|
Home > Archive > Electrical code Compliance > June 2005 > Conducted emission test to DC lines
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Conducted emission test to DC lines
|
|
|
| Does anybody know how to perform the conducted emissions test as per EN
55022 to DC supply lines? Can we use the same LISN we normally use to test
the mains supply lines (AC)? If so, how do we have to connect the DC lines
(0V, 24V)? What about the protective earth?
Thanks in advance.
Ana
| |
| Ed Price 2005-06-17, 11:31 pm |
|
"Ana" <lacecal@eis.uva.es> wrote in message
news:d76l2d$32p$1@cabezon.cpd.uva.es...
quote:
> Does anybody know how to perform the conducted emissions test as per EN
> 55022 to DC supply lines? Can we use the same LISN we normally use to test
> the mains supply lines (AC)? If so, how do we have to connect the DC lines
> (0V, 24V)? What about the protective earth?
>
> Thanks in advance.
>
> Ana
>
>
Yes, you can measure the noise on a DC power bus; this is regularly done in
military testing for 28 VDC power busses. The LISN inductor design is very
resistant to saturation, so the impedance is the same regardless of current
draw.
OTOH, why would you need to meet EN 55022 for a DC powerline? Are there
public mains somewhere that are DC?
--
Ed
WB6WSN
El Cajon, CA USA
| |
|
| Dear Ed,
Thanks for your promtp reply.
The thing is that the EUT we have to test have DC supply lines (0,24V) but
the manufacturer do not deliver the power source with it, so that any 24V DC
source can be used. Due to the equipment characteristics the product
standard EN 55022 applies though there´s nothing in this standard regarding
to DC supply lines. As you say in your post the LISN can be used to perform
this test but I´m not really sure about the earth connection since,
according to EN 55022 the LISN measures conducted disturbances "between the
phase lead and the reference ground and between the neutral lead and the
reference ground".
Thus, in a "floating" DC supply mains (0, 24V), how can the test be
performed? Do we have to connect the 0V line and the LISN earth terminal to
the reference ground plane? If so, we would only measure the conducted
emissions between the 24V line and the reference ground; is that correct?
Thanks again.
Ana
| |
| Ed Price 2005-06-17, 11:31 pm |
|
"Ana" <lacecal@eis.uva.es> wrote in message
news:d77aji$3k7$1@cabezon.cpd.uva.es...
quote:
> Dear Ed,
>
> Thanks for your promtp reply.
>
> The thing is that the EUT we have to test have DC supply lines (0,24V) but
> the manufacturer do not deliver the power source with it, so that any 24V
> DC source can be used. Due to the equipment characteristics the product
> standard EN 55022 applies though there´s nothing in this standard
> regarding to DC supply lines. As you say in your post the LISN can be used
> to perform this test but I´m not really sure about the earth connection
> since, according to EN 55022 the LISN measures conducted disturbances
> "between the phase lead and the reference ground and between the neutral
> lead and the reference ground".
>
> Thus, in a "floating" DC supply mains (0, 24V), how can the test be
> performed? Do we have to connect the 0V line and the LISN earth terminal
> to the reference ground plane? If so, we would only measure the conducted
> emissions between the 24V line and the reference ground; is that correct?
>
> Thanks again.
>
> Ana
>
>
LISN's have been around for over 50 years, although they have been in and
out and then back in favor for various standards.
An LISN does three things:
1. Provides a modest amount of filtering of the incoming AC or DC power.
That's nice, but you usually still need a big powerline filter or a
dedicated & quiet power source.
2. Standardizes the power source impedance that the device under test
"looks" into. Although EN 55022 (and other standards) specify the CISPR
50-ohm impedance curve, LISN design and the resultant impedance curve can be
anything the standard thinks is right to reasonably replicate the user
environment. (For example, one spec on satellite power busses defines a 5 uH
inductor LISN, while TEMPEST testing often uses much larger inductances.)
3. Provides for a convenient signal port that is frequency-selectively
coupled to the powerline. (You capacitively couple to the powerline; this
mostly rejects the low-frequency power, but passes the high-frequency noise.
And don't forget to verify the coupling factor during periodic calibration.)
