|
|
[<< Home](/home#3-front-end-design-panel-charge-3)
|
|
|
|
|
|
[<< Section 3.4](/3-front-end-design/3.4)
|
|
|
|
|
|
## 3.5 Cryogenic LNAs, Bias Monitoring and Control
|
|
|
|
|
|
The low noise amplifiers (LNAs) for ALPACA are being developed and provided by Arizona State University. The long wavelength instrumentation group of professors Christopher Groppi and Philip Mauskopf have developed several cryogenic LNA designs based on commercially available SiGe transistors for radio astronomy applications including kinetic inductance detector readout, heterodyne receiver IF amplification and quantum computing. ASU has provided over 100 of these amplifiers for dozens of institutions around the world, including the NASA GUSTO Explorer Mission of opportunity, where NASA flight qualification was required.
|
|
|
|
|
|
<div align="center">
|
|
|
<img src="../uploads/d127a0aa4663c04692aadc6daadd77b4/Gain_Bias_Sweep.jpg" width="540"></div>
|
|
|
<kbd>Figure 1: LNA Bias Power vs Gain parameter sweep for one prototype amplifier. The LNA delivers 35 dB gain across the ALPACA operating frequencies at less than 18 mW of power dissipation. The data presented here is from an earlier iteration of the LNA design. The prototype and production versions have a bias-tee where the RF line also carries the DC bias current to the amplifier. The exact curve for the production amplifiers is slightly different (also see Figure 3) but they satisfy the requirements identified (>35dB gain at <18mW power dissipation)</kbd>
|
|
|
|
|
|
The ALPACA amplifier is a modification of the ASU-LN2 10 MHz-2 GHz amplifier, optimized for the ALPACA band of 1300-1720 MHz. The amplifier uses a custom package designed for the ALPACA dipole insert and has an output bias tee to allow phantom biasing over the RF output line to eliminate the need for additional bias wiring. The amplifier delivers 35 dB of gain with 18 mW of power dissipation and 4-5K noise temperature across the band, meeting all ALPACA requirements. We have produced a prototype amplifier that was extensively tested, including measurements to characterize the noise parameters of the design by Prof. Leonid Belostotski at the University of Calgary, which resulted in a publication [^a]
|
|
|
|
|
|
![ALPACA Prototype amplifier package assembled.](../uploads/3940d10517ff1858884f4c7a17bfe6c3/ASU_package.jpg)
|
|
|
|
|
|
<kbd>Figure 2: One of the fully assembled cryo-LNA boards installed inside a prototype enclosure.</kbd>
|
|
|
|
|
|
We recently completed full cryogenic characterization of 10 prototype amplifier packages identical to the “production” versions, including measurements of cryogenic S-parameters for all 10 (results shown in Figure 3).
|
|
|
|
|
|
![Measurements of a lot of 10 ALPACA prototype amplifiers taken at 7.5 mW power dissipation (so gain is lower than requirement).](../uploads/dc58fba6e2c34785a4ef64a84f425c78/May2021_Prototype_Measurements.jpg)
|
|
|
|
|
|
<kbd>Figure 3: Cryogenic measurements of 10 final-design prototype LNA after repeated thermal cycling. Please note that the measurements were made at a slightly lower bias point (1.5V, 5mA; 7.5 mW dissipation) and that is why the overall gain is slightly lower than the required 35dB. The noise temperature (T<sub>N</sub>; right axis) of the LNA does not change significantly with bias power. </kbd>
|
|
|
|
|
|
The performance of the amplifiers shows very good consistency and all meet ALPACA requirements. ASU is now building 180 production amplifiers (138 for the instrument and 42 spares). The mechanical enclosures are complete and the fabrication of the circuit boards are underway by the vendor that produced the GUSTO flight amplifier boards. ASU will also provide DC bias boards for the amplifiers (8 channel bias boards on standard 3U x 160 mm cards), and 8-channel flexible IF stripline circuits similar to those flight qualified and delivered for the GUSTO mission.
|
|
|
|
|
|
### Rigorous Cryogenic Test and Measurement
|
|
|
|
|
|
Once fabrication and assembly of all parts for the LNAs are completed, ASU will test the production amplifiers. Our previous experience with our SiGe amplifiers is that nominal S-parameter performance at room temperature is an extremely good predictor of good performance at cryogenic temperatures. We have developed a test procedure to increase speed and decrease cost of testing while still providing high performance.
|
|
|
|
|
|
1. All amplifiers will be cryogenically heat cycled from room temperature to 15K six times in a non-operational configuration.
|
|
|
2. All amplifiers will then be tested for S-parameters at room temperature. Performance will be compared to the lot of 10 prototype amplifiers.
|
|
|
3. Any amplifiers that fail step 2 will have their boards replaced. We will repeat steps 1 and 2 until all amplifiers pass.
|
|
|
4. We will randomly select 20% of the amplifiers and perform full cryogenic testing (cold S-Parameters and noise measurements) on those 36 amplifiers. We expect all 36 to perform nominally. If an unacceptably high failure rate is observed, we will determine the root cause and mitigate through additional warm and cold testing.
|
|
|
|
|
|
[^a]: [Sheldon, A. et al., Cryogenic Noise-Parameter Measurements: Recent Research and a Fully Automated Measurement Application, IEEE Microwave Magazine, Aug 2021)](https://ieeexplore.ieee.org/document/9475609).
|
|
|
|
|
|
[Section 3.6 >>](/3-front-end-design/3.6) |
|
|
\ No newline at end of file |