PROFIBUS shenanigans

Im working on a large machine, been down for 2 days now. I am getting Profibus errors of every type possible for a particular cell of the machine. All slaves are showing o.k. I can Isolate one saw assembly and the machine will work without errors, so the problem lies in there obviously, but the only thing profibus has to do there is talk to 2 motors. each one reads ok, but as soon as they are used in cycle ALL errors happen. Any ideas? I have checked the cables for fractures, etc. and they check out fine. We are using SIMATIC micro pc.

You are not describing your setup at all, but I am guessing there are VFDs since you mention that the Profibus network has to "talk to 2 motors". And if the VFDs are not running, the bus is OK. When the problems VFDs startup, you get bus errors.
If the above is correct, I would check screening and earthing.

Or do you mean that if 1 of 2 are connected, the system can run, whereas if 2 of 2 are connected it cannot run ?
If that is the problem, then you should check equipotential earth, as well as termination.

Please describe the profibus system in detail (masters, slaves, bus speed, bus profile).
And describe the whole thing a lot more.

JesperMP said:

I would check screening and earthing.

Click to expand...

+1

Once had equipment which was causing a major headache, the installers used the stainless adapters which looked to be causing an issue. Luckily i had a few of the plastic ones at the bottom of my bag and as time was tight we swapped them out and was OK.

Thanks for the quick reply!
I have thought about shielding before when these errors were intermittent, but didnt have the chance to fully investigate during production. I dont think the VFDs are causing the issue unless there is a noise problem like you suggested. I have gone to the slave devices local to the cell and checked for proper grounding and made sure the termination points have the shielding going to ground.

setup in brief...

4 cells to the machine. only one at fault. the faulty cell has 4 saws that have a Y axis motor and a Z axis motor. each motor is controlled with a VFD. during cycle, Y axis fires to drive the saw/die assembly to the part. then Z axis drops the saw and cuts part. return home and xfer part to next cell. It is hard to pinpoint the exact moment of fault due to guards in place. Either way, I would think it would be either of these motors Profibus cables being hit with EMI. seems legit?

DEUTOPLASM said:

I have gone to the slave devices local to the cell and checked for proper grounding and made sure the termination points have the shielding going to ground.

Click to expand...

By this you mean that the cable shield is tied to a big fat earth bus-bar, as described in the profibus application manual ?
Just terminating the shield in the connector is not enough.

The ideal way to troubleshoot this would be to connect a profibus tester, it would give you a very clear picture of what is causing the problem.

do the profibus cables connect to any encoder or field device, is there any movement of the profibus cables along the motor / encoder?

I don't know if you already did that, but you have to be sure that your power cables to, and from your VFD's are properly grounded.
Bad grounding from motor cables can cause this kind of problems.
Once I had a cabinet with corrosion on the connection from the copper earth busbar to the building ground, causing me all kinds of communication errors.

Got it! Thanks everyone. Turns out someone had taken apart one of the profibus connectors before. We are using the steel ones. Inside the connector housing is a small bar that clamps around the shielding. It was missing, which gave a weak path to ground. I fixed that and boom. Were back in bidness!

DEUTOPLASM said:

Got it! Thanks everyone. Turns out someone had taken apart one of the profibus connectors before. We are using the steel ones. Inside the connector housing is a small bar that clamps around the shielding. It was missing, which gave a weak path to ground. I fixed that and boom. Were back in bidness!

Click to expand...

Even if you have fixed the problem for now, the cause and remedy indirectly points to that you have not done this:

jesperMP said:

By this you mean that the cable shield is tied to a big fat earth bus-bar, as described in the profibus application manual ?
Just terminating the shield in the connector is not enough.

Click to expand...

Incorrect. The shielding for the profibus cable wasnt secured properly to the metal connector housing. The profibus connector plugs into a module that happens to be tied to the bigass earth ground bus bar.

No no no. It is not good enough.
It does not matter if the module that the connector is plugged into is tied to a bigass bus-bar.

If you have a simple profibus network with just a few stations within the same control cabinet, then OK just rely on the earth connection via the connectors.
But if you have many stations, or profibus connections via cabinets, or profibus cable in the field, or if there are VFDs or other "nasties", then you absolutely MUST tie the Profibus screen to the earth bus-bar via a big fat clamp that goes directly over the exposed screen on the cable.

Oh yeah, I know what youre saying. That may very well help us in the long run, but for now I just wanted to get the thing going. Given the severity of this downtime im sure I can have some extra lugs ordered and fitted. Thanks again for the advice