Introduction
The increase in operating frequency of RF semiconductor devices presents challenges when interfacing to the device under test (DUT) through automated test equipment (ATE). One of the critical challenges in testing large volume, mmWave RF semiconductors is maintaining repeatable measurements through many cycles of docking/undocking test fixtures.
mmWave Uniqueness
Higher frequencies (shorter wavelengths) mean that the electromagnetic wave will travel more on the surface of the transmission line (skin effect) . For any material, the losses increase as the frequency increases. Connectors/Interfaces used in mmWave test must be fabricated from materials that support the shorter wavelengths, and the geometry of the interface internal structures must also be smaller. In short, extreme mechanical precision is required in all fabricated parts of the RF interface to ensure repeatable, accurate measurements.
Common Issues With ATE RF Interfaces
In the context of ATE, blind mate interfaces often suffer because of misalignment or incorrect mating force. Either misalignment or incorrect mating force can contribute to a resonance at certain frequencies, creating a loss of energy through the transmission line. This is known in the industry as a "suck out". For the case of incorrect mating force, this is illustrated in the figure below. For misalignment a similar phenomenon is observed.
Figure 1 - Mating Force Plot
Below is an example of a real world "suck out". You notice that at around 30GHz, unexpected loss of energy can clearly be observed. In this particular case, a manufacturing defect of five thousandths of an inch (5 mils) in the blind mate interface connector cavity is the cause.
Figure 2 - 30GHz "Suck Out"
TS-9XXe-5G 1.85mm Blindmate Solution
The RF blindmate interface on the TS-9XXe-5G:- 1.85mm (67GHz) geometry (usable up to 53GHz, Keysight VNA limit).
- Sexed design with floating interface.
- Keyed fixture, using both coarse and fine alignment pins.
- Single locking cam, dual release mechanism.
- All mechanical tolerances are held at +/- 0.001".
Figure 3 - ATE Side of Blindmate Interface
Figure 4 - Fixture Side of Blindmate Interface
Materials used in TS-9XXe-5G RF Blindmate Interface:
- Outer conductor (housing) is 303 stainless steel per ASTM A582 and passivated per AMS 2700 Method 1 type 2.
- The material used meets the requirements of DFARS (Defense Federal Acquisition Regulation Supplement); DFARS requires that the steel must be smelted in the US or EU.
- Center conductor is BeCu Alloy UNS C17300 per ASTM B196, heat treated to condition TH04.
- Center conductor plating is gold per MIL-DTL-45204, Type II, Grade C, .000100”-.000150” thick over electro-plated Nickel per QQ-N-290A, .000150” - .000300” thick.
- Plastic Bead is Neoflon per ASTM D1430 or AMS 3650.
TS-9XXe-5G 1.85mm Blindmate Repeatability Demonstrated
Below is a composite (overlay) of the S21 (insertion loss) and S11 (return loss) measurements of our 1.85mm blindmate interface over 100 mating cycles. You will notice the marker values in the top right corner of the measurement window stays almost entirely within one tenth of a dB throughout the 100 mating cycles.
Conclusion
Marvin Test Solutions TS-9XXe-5G blindmate RF interface offers a robust, reliable and repeatable RF blindmate solution that allows users to utilize the full bandwidth of the system's 53 GHz Keysight VNAs.