- All
- Measurement and validation
- Modeling and high-speed design
- Signal integrity fundamentals
- Troubleshooting - test fixtures - and compliance
- WRT engineering services
Start by contacting Wild River Technology to discuss the design challenge, target data rate or frequency range, current test results, and project schedule. For a focused engagement, it is helpful to provide available PCB stack-up information, layout or interconnect geometry, simulation files, S-parameter data, TDR results, and applicable performance requirements. The team can then discuss the appropriate service, scope of work, and next steps. The company provides a contact form and scheduling options through its official website.
Yes. Wild River Technology offers training on S-parameter workflows, VNA calibration, measurement setup, and verification. Training is available on-site or virtually, including remote demonstrations. This support can help engineering teams improve measurement repeatability, understand their measurement data, and integrate reliable S-parameter models into their electronic design automation (EDA) workflows.
Yes. Wild River Technology can design and build test vehicles tailored to a customer’s technology roadmap. A test vehicle can help evaluate fabrication variables such as impedance, etch variation, and material properties, while providing structures for 3D electromagnetic modeling and measurement correlation. This gives engineering and fabrication teams objective data for validating manufacturing processes before committing to larger production runs.
The services include electromagnetic modeling of interconnects, vias, transmission lines, and RF connector launches, along with material identification and simulation-to-measurement correlation. Wild River Technology identifies Ansys HFSS and Cadence Clarity as tools used in its modeling workflows. Depending on the project, deliverables can include models, measurement comparisons, material identification reports, and supporting PCB libraries.
Can Wild River Technology characterize PCB materials for more accurate signal integrity simulations?
Yes. Wild River Technology offers a material identification service that uses dedicated test structures, precision measurements, and 3D electromagnetic simulation to characterize PCB dielectric properties. The process can determine dielectric constant (Dk) and dissipation factor (Df) in the X, Y, and Z directions, helping engineers account for material anisotropy. The program includes pre-fabrication modeling and post-fabrication verification to improve simulation accuracy.
An RF connector launch can contribute significant reflections and frequency-dependent loss to a high-speed measurement or serial link. Wild River Technology uses 3D electromagnetic modeling and measurement validation to optimize launch geometry for specific application goals. For serial link characterization, the optimization can account for the intended data rate and coding; for de-embedding applications, broadband impedance matching and return loss are especially important.
Wild River Technology develops test fixtures around the application’s electrical requirements, operating frequency, connector configuration, and measurement objectives. Its services include high-speed signal integrity fixtures, ATE device interface boards, SERDES characterization platforms, backplane fixtures, and RF connector launch optimization. The company describes fixture capabilities extending to 110 GHz, with the actual validated performance dependent on the specific design and test requirements.
Yes. Wild River Technology lists emergency engineering and redesign among its signal integrity services. This support is intended for urgent engineering problems, including failing or non-compliant designs. The team can evaluate the immediate technical challenge and determine an appropriate path toward diagnosis and corrective action. Availability and turnaround depend on the project’s scope and should be discussed directly with the company.
Yes. Wild River Technology offers SI analysis and debugging services that use simulation, TDR measurements, S-parameter analysis, and root-cause diagnostics to investigate electrical failures. Depending on the problem, the work may include examining impedance discontinuities, excessive channel loss, crosstalk, connector launches, PCB stack-up assumptions, and discrepancies between simulated and measured results.
Wild River Technology provides signal integrity modeling, analysis, and debugging; emergency engineering and redesign; 3D electromagnetic modeling; custom test fixture design and validation; RF connector launch optimization; precision VNA measurements; and PCB design and fabrication support. These services help engineering teams identify electrical performance issues, validate designs, and improve high-speed system reliability.
Wild River Technology offers signal integrity modeling, analysis, and debugging; precision VNA and TDR measurements; RF connector launch optimization; high-frequency test fixture design and validation; and channel modeling platforms. These capabilities help engineers investigate difficult electrical problems, validate simulations against hardware, and improve confidence in high-speed designs. The best starting point is to define the failure or performance target and identify the measurements, models, or design data needed to investigate it.
Engineers can compare measured and simulated channel responses against the applicable electrical specifications and channel budgets. Relevant checks may include insertion loss, return loss, crosstalk, impedance, eye opening, and bit error rate (BER). Depending on the interface, compliance analysis may also use Channel Operating Margin (COM) or statistical link analysis. The appropriate limits depend on the specific interface standard and system requirements.
A connector launch transitions a signal between a coaxial connector and a PCB transmission line. Poor pad geometry, via placement, grounding, or impedance matching can introduce reflections and excess loss that compromise both system performance and measurement accuracy. Electromagnetic simulation followed by physical validation helps optimize the launch and improve measurement repeatability at high frequencies.
PAM4 encodes two bits per symbol using four distinct signal levels. Compared with binary NRZ signaling, the smaller voltage separation between adjacent levels makes the signal more sensitive to noise, distortion, and linearity errors. At high baud rates, channel loss, crosstalk, reflections, and equalizer performance all contribute to the challenge of maintaining acceptable bit error rates.
An eye diagram overlays many received signal waveforms to reveal timing and voltage variations. Eye opening indicates the available margin for reliably distinguishing symbols, while eye closure can indicate noise, jitter, intersymbol interference, crosstalk, or channel loss. Eye diagrams are useful for evaluating equalization strategies and diagnosing high-speed serial link problems.
Passivity means a passive network cannot generate net energy. Causality means its response cannot precede the stimulus that causes it. Models that violate these physical properties can produce misleading simulations or unstable time-domain behavior. Checking and, where appropriate, correcting S-parameter data helps ensure it is suitable for trustworthy channel analysis.
A channel modeling platform provides physical test structures, measurement data, and supporting models that engineers can use to validate electromagnetic simulations and material assumptions. Wild River Technology’s CMP-50 and CMP-70 platforms support simulation-to-measurement workflows, with nominal frequency coverage up to 50 GHz and 70 GHz, respectively. Such platforms help teams establish more reliable design processes for high-speed interconnects.
The dielectric constant influences a transmission line’s propagation characteristics and impedance, while the dissipation factor contributes to dielectric loss. Real PCB materials can also exhibit frequency dependence and directional variation, known as anisotropy. Accurate material characterization helps engineers predict delay, impedance, and attenuation more reliably, particularly at very high frequencies and data rates.
Correlation establishes whether a simulation accurately predicts the behavior of a real, fabricated structure. Poor correlation can lead to incorrect design decisions and repeated board spins. Comparing measured and simulated S-parameters, TDR waveforms, and other responses helps engineers refine material properties, geometry, port definitions, and modeling assumptions.
Simulation predicts electrical performance using mathematical models of materials, geometries, components, and interconnects. Physical measurement captures the behavior of fabricated hardware. Simulation helps engineers evaluate designs before manufacturing, while measurement validates the actual result. Using both allows teams to identify modeling errors, manufacturing effects, and design problems that might otherwise remain hidden.
Fixture de-embedding mathematically removes the modeled or characterized electrical effects of test fixtures from a measurement, helping isolate the behavior of the device under test. It is important because connectors, launch structures, and PCB fixture traces can add loss and reflections that distort the results. Wild River Technology’s technical resources cover methods such as Automatic Fixture Removal (AFR) and measurement-based model techniques.
Calibration corrects systematic measurement errors caused by the instrument, cables, and measurement interfaces up to defined reference planes. Without appropriate calibration, measured reflections and transmission can include errors that do not belong to the device under test. At high frequencies, careful calibration, stable connections, and repeatable test procedures are essential for trustworthy results.
S-parameters describe how signals enter and leave a network at defined ports, including reflections and transmission between ports. For example, S11 describes reflection at port 1, while S21 describes transmission from port 1 to port 2. Engineers use measured or simulated S-parameters to model channels, assess loss and matching, and predict high-speed link behavior.
A VNA measures the amplitude and phase of signals reflected from and transmitted through a device over a range of frequencies. These measurements are commonly represented as S-parameters. Engineers use VNAs to characterize insertion loss, return loss, coupling, and other high-frequency behaviors, producing data that can support accurate channel models and simulation.
Time-domain reflectometry (TDR) sends a fast electrical edge into a device or transmission line and measures reflected energy. The resulting waveform helps identify impedance changes, discontinuities, and their approximate locations. TDR is especially useful for validating PCB traces, vias, connector launches, and test fixtures. Wild River Technology offers TDR measurement and signal integrity services for high-frequency applications.
