OT: On-site versus remote programming

[rguimond]

It is just me, or have other noticed that theyre more comfortable programming devices that they can see, as opposed to devices they can't see? The reason i ask is that I have a Redlion PTV sitting on my desk, clipped onto a DIN rail. A 24VDC power supply is also attached to it, connected to a surplus computer power supply cord with the receptacle end cut off. The PTV is connected to a network switch and I'm programming it over ethernet.

It looks bad and takes up room in my office. There's no good reason for having it here because there are no physical inputs or outputs that I can monitor (I'm just using it for OEE data collection). It could just as easily be mounted in a control poanel anywhere in the plant where there is a network connection. Subconsciously, however, there's a degree of comfort when I can see it.

I'm weird. That's all I have to say

 

[CapinWinky]

I have my test equipment in one of those overhead cubicle cabinet things across the room. Out of sight but still accessible. It's not a comfort thing, it's a Murphy's law thing; as soon as you make it too hard to get to, you'll need to get to it.

EDIT:
I prefer simulating PLCs when I can, but some platforms don't make it easy, so I use real controllers and comms are always easier to test in real life.

 

[James Mcquade]

Many oem programmers write code for machines that were in the process of being built. they have a sequence of operations and an I/o listing.

I have done this for 20 years until recently. I started using the plc 5 & 500 emulators to test my code about 5 years ago.

works great.

regards,
james

 

[boneless]

Testing purposes, I simulate most of my stuff. Problem is that the HMI we use is Beijer, PLC is Siemens. So when I get around to the final testing, I download it to a PLC.

Also, some of the safety stuff is just a hassle to simulate. Same goes for BUS.

I built in my test setup in the back of a cabinet here, where I can still access it and swap some wiring. It isn't the neatest looking installation, but I can close the door and I still have most of my storage

(and it keeps my wife happy that she doesn't trip over it all the time )

 

[mk42]

It's not a comfort thing, it's a Murphy's law thing; as soon as you make it too hard to get to, you'll need to get to it.

Speaking of Murphy's law:

I was involved in a project where the end customer had one giant network connecting all its plants, and each line/cell had identical IP addresses. They were isolated by NAT routers.

One day, someone went to download a simple change from his desk. He managed, instead, somehow, to download to a PLC in a plant in a different country, which obviously caused havoc.

Lesson learned: I prefer to only download to a PLC I can see, if for no other reason than being able to verify it is the right one.

 

[The Plc Kid]

Speaking of Murphy's law:

I was involved in a project where the end customer had one giant network connecting all its plants, and each line/cell had identical IP addresses. They were isolated by NAT routers.

One day, someone went to download a simple change from his desk. He managed, instead, somehow, to download to a PLC in a plant in a different country, which obviously caused havoc.

Lesson learned: I prefer to only download to a PLC I can see, if for no other reason than being able to verify it is the right one.

Sounds like they had a half As$ed network setup. If it was setup correctly you would never be able to do that.

 

[jgreenewv]

If I'm doing initial config or testing, or some monitoring, I'm fine being out of sight of the PLC. Otherwise, I want to be sitting or standing beside it, just in case.

As an aside, I interviewed with a machine builder a couple years ago. The position I was interviewing for was open because the previous engineer was sitting at his desk modifying code on the machine on the shop floor. Except he was actually modifying code on the identical machine in production at the customer's plant because he forgot the change the IP address and hadn't disconnected the VPN connection. Apparently repeatedly downloading to a production machine is not appreciated by the end user.

 

[mendonsy]

I can think of a few rare instances where I might consider loading programs remotely, but not many!! I want the ESTOP button within reach when I test something!!

 

[Tom Jenkins]

I've done remote programming and testing, including tuning machine controls and processes in full operation across the ocean. I am never totally comfortable with it, but has always been successful and is obviously cost and time effective. A few precautions that we always took:

1) the connection at the remote site must include a router, modem, or other device that is normally unplugged
2) before we tied in we always called the plant so they could alert their staff, plug us in, and then give us the OK
3) after we were done we would call the plant and ask them to unplug us. We also had standing instructions to unplug us after a set period of time, whether we had called them or not
4) we didn't do a remote link until a conventional on-site start-up had been completed, and we tried to have the remote work done by a person who had been to that site
5) we put warning labels on the equipment indicating it could be operated remotely and caution was to be observed - and stating that LOTO procedures were needed when servicing the machine