 |
|
 |
|
|
|
 |
The electrical properties of a Synergetix spring pin will depend very strongly on the location and number of return pins. In order to provide useful information to a designer, each of the nine different return pin configurations are measured using a specially prepared, custom socket.
For each configuration, the performance of the signal pin and its return path is measured, and then a model, assuming a uniform transmission line, is used to extract the characteristic impedance and time delay. The quality of the fit between the measured performance and the simulated performance of the uniform transmission line model is excellent to typically above 10 GHz.
The table summarizes the characteristic impedance and time delay of a signal pin for each of the nine return pin configurations. This is a model which can be used directly in a SPICE simulation to evaluate the system performance of the socket. In addition, the impedance mismatch of the pin to the typically 50 O of the system will be the first order factor influencing the insertion loss. Using this extracted model of the signal pin, the insertion loss can be simulated. The frequency at which the insertion loss dips below -1 dB is listed in the table.
Finally, the socket can be approximated by a single section lumped circuit model using a single L and C section up to 5 GHz.
The electrical performance of a single signal pin with a variety of return pin configurations is measured under two conditions: the far end of the probes open and the far end of the probes shorted. From these two measurements, the high bandwidth equivalent electrical circuit model of the probe for each return pin configuration is extracted.
GGB microprobes are used between the VNA and the test fixture circuit board. A custom, two layer circuit board, with a filled thru via, surrounded by a copper fill plane is used. The exact same pattern is used on the top and bottom sides of the fixture board. The copper fill on the bottom side, the surface in contact with the Synergetix socket, is used to bus all the various return pins. Numerous thru vias connected the bottom copper fill to the top copper fill planes.
The bottom of the socket is either left in contact with a piece of blank FR4 to provide an open surface or is pre-loaded against a gold plated FR4 coupon to provide the shorted measurement.
To be able to separate the contribution of the fixture from the fixture and socket combination, the fixture alone is measured when pressed up against a blanket piece of FR4 to obtain its open performance and a special pattern is measured which has the signal via shorted to the bottom copper fill plane to provide a reference of the shorted condition of the fixture board.
A network analyzer is used to perform the electrical measurements of the open and shorted conditions. In the open configuration, the one port S11 measurements of the fixture alone and the fixture and socket combination are taken. In the case of the shorted configuration, a two port measurement is made with both probes touching the same pads on the top of the fixture board. Because the impedance of the shorted fixture and the shorted fixture and socket combination is so low, a special technique is used to obtain accurate values of these two impedances. This method of using two ports on the same pad is analogous to a four port Kelvin measurement which gets around the probe contact impedance effects and allows artifact free measurements of socket impedances significantly lower than 1O , or loop inductances below 0.5 nH.
A simple model is created for the fixture and the signal pin with its associated return pins. In each case, the fixture is modeled as a short, ideal stet transmission line and the signal pin and its return pin are modeled as a short, ideal stet transmission line. For each line, there are only three parameters that completely describe their electrical properties: the characteristic impedance, the dielectric constant, and the dissipation factor. For each structure, the length of the transmission line is taken as the measured length of the structure, so no fitting of this parameter is required.
Four separate measurements are performed for each return path configuration: fixture open and shorted, and fixture plus socket open and shorted. The three parameters of each transmission line are simultaneously fitted from these four measurements. The values in the table are taken from the extracted fits.
To understand exactly how the Synergetix probe will perform in each specific application, a parametric values table has been created for each probe. Because probes can be used on a variety of pitches, values are calculated for each probe from their minimum pitch up to 1,27mm. Below is a sample of available data for each probe. (Each individual probe specification sheet lists the parametric values at the 2a configuration).
|
|
 |
 |
|
 |
|
|
 |
|
 |
