• No results found

Vibration-free Cooler for the METIS Instrument Using Sorption Compressors

N/A
N/A
Protected

Academic year: 2021

Share "Vibration-free Cooler for the METIS Instrument Using Sorption Compressors"

Copied!
6
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Physics Procedia 67 ( 2015 ) 411 – 416

1875-3892 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of ICEC 25-ICMC 2014 doi: 10.1016/j.phpro.2015.06.050

ScienceDirect

25th International Cryogenic Engineering Conference and the International Cryogenic Materials

Conference in 2014, ICEC 25–ICMC 2014

Vibration-free cooler for the METIS instrument using sorption

compressors

Y.Wu

a,

*, T. Mulder

a

, C. H. Vermeer

a

, H. J. Holland

a

, B. Benthem

b

, H. J. M. ter Brake

a

aEnergy, Materials and Systems, Faculty of Science and Technology, Unveristy of Twente, P. O. Box 217, 7500 AE, Enschede, The Netherlands bDutch Space B. V., P. O. Box 32070, 2303 DB Leiden, The Netherlands

Abstract

METIS is the “Mid-infrared ELT Imager and Spectrograph” for the European Extremely Large Telescope (E-ELT) that will cover the thermal/mid-infrared wavelength range from 3-14 micron. Starting from a pumped nitrogen line at 70K, it requires cryogenic cooling of detectors and optics at 40 K (1.4 W), 25 K (1.1 W), and 8 K (0.4 W). A vibration-free cooling technology for this instrument based on sorption coolers is under development at the University of Twente in collaboration with Dutch Space. We propose a sorption-based cooler with three cascaded Joule-Thomson coolers of which the sorption compressors are all heat sunk at the 70K platform. A helium-operated cooler is used to obtain the 8K level with a cooling power of 0.4 W. Here, three pre-cooling stages are used at 40K, 25K and 15K. The latter two levels are provided by a hydrogen-based cooler, whereas the 40K level is realized by a neon-based sorption cooler. Based on our space-cooler heritage, our preliminary design used sorption compressors equipped with gas-gap heat switches. These have maximum efficiency, but the gas-gap switches add significantly to the complexity of the system. Since in METIS relatively high cooling powers are required, and thus a high number of compressor cells, manufacturability is an important issue. We, therefore, developed an alternative cylindrical compressor design that uses short-pulse heating establishing a thermal wave in radial direction. This allows to omit the gas-gap heat switch. The paper discusses the adapted cell design and two METIS cooler demonstrator setups that are currently under construction.

© 2014 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the organizing committee of ICEC 25-ICMC 2014.

Keywords: vibration-free; sorption compressor; Joule-Thomson cooler; METIS instrument; E-ELT

* Corresponding author. Tel.: +31 53 4892127; fax: +31 53 4891099. E-mail address: y.wu@utwente.nl

© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

(2)

E-ELT has a primary mirror of 39 m in diameter, which captures 15 times more light than the largest optical telescopes operating at this time and it vastly advances astrophysical knowledge and allow for detailed observations of among others the first objects in the universe and planets in other star systems (Gilmozzi, 2007). E-ELT will have several science instruments, and METIS is one of them. METIS is proposed for optical and infrared wavelengths covering the L, M and N bands and offering imaging and medium-resolution spectroscopy over the range of 3-14 microns (Brandl, 2010, Brandl, 2012). Like in most infrared optical systems for astronomy, cryogenic cooling is essential for the METIS instrument to achieve a high resolution. Multiple temperature levels are required for the different METIS components. In configuration trade-offs the number of required temperature levels has been reduced to four. An operating temperature of 8 K is required for the N-band detector, 25 K for the N-band imager and 40 K for the LM-band detector. While other cryogenic units in the METIS have to operate at 85 K. The 85 K level will be established by a liquid-nitrogen (LN2) bus whereas 40 K, 25 K and 8 K have to be provided by a cryocooler. One of the challenges for developing the cryogenic cooling system of the METIS instrument is limiting the vibrations introduced by the cooling system. Mechanical coolers such as Stirling, Gifford-McMahon or even pulse-tube coolers vibrate intrinsically because of the active moving parts and the oscillating pressurized working flow. Dedicated designs in the control electronics and the mechanical structure can reduce the vibrations at the detector level. However, this is complicated and costly and it is still questionable whether the reduction is adequate. Significant degradation in the imaging quality may arise from vibrations at the detectors that are not properly reduced. Equally important is the short-term temperature stability of the cooling system at the cryogenic interfaces for preventing calibration errors due to changing detector temperatures.

A vibration-free cooling technology based on sorption coolers is proposed for the METIS instrument. Sorption Joule-Thomson (JT) coolers have been developed for over a decade at the University of Twente in collaboration with Dutch Space (Burger, 2007, ter Brake, 2011). A sorption-based JT cooler has no active moving parts so it is free of vibration and EMI and has long-life potential. Furthermore, the JT cold stage is flexible to integrate with an instrument like METIS which has a complex mechanical structure and requires distributed cooling at multiple temperature levels.

