A 5000 h RF life test on 330 W RF-LDMOS transistors for radars applications O. Latry a, * , P. Dherbécourt a , K. Mourgues a , H. Maanane b , J.P. Sipma b , F. Cornu b , P. Eudeline b , M. Masmoudi c a GPM UMR CNRS 6634, Université de Rouen, 76801 St-Etienne-du-Rouvray, France b THALES AIR SYSTEM, ZI du Mont Jarret, 76520 Ymare, France c IUT, Université de Rouen, Rue Lavoisier, 76821 Mont-Saint-Aignan, France article info Article history: Received 2 July 2010 Accepted 16 July 2010 Available online 8 August 2010 abstract A reliability test bench dedicated to RF power devices is used to improve 330 W LDMOS in a radar con- ditions. The monitoring of RF power, drain, gate voltages and currents under various pulses and temper- atures conditions are investigated. Numerous duty cycles are applied in order to stress LDMOS. It shows with tracking all this parameters that only few hot carrier injection phenomenon appear with no inci- dence on RF figures of merit (P out or PAE). Robustness and ruggedness are shown for LDMOS with this bench for radar applications in L-band. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction The huge demand for cost effective, high gain and high perfor- mance give a place to LDMOS components. They have better per- formances than bipolar transistors and a cost effective compared to GaAs. Today, they are used on base stations, TV broadcast or in radar applications with high capabilities particularly in terms of output power and Power Added Efficiency (PAE). The optimization design of LDMOS operating under continuous peak power in RF applications has been conducted in order to im- proved their reliability [1] for high-power IC developments [2]. Their reliability has been conducted about engineering [3,4]. In the radar field, one of the major subject of concern is the reli- ability of RF power amplifiers [5,6] (whatever the RF transistors technology employed) submitted to RF pulses with a high drain– source DC bias for maximum output power under wide tempera- ture range. In order to qualify new 50 V RF-LDMOS reliability for L-band ra- dar applications, a 5000 h pulsed RF life test has been conducted on a dedicated RF L-band test bench in operating modes [7]. 2. General RF high voltage LDMOS transistor performances The RF-LDMOS device under test is a commercial L-band 330W operating in class B at saturation and 50 V DC biasing. Indeed, these performances are given in conditions of width pulse 300 ls with a duty cycle of 12%. The Power density is equal to 1.89 W/ mm. More than 59% of drain efficiency, and a gain of 17 dB can be obtained around the L Band frequency at 1400 MHz. The junc- tion temperature does not exceed 100 °C for a flange temperature equal to 65 °C. The thermal resistance is 0.13 °C/W. 3. RF pulsed life test This bench is able to keep track many parameters like voltages, currents, base-plate temperature, peak power. The Fig. 1 represents the component under test placed on its test fixture, supplied by DC power and connected to RF connector (type N). In order to obtain operating conditions with mission profile for radar applications, we have carried out three different life tests (see Table 1). During the life test, the goal is to study the compo- nent performances in actual working situation to ensure that it will maintain a good performance level. Therefore, a RF life test bench has been developed [8]. This bench operates in radiofrequency pulse mode. The device under test is placed on a thermal module in order to maintain a constant flange temperature. The command unit manages DC supply voltage, temperature regulation and RF signal monitoring. 4. Results and discussion The bench allows to record temperature, currents and voltages (gate and drain), the input power, reflective and output powers. During the three runs the output power has not drifted as we can see on Fig. 2. The increase of power during the last run is due to the operating temperature (10 °C) of the pulsed RF life test lower than runs 1 and 2 (80 °C). It is the same case about all the P out = f(P in ). The curves have been measured before and after each run in order to have comparisons with each temperature condition. Fig. 3 summaries these results and shows perfect RF pulsed 0026-2714/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.microrel.2010.07.086 * Corresponding author. Tel.: +33 232 95 51 22. E-mail address: olivier.latry@univ-rouen.fr (O. Latry). Microelectronics Reliability 50 (2010) 1574–1576 Contents lists available at ScienceDirect Microelectronics Reliability journal homepage: www.elsevier.com/locate/microrel