Author Topic: Troubleshooting Puzzling Earth Issue  (Read 1260 times)

0 Members and 1 Guest are viewing this topic.

Offline ed_reardonTopic starter

  • Regular Contributor
  • *
  • Posts: 131
  • Country: gb
Troubleshooting Puzzling Earth Issue
« on: July 06, 2017, 07:23:16 am »
Hello all,
           I am currently dealing with a device which seems to have a bizarre earth issue.

The device is a piece of industrial equipment,  it is being powered by a DIN rail Class-I AC/DC SMPS (TDK Lambda series) and operates on 24VDC from a 230VAC supply.   The internal metal work of the equipment (which is bonded to the earth pin of the incomming power to the power supply) is sitting around 55VAC above ground (at 50Hz precisely so it's mains referred) when measured with a meter.

This is causing a 'nibble' to operators whenever the case is opened.

Probing the +/- output of the PSU shows there is negligible AC component, (around 70mVAC) as well as the + or - pins to ground, which again show negligible AC components.  The DC cable run from the PSU to the equipment is short and I can't see how it could inductively pick up anything along the way.  Essentially I've got a very noticeable 55VAC 'appearing' from nowhere inside the unit,  there is nothing inside the equipment which could potentially generate this stray voltage (it's an instrument amplifier and a few solenoid valves).

Any tips for dealing with bizarre earth faults? I've not had a good chance to look at it yet,  but it's a bit of  headscratcher.  If the DC lines contained a large AC component I'd understand, but this voltage appears from 'nowhere'.

Being an instrument amplifier, having a stray 55VAC in the unit isn't helping with accuracy either!  |O


Any suggestions,  could this be PSU related or is it being coupled in from somewhere else?



 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Troubleshooting Puzzling Earth Issue
« Reply #1 on: July 06, 2017, 07:52:06 am »
Is the incoming earth (which is bonded to the chassis) shorted to the 24V DC input "- "?  If it isn't, connecting them may change the situation - beware grounding (earth ground) in industrial systems is a beast that often isn't properly planned at system level, so you might introduce unwanted earth currents.

Does the instrument amplifier isolate the 24V input from its circuit ground? Doing so isn't unusual in idustrial equipment.
If so, the source of the AC might be somewhere in the cabling / sensors connected to the amplifier.

Otherwise: usually such coupled AC is rather high impedance, so measuring the 24V DC input to earth could give you other results than measuring chassis to earth if you use different instruments / ranges.
Safety devices hinder evolution
 
The following users thanked this post: ed_reardon

Offline DenzilPenberthy

  • Frequent Contributor
  • **
  • Posts: 408
  • Country: gb
Re: Troubleshooting Puzzling Earth Issue
« Reply #2 on: July 06, 2017, 09:30:09 am »
I'd check the earth continuity of your supply.

Your power supply will typically have some emc filter on its mains input like this:



Note the X and Y capacitors. In particular the Y capacitors between the mains and earth.  If the earth connection is left floating (or doesn't exist like in a lot of class 2 stuff like DVD players etc) then the 'earth' will actually float to around half of the mains voltage.  Usually the capacitors are small enough that they can't pass enough current at 50 or 60 Hz to do any damage.  If your system is designed to be properly earthed but there is poor earth continuity from the supply then this would cause the problem you are describing...
 

Offline DenzilPenberthy

  • Frequent Contributor
  • **
  • Posts: 408
  • Country: gb
Re: Troubleshooting Puzzling Earth Issue
« Reply #3 on: July 06, 2017, 09:38:31 am »
also... I see you are in the UK. If once you've sorted this you find that the current in the earth conductor is more then 3.5mA then you should check section 543.7 of the BS7671 'wiring regs' as you may need to take precautions as this counts as "Equipment having high protective conductor currents"

 

Offline vealmike

  • Regular Contributor
  • *
  • Posts: 192
  • Country: gb
Re: Troubleshooting Puzzling Earth Issue
« Reply #4 on: July 06, 2017, 02:13:25 pm »
I'll echo what Denzil has said.

You say that you measure 55V on the chassis, which is bonded to earth. Where are you taking your reference? i.e. Where is the other end of the meter?

If operators are getting a nibble, then the chassis is not at earth or the operator is not at earth.

Personally, since you're in the UK, I'd start by giving this a class 1 (IIRC) PAT test, to measure impedance from chassis to earth. Use the mains lead that's giving trouble.


It's just a resistance check, but it's done with a much higher current than your multimeter will use in order to measure much smaller resistances.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf