February 2015
20
T
ools
Ethernet originated in the world of computer networking but is now
well established as the communications interface for countless elec-
tronic devices and systems. For example, this standard is used in the
automotive industry to control robots in production systems. Develop-
ers must ensure the interoperability of Ethernet interfaces. The neces-
sary tests must be determined along with acceptance criteria and how
to quickly detect any possible design flaws. Ethernet compliance tests
allow comprehensive verification of interfaces based on standardized
test sequences, thereby providing support to hardware developers as
they work to debug and release their designs.
Ethernet was developed in the 1970s by Robert Metcalfe for use as a
communications protocol. Beginning in 1980, Ethernet was standard-
ized by the Institute of Electrical and Electronics Engineers (IEEE) 802
working group and then continually developed. 10Base-T, 100Base-TX
and 1000Base-T are the most popular electrical Ethernet standards.
For switches and servers, interfaces with 10GBase-T Ethernet are also
being used to allow higher data throughput. All of these interfaces use
two or four twisted pairs typically with RJ-45 connectors.
The Ethernet standard with the lowest data rate (10Base-T) is based
on a signal with Manchester coding. The other standards considered
here with higher data rates use more complex coding schemes for data
transmission along with up to 16 electrical signal levels. IEEE has spec-
ified compliance tests for the electrical characteristics of Ethernet inter-
faces. The documentation describes comprehensive tests of transmitter
signal quality and some tests of receiver signal quality. The specifica-
tion defines test setups, test sequences and special test modes. The user
is expected to manually activate the test modes when performing the
compliance tests, e.g. by setting the appropriate register entries. Details
can be found in the documentation for the Ethernet chip that is used.
Figure 1 shows an example of test mode 1 for the 100Base-T transmitter
test used to measure the quality of the 100Base-T signal (peak voltage,
maximum droop, differential output template).
Developers need to perform Ethernet compliance tests on components
or devices as part of basic R&D or when debugging their designs during
integration. For analysis and verification applications, complete solu-
tions are available based on digital oscilloscopes. Rohde & Schwarz
offers the software options and test accessories for its R&S RTO oscillo-
scopes that are required for performing Ethernet compliance tests. The
test accessories include a test fixture equipped with all interfaces from
10Base-T to 10GBase-T, making it simple to connect the oscilloscope
probes to the DUT signal lines (figure 2).
Ethernet compliance tests are very demanding on the oscilloscope. It
must provide a sufficiently high dynamic range while maintaining a low
noise level. During transmitter distortion tests, for example, the trans-
mitter signal distortion must not exceed a value of 10mV even with
an unwanted signal of 5.4V (Vpp) and 20.833 MHz. The oscilloscope
should be designed with these requirements in mind, ensuring exact
results in this critical test.
Users should select a test
solution that is certified for
compliance testing by an
internationally
recognized,
independent Ethernet vali-
dation lab. The University of
New Hampshire InterOp-
erability Laboratory (UNH-
IOL) confirmed that “the R&S
RTO oscilloscope and R&S
ScopeSuite correlate with the
time-tested techniques and
methodologies of UNH-IOL.”
When putting new hardware
designs into operation, devel-
opers tend to be under signif-
icant time pressure, making
it important to perform the
Figure 1. Signal acquired by an oscilloscope from an Ethernet chip in
mode for 100Base-TX tests for peak-to-peak jitter measurements
Figure 2. The R&S RT-ZF2 Ethernet test fixture offers all interfaces
from 10Base-T to 10GBase-T for Ethernet compliance tests. It connects
the oscilloscope probes to the various DUT signal lines.
oscilloscopes. The outstanding dynamic range and the low-noise front-ends ensure
accurate re ults even in critical transmitter distortion tests. The co plete solution also
includes the two software ptions R&S RTO-K22 for 10Base-T to 1000Base-T a d R&S
RTO-K23 for 10GB se-T, al ng with the R&S RT-ZF2 test fixture. Testing of the
Ethernet interfaces is extensiv ly automated to deliver precise results th t can be fully
docu ented to meet
use requirement .
10Base-T
100Base-TX
1000Base-T
10GBase-T
Standard
IEEE 802.3
clause 14
IEEE 802.3
clause 25
IEEE 802.3
clause 40
IEEE 802.3an
Coding
Manchester
coding,
unidirectional,
2 twisted pairs
4B5B, MLT-3,
unidirectional,
2 twisted pairs
8B10B, PAM-5,
bidirectional,
4 twisted pairs
128-DSQ, PAM-
16,
bidirectional,
4 twisted pairs
Signal levels
Manchester level
change
3 levels
5 levels
16 levels
Transmission
bandwidth
10 MHz
32.5 MHz
62.5 MHz
500 MHz
Table 1. Protocol characteristics of various Ethernet standards
((Bild für Leadpage, kann laut Dummy entfallen))
Table 1. Protocol characteristics of various Ethernet standards