Wild
Crosstalk occurs when electromagnetic fields from one signal path couple energy into another. It can result from closely spaced traces, poorly controlled return paths, connector coupling, or inadequate isolation between channels. Mitigation techniques include optimizing trace spacing and routing, improving reference-plane continuity, controlling connector geometry, and validating the design with electromagnetic simulation and measurement.
Insertion loss describes how much signal energy is lost as a signal travels through a channel. Return loss describes the signal energy reflected back toward the source because of impedance mismatches. Both matter: excessive insertion loss reduces the received signal, while poor return loss can introduce reflections and waveform distortion.
A SerDes channel is the complete electrical path between a serializer transmitter and a deserializer receiver. It may include IC packages, PCB traces, vias, connectors, cables, and backplanes. Its performance depends on insertion loss, return loss, crosstalk, impedance consistency, timing distortion, and the transmitter and receiver’s equalization capabilities.
As data rates increase, the available timing and voltage margins shrink, while channel losses, reflections, crosstalk, and manufacturing variations become more consequential. PAM4 signaling also uses multiple voltage levels, making noise and distortion more significant relative to the distance between adjacent levels. Successful designs require better modeling, measurement, and control of the complete channel.
Signal integrity (SI) is the ability of an electrical signal to travel through an interconnect and arrive at its destination with sufficient timing accuracy, amplitude, and waveform quality for reliable operation. At high data rates, PCB traces, connectors, vias, packages, and cables behave as interconnected transmission lines. Their losses, impedance discontinuities, and coupling can distort signals and increase bit errors.
“We’re building a new family of systems—can you support a test vehicle build?”
Yes, we can help you. We recommend 2 options, both of which WRT can help you with.
- Build a Test Vehicle (TV) suited to your technology roadmap that addresses issues of
- Fabrication validation (impedance, etch variation, etc.)
- 3D EM optimization of key connectors, ability to directly measure with high
integrity de-embedding - Use TV to validate a concerted measurement process, including S-parameter
quality metrics (causality, passivity, etc.) - Cross sectional analysis to determine actual “as-fabricated technology”, versus
“as-designed”. - Material identification fitting for EDA, including X-Y and Z-axis Dk, Df for
glass/resin laminates - Testing and validation of via designs and crux breakout regions.
- The CMP-50 is a VERY good modeling and validation platform for activity focused on the
10G to 56G PAM4 space. With purchase we conduct a 45-minute overview of Channel Modeling
methods. Please contact us for more info or to set up a time.
“When should/shouldn’t I use forward couplers for injecting crosstalk?”
We don’t recommend using forward couplers for injecting crosstalk into serial link systems.
There are no forward couplers in backplane or general serial link systems. Generally, they are wideband systems with channel characteristics unlike PCB t-lines. True power spectral density of the crosstalk is impacted by dielectric, and metal losses in actual PCB transmission systems, and those systems represent models that are used for COM and general serial link analysis. You can’t do any of this with forward couplers.
You should seriously think about using PCB system generated crosstalk, like our XTALK-56 platform. It includes basic crosstalk coupling, via coupling, backplane/connector emulators that include the ability to add crosstalk.
Finally, by having crosstalk structures that have the same loss as our PCB platforms (ISI and XTALK) you can directly determine, easily, how crosstalk impacts transmission reach penalty. By including return loss degradation structures in the XTALK platform you can perform SERDES design of experiments with variables loss, crosstalk, return loss floor through the Nyquist region. Sorry, you can’t do that with forward couplers.
Pluou can add both clock and data to crosstalk in the XTALK platform.
Adding noise to RX input isn’t crosstalk, it is simply noise.
Finally, forward couplers add strange deterministic jitter profiles, while PCB transmission does not. PCB transmission adds DJ exactly like normal transmission PCB.
“When should/shouldn’t I use phase shifters for skew?”
First of all, PCB skew is not phase shifter skew, which often has bandwidth issues. Phase shifters work well for simple narrow band carrier based signaling systems, but not broad-band digital systems.
PCB skew, such as provided by our SKEW-112 platform, has following advantages over phase shifters:
- The skew is calibrated using NIST traceable methodology and recorded to <100femtoseconds or 0.1psec accuracy, there is no calibration. Skew calibration is really challenging, we do all that work for you, and provide program table. We are creating skew matched calibrated cables every day, so we have special measurement techniques using both VNA and TDR. SKEW-112 is easy to use, WRT team provides apps support.