The system level design of the METIS sorption cooler has been reported in a previous study (Wu, 2014). In this paper, we present the recent developments, including new sorption compressor design, the corresponding adaptation in the system design, and the experimental demonstrators.

2. METIS cooler chain

A sorption JT cooler chain was designed to meet the requirements of the METIS instrument (ter Brake, 2012). As shown in Fig. 1, the chain includes three cold stages thermally linked in parallel, to provide cooling at 40 K by neon, 25 K by hydrogen, and 8 K by helium. The helium stage is driven by a single-stage sorption compressor and uses four counter flow heat exchangers (CFHXs), followed by a JT restriction and a cold-tip heat exchanger (CHX). In order to maximize the achievable performance of the helium stage, pre-cooling heat exchangers are applied at 40 K, 25 K and 15 K. The lowest precooling stage at 15K is most critical in this respect. It has the highest effect on the overall efficiency of the cooler chain. Here, we can use the heritage from a previous 15K hydrogen-based sorption cooler project (ter Brake, 2011). Because the cooling temperature of 8 K is above the critical temperature of helium, the gas will not liquefy during expansion and a well-designed gas heat exchanger is needed to transfer the heat load to the cold helium gas at the cold tip. A 15 K operating temperature requires a hydrogen low pressure of 0.1 bar. Since the required high pressure is higher than 20 bar, a two-stage compressor is required in the hydrogen cooler. Finally, a neon-operated stage delivers the required cooling power at 40 K. This stage uses a single-stage compressor and its cooling capacity is split into cooling of the METIS L/M-detectors and precooling of the helium and hydrogen stages. The sorption compressor cells are thermally attached to a heat-sink that is cooled by a pumped LN2 bus. Saran carbon, which is well studied and has been used successfully in earlier developments at the University of Twente, is used in all compressors.

A quasi-static thermodynamic model (Wu, 2014) was built to optimize the operating parameters such as the working pressures, the heating temperatures and the heat sink temperature of the sorption compressor cells. Based on this model, an analysis of the total input power sensitivity with respect to the effectiveness of CFHXs was carried out and a set of required CFHX effectiveness was determined for the further design.

(3)

Fig. 1. Conceptual design of METIS cooler chain

3. Switchless sorption compressor concept

Fig. 2. Schematic of a sorption compressor cell. Top: GGHS configuration; Bottom: switchless configuration

The original baseline design of the sorption compressor for METIS cooler was based on a gas-gap heat switch (GGHS) configuration that is designed for space applications. An effective GGHS is difficult and expensive to realize. In the case of a relatively large-scale cooler for ground application such as the METIS cooler, complexity and costs may become more important whereas the cooler efficiency is not that critical compared to a space cooler. Therefore, an alternative switchless sorption-compressor design using short-pulse heating was developed (Wu, 2014). This concept replaces the GGHS by a switchless insulation layer as shown schematically in Fig. 2. The solid insulation layer has a higher conductance than the GGHS in the OFF state. In order to reduce the conduction loss to the heat-sink during the heating phase, a short high-power pulse heating is required. Rapid and short heating will generate a radial temperature gradient in the adsorbent that results in a degradation of the performance. This loss of

ph pl pi ph pl ph pl 40K 15K 8K 25K Heat

loads Internalthermal links

H2-CFHX-1 H2-CFHX-2 H2-CFHX-3

Ne-CFHX

He-CFHX-1 He-CFHX-2 He-CFHX-3 He-CFHX-4

C ontainer B ody (thin-wall tube) H eat-s ink (clamps )

C ontainer E nd C ap

H eater (hot part) H eater (cold part) G as Inlet tube

S pacer s upport G as channel

Ads orbent C arbon G G H S ads orbent

G G H S tube Thermal s ens or

F ilter

Wiring H eat-s ink (thin-wall tube)

G G H S heater

Wiring

H eater (cold part) Wiring Thermal s ens or

H eater (hot part) G as channel

Insulator

H eat-s ink (cooper clamps ) H eat-s ink container (thin-wall tube) Ads orbent C arbon

(4)

A dynamic thermal model was built and experimentally validated, for evaluating this switchless concept and for optimizing the compressor-cell dimensions and operating parameters. The dynamic model can simulate the radial temperature gradient in the cell while the pressure and the amount of working fluid stored in the cell are calculated versus time. A typical simulation result for the helium stage compressor cell is presented in Fig. 3. In the compression phase, a high-power pulse heating of 150 W is applied, resulting in a rapid temperature increase in the heater and the inner carbon. It is noticed that a part of the heat is lost via conduction to the heat sink. A large radial temperature gradient of more than 130 K is built in the carbon as shown in Fig. 3 (d). The pressure in the cell increases to the high pressure of the JT cold stage, then the high-pressure check valve opens and the working fluid flows out of the cell at a constant pressure. Although the heating power is switched off after 9.1 s when the average temperature of the carbon reaches the optimum high temperature, the out-flow continues for a few seconds since the temperature wave in the carbon moves from inside to outside. Once the out-flow stops, the pressure decreases as the carbon is cooled down by the conduction heat transfer to the heat sink and re-adsorbs gas. The pressure-reduction phase takes about 20 s, followed by the in-flow phase. The in-flow rate peak is less high than the out-flow peak but it lasts longer as the temperature difference between the carbon and the heat-sink decreases.

