On 03/16/2015 09:40 PM, Karlsson & Wang wrote:
> A shield is primarily intended to prevent electrostatic coupling from
> the outside world. So by grounding in the consuming end the shield
> will get the ground potential of the consumer and the signal cables
> will be shielded from different external electric fields. This should
> motivate why as you say the shield should be connected in this end
> only. If there are current there is also a potential difference.

There are generally three cases to consider:
1) a shield as a Faraday cage; this means that only one side of the
shield may be connected to go into the earth/ground star-point where you
maintain an absolute reference of zero (0V). The shield is used to dump
all (most) cable-internal EM radiation into a low-impedance star-point
via the shield.

2) a shield as in a coaxial conductor; here the signal is contained in
the cable and the shield is part of the conducting circuit. This works
on basis of very tightly controlled cable properties and is generally
limited to a specific frequency range per cable specs.

3) all other cases; the shield is connected at both sides and is
intentionally or unintentionally part of the conducting circuit. This
case will generally give you worse results in terms of EMI emissions and
protection. The shield will radiate and may act as an antenna to worsen
the situation.


> I consider the VFD to be a noise source since it have common mode
> voltage which will emit an electrical field. There is also a
> capacitance between the VFD cables and shield. Since Shield impedance
> on high frequency is far from zero the shield around the VFD cables
> will not be at GND potential. The most common method is to increase
> common mode inductance by a filter but I have also seen active
> filters which reduce the common mode voltage and multiple step
> voltage inverters.

With respect to VFD systems; the amount of junk they produce depends on
the quality of the converter.

The best version generates a relatively pure sinusoidal output and the
EMI it generates is very minimal. Such VFD can normally be connected
without problem with unshielded cabling.

The lesser quality emulates a sinusoidal output, but has substantial
higher harmonics. These VFDs are not too shabby, but they can cause
interference. The best solution is to filter the output before putting
it on (long) cables. No shielding is required when the harmonics are
under control, but it generally does not hurt to use a Faraday cage type
shield.

The cheap VFDs are poor substitutes and generate nearing square-wave
output. The amount of EMI from higher harmonics is high and is often
difficult to filter at the source. These VFDs should be used with both
output filters and shielded cabling.

The remaining problem that may arise is EMI from the motor. The
remaining harmonics may radiate from the motor just as easily. That
cannot be solved with shielding of the cables. You must ensure proper
earthing of the motor as well and it should be enclosed in a proper
metal casing.

It should also be noted that VFD frequency changes cause harmonics in
the output. If you turn on/off the hard way, then you can introduce some
transients that are very hard to control. The best way is to control the
up-/down-going frequency such that no abrupt changes can occur and
therefore no transients are allowed to be created due to too fast changes.

-- 
Greetings Bertho

(disclaimers are disclaimed)

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to