The concept of conducted emission testing is to measure the noise that the
device under test is contributing to a common (shared by other equipment or
users) power bus. If your device connects directly to the common power bus,
then the test is reasonable. However, if you are the only user of the power
source, then perhaps conducted emission information has no value.
If you have to put anything between your device and the common power source,
then that additional device becomes an unknown variable. Does it filter and
reduce your device's noise? Maybe. Does it contribute noise of its own onto
the power bus? Maybe that too. So, if your device needs 24 VDC, and you want
to power it from a typical AC public mains, you obviously need an external
power converter (like a "wall-wart"). And the boundary of your product must
then include this converter; so you need to measure conducted emissions at
the wall-wart to mains interface. If you provide the wall-wart, then this is
obviously part of your equipment. If you allow the customer to provide the
converter, then you are in a grayer area, and I think you would have to test
your product using a typical commercially available third-party wall-wart
(accepting whatever problems may come along with this uncontrolled [by you,
that is] item).
Getting back to your questions on connections, remember that all standards
which include an LISN also define a ground reference plane. LISN networks
are inserted into all powerlines. On a single-phase line in the USA, that
would be the high and the neutral lines (you need 2 LISN's). Same for
Europe, except that both lines are "highs". And, IIRC, in Australia, one
side of the 230 VAC pair is grounded. In all cases, the supply and return
power paths have LISN's in them. For USA three-phase, you need four LISN's
(one in each phase and one in the neutral). Safety ground wires are not
measured.
If your device has no ground connection, then common mode noise will flow
down both power wires, through the LISN's, and into the ground plane (and
this will mean you will have problems with radiated emissions). Differential
mode currents will flow down one power lead, through its LISN to the ground
plane, back through the other LISN to the other power lead, and back up to
your device. The voltage that you see at an LISN signal port is due to a
combination of common and differential currents.
Military vehicle systems using 24 VDC or 28 VDC have a supply and a return
line (2 LISN's), although the power bus negative is usually referenced
(grounded) to the vehicle chassis near the battery or alternator. In this
case, you can place a wire between the return-line LISN (power-source side)
and the ground plane to emulate the grounded vehicle power system.
One futher caution; remember to always use an attenuator on the LISN signal
port. It may not be nice to "throw away" 10 dB of sensitivity, but LISN's
can develop a significant transient voltage due to unexpected power
interruptions. The attenuator protects the expensive mixer diode in the
front end of your analyzer.
--
Ed
WB6WSN
El Cajon, CA USA
| |
| Michael 2005-06-17, 11:31 pm |
| "Ana" <lacecal@eis.uva.es> wrote in
news:d77aji$3k7$1@cabezon.cpd.uva.es:
quote:
> Thus, in a "floating" DC supply mains (0, 24V), how can the test be
> performed? Do we have to connect the 0V line and the LISN earth
> terminal to the reference ground plane? If so, we would only measure
> the conducted emissions between the 24V line and the reference ground;
> is that correct?
Is your EUT a relatively small device powered by a "wall wart" as Ed
already asked or is it equipment powered from a 24V centralized supply like
used in some wireless telecom installations?
If the latter is the case, you will have to test it according to EN 300386
which refers to EN 55022 (as you mentioned) and uses the 50µH/50Ohms LISN
as specified in CISPR 16-1.
The grounding might not be specified by the equipment manufacturer, but by
the guidelines for the installation or a relevant standard. I am not that
familiar with 24V installations in this area, but 48V installations for
example have positive ground on the DC supply voltage.
HTH
Michael
| |
|
| Dear Ed & Michael,
Thanks a lot for your support; it´s great to have the chance to learn from
such an experts as you.
Hope I´ll finally manage with the test, though I still have some doubts
about it.
I´ll tell you about it.
Best regards,
Ana
| |
| Michael 2005-06-17, 11:31 pm |
| "Ana" <lacecal@eis.uva.es> wrote in
news:d7mg5t$na0$1@cabezon.cpd.uva.es:
quote:
> Dear Ed & Michael,
>
> Thanks a lot for your support; it´s great to have the chance to learn
> from such an experts as you.
>
> Hope I´ll finally manage with the test, though I still have some
> doubts about it.
>
> I´ll tell you about it.
Ed provided a good explanation - I have to correct myself: EN 300386 is
_only_ valid if your equipment is used in telecom applications.
Regards,
Michael
|
|
|
|
|