cost, and the relay outs, used within ratings, should be fine. One thing that may make the decision easier is that you often need isolated contacts when replacing relay logic. The MLs I've seen had bussed relay contacts, usually broken up in groups. If that will work, I'd probably use the PLC relay outputs.

If you can't have the common contacts, then I'd get the solid state outs and use relays. In either case use suppression or you don't have to worry about which will be more reliable because neither will be. This is neglected a lot these days and then the relays are called unreliable. If you can see a relay arcing from the other side of the plant, it's not going to be reliable.

Regards,
cww