- The SKEW-112 has channel performance like breakout regions or common transmission paths for Serial Link systems, which means SCD21 is reasonable (diff to common mode), and relates to the skew. Backplanes do not have simple skew, the have coupled systems, and they do not have phase shifters. Simple phase shifters offer no coupling, so it only mimics TX skew, not channel skew.
- Signal integrity is exact same for all lengths, so doing Design of Experiments works well since channel performance does not change like it does with phase shifters.
- 1.5 VSWR is 14dB, SKEW-112 maintains over -20dB return loss through 2X Nyquist. For PAM4 systems return loss or S/N ratio through 1X Nyquist is very important.
- How is customer going to calibrate skew between set ups? SKEW-112 will work in multiple locations using simple Excel spreadsheet.
WRT applications staff will also calibrate skew of your cables (WRT will do this complimentary right now, later this will be optional) and have our own 2psec, 1psec, or 0.375psec max skew matched cables for purchase. We also provide the measured S-parameters of cables.
“What should we first do before engaging with WRT for services?”
Our experience dramatically helps customers accelerate their signal integrity prowess, but it takes a bit of systematic effort.
First, we need a mutual NDA to have a substantive first meeting. Otherwise, we can speak in general terms, which mostly doesn’t work.
Next, it is a good idea to start a document outlining prioritized signal integrity issues existing presently, also future proofed for next generation activity. Common signal integrity issues include:
- Material ID
- Via design
- TDR impedance of transmission lines using both EDA, time domain S-parameters, and direct TDR impedance analysis
- Differential analysis and differential via design methodology
- Benchmarking EDA tools like Clarity and HFSS
- Signal Integrity training for either measurements or EDA simulation tools
- Measurement analysis using either TDR or S-parameters
- Elements of stack-up design
- Advancing simulation to measurement methodology.
We initially help our customers with an option to add consulting packages for specific objectives.
It would be good to have team consensus on max frequency, data rates, coding, etc.
The Wild River engineering team looks forward to working with you folks.
“Why do forward couplers fall short for aggressor injection in SERDES Testing?”
Forward couplers are often proposed for aggressor signal injection in SERDES testing—but they introduce fundamental limitations that make them unsuitable for accurate crosstalk modeling. In real-world backplane and serial link systems, forward couplers are not present. These systems are inherently wideband, and their transmission characteristics—defined by PCB traces, vias, and dielectric loss—are vastly different from what a directional coupler can emulate.
Crosstalk behavior in practical PCB systems is shaped by dielectric and conductor losses, along with layout-specific factors like via structures and connector transitions. These are critical variables for accurate Channel Operating Margin (COM) and serial link analysis, and none of them are captured using forward couplers.
Instead, consider using a purpose-built PCB crosstalk platform like XTALK-56. It’s designed to replicate realistic inter-symbol interference (ISI) and crosstalk characteristics across a range of channel types.
XTALK-56 includes:
- True PCB-based crosstalk coupling
- Via and backplane/connector emulation
- Return loss degradation structures
- Support for simultaneous clock and data interference
By modeling loss and crosstalk together, XTALK-56 allows engineers to quantify how these impairments affect transmission reach and perform SERDES design of experiments (DoE) with precision.
Crucially, adding noise to the RX input is not crosstalk—it’s just noise. Forward couplers inject deterministic jitter profiles that don’t match the stochastic nature of real PCB crosstalk. XTALK-56, by contrast, introduces jitter in a way that reflects actual high-speed PCB transmission paths.
If your goal is to evaluate the true impact of crosstalk on SERDES performance, I strongly recommend exploring XTALK-56 over any forward coupler-based approach.
“Are test platforms available for evaluation?”
Wild River Technology does not typically offer test platforms for general evaluation. However, in select cases, we may provide access to a test platform based on the specific needs and qualifications of the project. Evaluation opportunities are reviewed individually and may depend on factors such as technical alignment, timeline, and intended use.
If you’re interested in evaluating one of our platforms, please reach out to Wild River Technology Sales to discuss your requirements. We’re happy to explore whether a tailored evaluation arrangement makes sense for your situation.
“How much does your xxx cost?”
Please contact us at sales@wildrivertech.com for a quote.
