Michael Steinberger

Eventually, any stationary system you’re working on will use electrical repeaters. This article explains why and demonstrates the type of analysis that’s required when designing with repeaters.

While at first it may seem to be impossible or impractical to create an IBIS-AMI (Algorithmic Modeling Interface) model of DFE (Decision Feedback Equalization), it’s actually quite easy so long as you understand some unavoidable artifacts in the results.

Bert-

The delays in Figure 2 were taken from TDR plots synthesized from the measured S parameter data. I probably forgot to divide by two to account for the fact that they were TDRs.

Regarding an experiment to test for anisoptropy in the dielectric constant, the method I have in mind is to repeat…

Bert-

Thanks for the very clear description of your work and results. I think it’s helpful to have more than one point of view.

There are two key technical points that prevent me from being able to go along with your theory. Also, I have a larger technical point that perhaps you could shed some…

Published data indicates that vias are longer electrically than one might at first expect. The explanation that has been offered is that the effective dielectric constant for waves propagating perpendicular to the plane of the board is higher than that for waves propagating in the plane of the…

Time domain simulations of high speed serial channels are really computational experiments rather than mathematical evaluations. They have confidence limits just like any physical experiment, and users should determine what those confidence limits are. This article demonstrates how these confidence…

Cody-

Thanks for the question. For the data shown in Figure 5, the ground vias were 0.2” from the signal via.

This distance is three or four times the distances found in the via field under a connector. The result is that the ground vias were only effective up to a frequency which is a third or a…

A couple of months ago, I was discussing PC board via electrical behavior with a respected colleague, and he happened to mention that the ground return currents for the via would find the nearest ground via and follow that path. Five years ago, I would have said exactly the same thing.

Thanks for the additional information. In SiSoft’s Quantum Channel Designer, we use a not-so-naive approach to IFFT, and so have not run into this problem.

Regarding the construction of a VNA, the clunking you’re hearing is due to electromechanical switches, and not couplers. The couplers are,…

This was one of the four or five examples of butchered S parameters that I prepared for the “How to Avoid Butchering S Parameters” panel that I hosted at DesignCon2011. These examples covered the range of problems I’ve seen, and the S parameter quality measures would have only ever identified one…