Failure of PCBs due to leakage currents

rejoe.koshy

Member
Join Date
Dec 2011
Location
kolkata
Posts
193
Hello guys,
I have an injection molding machine , which uses a PLC to control its process.After running the machine for few days, one fine morning I suddenly observed that the HMI panel had smoked up.

The HMI unit was dismantled & it was observed that the RS232 communication cable has been completely burnt & HMI was in a inoperable state.It was sent for repair but the OEM has declared the HMI board completely dead, with most its components damaged.

They are telling me that the some leakage currents have found their way in to the PCB & destroyed it, but I am not able to stomach the fact.They are suggesting the use of isolation transformer to prevent such an occurrence, but I have never come across a communication cable getting damaged to this extent due to leakage currents!! & why would it happen suddenly..

Pls help!
 
Hello guys,
I have an injection molding machine , which uses a PLC to control its process.After running the machine for few days, one fine morning I suddenly observed that the HMI panel had smoked up.

The HMI unit was dismantled & it was observed that the RS232 communication cable has been completely burnt & HMI was in a inoperable state.It was sent for repair but the OEM has declared the HMI board completely dead, with most its components damaged.

They are telling me that the some leakage currents have found their way in to the PCB & destroyed it, but I am not able to stomach the fact.They are suggesting the use of isolation transformer to prevent such an occurrence, but I have never come across a communication cable getting damaged to this extent due to leakage currents!! & why would it happen suddenly..

Pls help!

I doubt "leakage" currents have caused this damage.

I would suggest somewhere in the cabling the RS-232 cable has become damaged, and touched a "hot" live component....
 
I doubt "leakage" currents have caused this damage.

I would suggest somewhere in the cabling the RS-232 cable has become damaged, and touched a "hot" live component....

Exactly, but then I checked the vicinity of HMI unit for 230 V connections , but couldnt find any.
 
Run a wire between the 2 chassis grounds, see if there is current passing through it. Different ground potential between the 2 serial ports will do exactly that.

You can get optical Serial Isolators that can protect against this condition .
 
But what is more intriguing to see , was the fact that the SMPS , which was feeding power (24 v dc) to the HMI unit was intact.Any leakage,spikes would have definitely destroyed the power source has well.

Moreover , I looked at the cable closely enough & found that only a particular portion of the cable had burnt out severely.If some high current would have found its way through the cable, it would burnt the cable across its entire length.
 
If you found a burned spot in the RS-232 cable, then you can be pretty sure the relatively delicate transceivers at at least one end of it were fried.
 
But what is more intriguing to see , was the fact that the SMPS , which was feeding power (24 v dc) to the HMI unit was intact.Any leakage,spikes would have definitely destroyed the power source has well.
You might have thought so, but no. I had a Beijer HMI that had obvious overvoltage burn damage to components, fortunately for me it was the only thing in the box damaged. I think the spike finds the weakest component, that then acts like a fuse, or maybe strange things can happen at high frequencies. All I know is that ended up in the middle, Beijer said it wasn't their fault, my customer said they hadn't done any welding since fitting the panel, and I ended up paying the bill. The system had diodes to protect against reverse spikes and voltages, and since then I fit MOVs to hopefully kill any spikes.
 
Moreover , I looked at the cable closely enough & found that only a particular portion of the cable had burnt out severely.If some high current would have found its way through the cable, it would burnt the cable across its entire length.

Wouldn't this depend on the resistance of the end connections? Did both ends get damaged or just one? The end that didn't seem to have a fried cable, if still working, how is the RS-232 comm channel? Isolated?

This being said, shouldn't the start of the burnt area indicate where the event must have started?
 

Similar Topics

I have a CNC cutting equipment, all the components are Fanuc from the CNC CPU A02B-0307-B822, a PLC AIF01A and a panel i, on the panel i (operator...
Replies
1
Views
45
Wizards, It has been a few, but you all have always done me well. I have acquired a 1769-L33ER and want to use it as my collection PLC to...
Replies
5
Views
463
Spoke to a few outlets including Emerson and some distributors. Apparently updating firmware on subsequent modules past the CPU still uses...
Replies
1
Views
349
Hello Everyone, i Have im my Industry a Endress & Hauser Promag400 this has a screen that constantly have that error, it says to wait, somebody...
Replies
2
Views
425
Hi all, One of the Kinetix 5500 in a machine is getting a Feedback Device Failure, subcode 66: INTERNAL RESOURCE ACCESS ERROR. Studio 5000 V.34...
Replies
4
Views
804
Back
Top Bottom