JULY 2018 I DESIGN007 MAGAZINE 73
els; any investigation of the losses would be
completely useless without such models. Thus,
any stackup exploration or planning stage must
include building small validation boards or test
coupons to verify the data obtained from the
PCB manufacturer and to identify actual geom-
etry adjustments and the conductor roughness
models. The coupons should have two seg-
ments of transmission line (single-ended or
differential) with different lengths per each
layer with unique dielectric. The coupons must
be cross-sectioned after S-parameters of the
line segments are measured. This is the most
important step of the systematic approach to
design predictable interconnects.
DESIGN007
Simbeor software was used for all computa-
tions provided in this article. All corresponding
Simbeor solutions are available upon request to
learn the "sink or swim" process.
References
1. M. Marin, Y. Shlepnev, 40 GHz PCB Inter-
connect Validation: Expectation vs. Reality, an
award-winning paper from DesignCon2018,
January 31, 2018, Santa Clara, California. Pre-
sentation with complete report is also avail-
able here.
2. Y. Shlepnev, Sink or Swim at 28 Gbps, The
PCB Design Magazine, October 2014, p. 12-23.
3. Y. Shlepnev, Unified approach to intercon-
nect conductor surface roughness modelling,
2017 IEEE 26st Conference on Electrical Per-
formance of Electronic Packaging and Systems
(EPEPS2017), October 15-18, 2017, San Jose,
California.
4. Y. Shlepnev, Broadband material model
identification with GMS-parameters, 2015 IEEE
24st Conference on Electrical Performance of
Electronic Packaging and Systems, October
25-28, 2015, San Jose, California.
Yuriy Shlepnev is founder and
president of Simberian Inc.
The first week of July 2018 marked the last in a series of
ground tests demonstrating the capabilities of the Tendon
Actuated Lightweight In-Space MANipulator (TALISMAN)
robotic arm; the Strut Assembly, Manufacturing, Utility &
Robotic Aid (SAMURAI); and the NASA INtelligent Jigging
and Assembly Robot (NINJAR) components of the Com
-
mercial Infrastructure for Robotic Assembly and Services
(CIRAS) project.
Earlier this year, the team of engineers manipulated the
newer, longer arm back and forth from folded to extended
positions to demonstrate that it is fully operational, then
they showed it could pull a truss out from being stowed in
a compartment. In this demonstration, the TALISMAN arm
was used to move a solar array from one truss section to
another and to install the array.
SAMURAI, the robotic hand that passes truss parts,
and NINJAR, the jigging robot that holds the pieces
in place while they are fastened, have similarly
been put to the test this year. The team first used
a remote control to operate the two robots and
assemble truss segments. This most recent test
accomplished an autonomous truss build, using
only code and no remote control.
CIRAS is a part of the In-Space Robotic Manufac-
turing and Assembly (IRMA) project portfolio, man-
aged by NASA's Technology Demonstration Missions
Program
and sponsored by NASA's Space Technol-
ogy Mission Directorate. Click here for more.
NASA, Partners, Advance In-Space Assembly Robotics
