Why pay for PLC programming? continue

[CaseyK]

Methinks that the programming software should be built in, like in Moellers "easy".

Plug the laptop into the RS232 or 485 ports, and the internal plc software will interface with the laptop, and you just program away.

Beceaus software is in the PLC, it is FREE!

See, KISS, all you need is a DB-9 to a DB-9. or something like that!

regards.....casey

 

[akreel]

Methinks that the programming software should be built in, like in Moellers "easy".

Is it REALLY???
Man, I think I played with one of these and I didn't know that!!!
I installed the software off a CD.


I had trouble trying to get the Moeller to use a modem. Could it be something in the "built-in" software that was giving the comm-link heartburn?

AK

 

[CaseyK]

Google up Moeller Easy demo and you should find a link showing how it is done. CD should have had an interactive demo too, but my CD is pretty old.

regards.....casey

 

[Terry Woods]

This is the same ol' argument about "My guy in a Red suit can beat your guy in a Blue suit".

What people seem to miss is that if you take off the suits, they're both just a couple of guys.

Each is capable of doing the same thing.

A CNC, PLC, and a PC are controlled by the same kind of critter... although there is no doubt that the PC is using the latest and greatest version of that critter.

I wonder if there are many PLC's out there still being controlled by the 8088 (or close to the 8088).

I wonder what the average age is of the CNC and PLC technology.

I know that NASA has a hard time moving away from that era of CPU... with good reason. I think that, maybe, the 8088 is the definition of "solid".

As I recall, the advent of a math co-processor is a relatively new thing for higher-end PLC's. (Some of you younger guys can't even relate to the idea of a processor without a co-processor... however, there was a time...)


Be that as it may...

The primary difference between the PLC and the PC is the "package" around the processor.

The "package" consists of harware components and software components.

You can go with either a PLC Manufacturer sanctioned package or you can build your own. If you go with a PLC Manufacturer sanctioned package then you are also using their software.

If you build your own, then you are not necessarily bound to one software package or another... you have access to a variety of packages. Those software packages are much more un-restricted in terms of their access to the processor (and co-processor) than the manufacturer-provided packages.

As always, this is good news for the Formula-1 Racer Mechanics, and not of much interest to the backyard-mechanic.

Depends on what you want...

However, in terms of what you can afford, there is a lot to be said for generally available, and reasonably priced, components and generally available, and reasonably priced, ways to control them.

Ultimately, the game is reliability. The PLC Manufacturers promise reliability. That is fine, until the particular component fails. Believe it or not, it happens!

Aside from DOA's (and soon to be almost-DOA's, regardless of why), all system-reliability depends on how equipment is used. This boils down to the developer and his skills.

Any idiot can design a piece of **** that is bound to fail. This can happen with PLC Manufacturer sanctioned components as well as generally available PC controlled componets.

The bottom-line is the skill of the developer... not the package delivered.

The DOW lost .52 this month!

The Red Sox, the Packers, the DOW... all of the superstitions are on our side, Tim!

 

[JesperMP]

Terry, I read your post as it is saying that the hardware platform, PLC or PC, is irrelevant to a systems reliability.

Ultimately, the game is reliability. The PLC Manufacturers promise reliability. That is fine, until the particular component fails. Believe it or not, it happens

Even if it happens, isnt there a difference in the frequency with which PLC and PC hardware fails ? And isnt that significant ?

.. all system-reliability depends on how equipment is used. .. The bottom-line is the skill of the developer... not the package delivered.

How can a developer secure the same hardware reliability of PLC and PC based systems ?
If you know some tricks, I'd love to know them.

By the way, from the other posters it looks like there is common design strategy for (some) CNCs to let the CNC control be by a PLC, and the conversion of CAD drawings to machining instructions be done by a PC program. The converted instructions are then downloaded to the PLC for the actual CNC machining to happen.

 

[msinclair]

I think the biggest hardware culprit for (lack of) PC reliability is the hard drive. Probably the power supply after that. Compared with the "no-moving-parts" (and industrially hardened) PLC I would suspect that the mean time to failure for a PLC is between 4 and 10 times that of a PC.

That differential gets greater when you look at how the PC gets used. If you are absolutely disciplined about it and ONLY install what you NEED on a PC, things go better. But people will be people, and you will one day find someone playing DOOM on the "Control PC" just because the PC can do it. Even if nothing "foolish" is done, one of the big selling points is how easy it is to put your HMI in the same package . . . and network things together . . . and extract data and send it to the company ERP system . . . and . . . Well, the more things you try to leverage off of the PC platform, the more failures you get.

The PLC being closed/proprietary/etc is a limitation in many respects, but is a benefit in that the well intentioned can't mistakenly add liability (and the malicious have a harder time of it too!).

Marc


PS: As Terry said though - you can certainly program a PLC to fail. Both PC and PLC systems are still hugely at the mercy of the developer.

 

[panic mode]

harddrive is mechanical device and as such prone to failure.
bigger problem with pc is software. what people understand as PC nowdays is not a commputer running DOS and very small dedicated application. everyone wants some sort of windows on it - and that's the problem. there are simply too many things running on the PC nowdays to make it reliable.
software problem is still present. sure you can replace one
harddrive with another (or graphic card, or anything else...).
and it might be smart move to replace whole pc since they are
getting cheaper and more integrated anyways.
but what do you do with software that runs on it? How many of you have tried to install something as generic and popular as Win9x on newer PC which happened to be "designed for XP"?
Sure, you just need the name, phone number and address of that same smart *** who made sure his product was working fine - on Win3.1 or Win95. Yes you are making something work on XP Pro right now and you are so proud. Just wait till XP 2006 comes out with DirectX 17 and servicepack 13.
How long till failed system is up and running again the way it's supposed?
How long till you find someone answer simple question on something as widespread as Windows or Outlook? How many guys here had problem with something as simple and silly as computer mouse problem with software packages we use daily? Would there be enough of guys to answer not so silly question for something not as popular, something that was working quietly for few years and now it's broken?
when PLC breaks, every electrician knows how to replace broken part or download program.
How many still used PCs older than 15 years you know where to find right now? I can think of two or three. But I sure know at least 50 machines by name, location and function that have PLCs older than that. It looks like PLCs survive much longer before they are ready for garbage.

 

[kennyb]

How many guys here had problem with something as simple and silly as computer mouse problem with software packages we use daily?

yep. who has tried to hot-swap a mouse (essential for the operator) on a system (e.g. intouch) which cannot be rebooted while production is running - and don't even get me started about replacing the mouse with a touchscreen for the pc.

we have a network of 19 pcs controlling robots and we are starting to have a few weird problems on startup after a full system power down and it definitely seems to be a windows problem - failure to boot all the required applications correctly, profibus driver not always loading etc. only the pc reset button works, a hard or soft restart has no effect. i'd like to be able to use the pc as it has more capability in certain applications but the reliability is still a long way off. maybe when mr gates finally acheives world domination he will oversee standardisation of all hardware manufacture.(on second thoughts take a look at every release of windows for quality,reliability,etc)