CE Pro February 2023 - 22

Networking+Home Control
HDMI CORNER by DPL Labs
Cable Failure Illustrates Shield
Issue with Internal Emissions
HDMI's Rev 2.1a mandates
radiated emissions testing.
It uses a relatively new
test process designed for
multimedia applications
designated as CISPR 32. Due
to Rev 2.1a's huge operating
spectrum and the possibility
of interference with other
devices, this new emission
standard is required in addition
to normal digital video
testing processes before
cables are released.
It is pretty much the science
of wire topology, proper
shielding, short and tight
mechanical connections,
and in some cases, RF filtering
on some ancillary wires
used within the interface.
An engineer practiced in the
art of RF emissions knows
just what to look and test
for... but other emissions
can occur within the cable
that can raise havoc with
the signal. One such case
came into DPL Labs HQ, it
was one of those irritants
that were intermittent and
almost impossible to find.
After almost six months
of planning, this brand developed
a new HDMI cable
designed under 48G standards
and was available in
several lengths: 1, 2, 3, 5 and
7 meters. Almost as soon as
the product was shipped, integrator
complaints started
coming in. It was an odd
one. The video would come
and go with no discrete
pattern and only with the
22 | CE Pro February 2023
Fig 1 shows the 19 channels within an HDMI jacket, and in
Fig 2's case, DDC wires buried inside the Ethernet shield
received interference.
be in a differential (balanced)
configuration using
HotPlug as the inverted
channel. Getting Ethernet
down the pipe required two
channels: one inverting and
one non-inverting, identical
to all the other video
channels in the interface.
These differential pairs are
housed inside a pretty hefty,
grounded shield, but in reality,
the true HDMI established
ground is carried by
the DDC ground and is not
necessarily shielded - this
is most important to note.
The reason the cable was
5- and 7-meter cables. We
went through DPL Labs
preliminary test measurements
for a clue. We found
a small spike popping out
of the signal modulation on
the DDC channel. That was
it, except for that this DDC
lane also had abnormally
high capacitance loading -
which is often found and
can cause video failure due
to HDCP, but not this time.
With DPL's new instrumentation,
every channel
within the interface can
be tested at once and in
real-time to see exactly
what's being produced with
every function fully operational
and loaded. Identifying
this unwanted noise is a
cakewalk by just turning off
each channel until the noise
goes away. To understand
this phenomenon, one must
have an understanding of
how the cable functions.
The cable uses 19 wires
within its jacket shown
in Fig 1: eight for video
transmission, one for power,
one HotPlug, one CEC, two
DDC, and one that was
never used, Pin 14. After
Rev 1.4 Ethernet and ARC
were added to where the
Ethernet channel needed to
defective was that when
the cable was designed
the two DDC wires were
buried inside the Ethernet
shield. Fig 2 exemplifies why
radiated emissions can be
so deadly. Because the DDC
wires were inside the same
braided sheath as the Ethernet,
the emission from that
differential pair were firing
directly into the adjacent
DDC wires, disturbing Status
and Control which includes
HDCP and EDID. The longer
the distance, the bigger the
antenna and the worse the
breakdown. In this case, if
the brand had tested this in
full operational mode before
it was released, it may have
eliminated big losses.
JEFF BOCCACCIO,
president of DPL Labs, can
be reached at info@dpllabs.
com, or visit dpllabs.com.
cepro.com
http://www.dpllabs.com http://www.cepro.com
CE Pro February 2023

Table of Contents for the Digital Edition of CE Pro February 2023
