[SOLVED] what are the risks for isolating test equipment by negating earth ground

[Zak28]

Are there any risks from isolating test equipment by removing the ground prong from the power cable?

 

[FvM]

Test equipment, e.g. an oscilloscope, is usually designed to be operated with PE connection. Leakage currents and insulation parameters are not suitable for operation without it. Operation manuals often hint to this fact explicitly.

The only way to operate if safely without earth connection is an external safety transformer, provided that you don't connect a hazardous contact voltage by the signal cables.

 

[dick_freebird]

Two kinds of risks. One is electrical safety, without the
ground prong any piece of equipment could have a case
voltage anywhere between positive and negative line
crest voltage, DC or AC. Potentially (heh) unsafe.

Then, the variation in case voltages may cause new
ground-loop currents and these can get into measurement
results, without you knowing. Establishing ground in a
way you understand and control, is better. Safety,
power and signal grounds have some relation; what
that is, or should be, is best figured out explicitly.

 

[Zak28]

Do isolation transformers provide earth ground or just ground to its chassis?

 

[Easy peasy]

If the 'scope ( for example) has RFI caps from P & N to earth, then then metal chassis will sit at half the mains volts - if you have an all plastic unit then less worries - but the gnd part of the BNC connectors for the scope probe will still be at 1/2 Vmains and could damage some cktry when connected via the gnd alligator clip.

If you isolate with an isolating xfmr - you can put a 500V 2 - 5W 1Meg resistor from the chassis to earth simply to keep the metal on the 'scope at a known potential.

However if you clip the scope gnd to a hi V potential or a switching node ( 0-420V of an half bridge for e.g.) then you will radiate from this point as well as a safety issue if you touch the scope ...

 

[FvM]

Do isolation transformers provide earth ground or just ground to its chassis?

The idea of an isolation or safety transformer is to provide a supply voltage without earth connection. Respectively they are exposing the output voltage at a two pin socket with no earth contact. The transformer chassis may be either isolated or connected to earth ground, but that's irrelevant for the output.

 

[KlausST]

Hi,

If the 'scope ( for example) has RFI caps from P & N to earth, then then metal chassis will sit at half the mains volts

This usually is done with devices that don't have an Earth connection, with safety isolation and plastic case.
(Wall wart power supplies)

In EU I think it is not allowed to design a mains powered device with metal case but without Earth connection to this metal case.

The benefit of using an isolating transformer in combination with a scope is, that the scoppe_GND is not connected to Earth_GND anymore. This avoids GND loops via Earth_GND. But you loos the safety. It should be done only by persons that are aware of the additional risk....and fir sure on their own risk.

The safety problem is not the mains voltage...the safety problem is when one connects the scope_jig_GND to a dangerous voltage referenced to earth_GND.
* Without an isolating transformer the scope case voltage is limited, current flows agains earth_GND and the RCD or the fuse trips.
* With an isolating transformer .. not only the cable and the connectors (BNC) carry dangerous voltage, but also the metal case. And this is not visible...

If you want to break Earth_loops, its more safe to use the isolating transformer at the DUT.

Klaus

 

[Nik_2213]

Plan B is to have a commercial wall-wart and run your test equipment on low voltage.

True, a lot of line-powered equipment is 'double insulated', but you are connecting wires, running signals in and out. Who knows what voltages you'll get, eg when the load on a 'harmless' up-converter goes open circuit ? Will your equipment's input protection circuitry endure ?

Take a look at MIDI wiring. That's digital, but still only grounds at one end of signal cables to prevent ground-loops' hum or, worse, phasing errors. And, yes, MIDI opto-isolates *everything* in-bound. Safety-overkill, totally, but it works.

 

[treez]

Are there any risks from isolating test equipment by removing the ground prong from the power cable?

Yes you can severly overstress the insulation in the scope.

Cutting earth out of scope plug is a no-no.

It will also mean more noise problems most probably.
Earth performs both safety and noise control functions....never one or the other......always both.

 

[Zak28]

Would grounding an isolation transformer chassis to earth running wiring separate from residential ground resolve safety/noise issues?

 

[dick_freebird]

An isolation transformer will free you from ohmic connection
to the facility ground net. This is good as far as it goes
(who knows what other people are pushing onto the wire?).
But now the signal and safety grounding is on you, to put
proper.

Transformer-floated bench equipment can show high AC
amplitude on high impedance measurements referred to
some other ground. This can also be ESD-like to your DUT
(have seen CMOS RF parts with ESD damage as low as
10V, gate ox rupture takes very little current).

Everybody should be grounded and nobody should be
sending current through the measurement reference
ground.