Based on the operating condition optimization the dimensions of the switchless sorption compressor for METIS cooler were optimized. The results are presented in Table 1. The dynamic model was also used for predicting the performance and the required cell numbers of the METIS cooler. A comparison was made between the GGHS and the switchless configuration. The later uses 12% more input power, however it requires 13% less cells.

Fig. 3. Typical simulation results of the helium compressor in the METIS cooler, from the dynamic sorption compressor model. The vertical dash lines splits the whole cycle into 4 phases, i.e. compression, out-flow, pressure-reduction and in-flow.(a) the heating power and the conduction heat flow rate to the heat sink; (b) the pressure in the compressor-cell; (c) the mass flow rate generated by the compressor-cell; (d) the temperatures at the different radial positions of the compressor-cell;

(5)

Table 1 Dimensions of the switchless sorption compressor for METIS cooler

Dimensions Unit Neon stage 1st stage Hydrogen stage 2nd stage Helium stage Container wall thickness mm 1.30 0.40 0.40 0.25

Insulation thickness mm 2 1.5 2 1.25 Material of the insulation - Kapton Kapton Kapton Kapton Diameter of the adsorbent mm 14.5 14.5 14.5 14.5

Length of the cell cm 50 50 50 50

4. Demonstrators for METIS

Fig. 4. Schematics of the helium JT cold stage demonstrator

Fig. 5. Schematics of the helium sorption compressor demo; 4 cells are arranged in parallel mounted on the GM cooler cold stage (only one cell is shown in the figure).

4.1. Helium JT cold stage

The JT cold stages of the METIS cooler consists of several counter flow heat exchangers (CFHXs), pre-cooling heat exchangers (PreHXs), Cold-tip heat exchangers (CHXs) and JT restrictions (see Fig. 1). The helium stage occupies almost 60% of the total input power and 65% of the total number of cells, while the coldest CFHX of the helium stage requires a high efficiency of 99.8%. Therefore, a demonstrator of this coldest part of the helium cold stage is under development. This demo setup will be representative of the METIS cooler in terms of cooling capacity, mass flow rate and cooling temperature, but will be precooled by a GM cooler (Sumitomo RDK408D2) and driven by a conventional compressor. The aim is to validate the cold-stage performance, in particular regarding the heat exchangers. Fig. 4 schematically presents this demo setup. A mechanical compressor drives the JT cold stage. Since this compressor will generate a much larger mass flow rate than that required by the demo, a bypass

G M c ool e r 1 5 K r a d ia tio n s h ie ld 70 K r ad iat io n sh iel d V a c u u m b e ll ja r C F H X -1 300-70 K C F H X -2 70-15 K C F H X -3 15-8 K CHX@8 K 15 K 70 K PreHX-1 PreHX-2 T5 T4 T6 T3 T8 T7 T2 T1 JT restriction MF C P1 P2 Bypass valve 1 Rotary valve H3 H2 H1 JT Compr. Bypass valve 2 Thermal straps GM Compr. B yp ass l o o p f o r f ast co o li n g d o w n G M c ool e r 7 0 K r a d ia tio n s h ie ld V a c u u m b e ll ja r 40 K 67 K T3 P1 P2 Rotary valve MFC 7 0 K He a t-s in k S o rp ti o n co m p resso r cel l Thermal straps T6 T5 T4 T2 T1 H1 H2 CFH X 300-70 K Check valves ph lin e pl lin e pl buffers ph buffer Restrict valve GM Compr.

(6)

precooled to 15 K before entering the last CFHX (CFHX-3). CFHX-3 is configured as a tube-in-tube heat exchanger with 1/8*0.028 inch inner tube and 1/4*0.049 inch outer tube. In order to achieve a 99.8% efficiency, it has to be 7.5 m long. With additional margin, a 11.44 m CFHX was made for the setup. The JT restriction is made of a capillary tube with a length of 15 cm and an inner diameter of 0.203 mm. The CHX at 8K is made of 1/8*0.028 inch tubing wound on a copper cylinder. The required heat transfer area of the CHX has to be carefully calculated in order to transfer 0.5 W (with 25% margin) cooling power at 8 K. The calculation shows that 0.8 m long tube can result in a 99.7% efficiency for the CHX, while 1.3 m long tube is finally chosen. Furthermore, as shown in Fig. 4, a bypass loop was applied to create a large flow and cool the sub-15 K part faster during the cool-down process. The bypass loop will be closed by closing bypass valve 2 when the cooler cools further to 8 K.

4.2. Scaled helium sorption compressor

A scaled-down version of the METIS helium compressor is under development in parallel to the helium JT cold stage demo. It is equipped with 4 cells of the same dimensions as those for METIS and it operates at the adsorption temperature of METIS (70 K). This demo setup is schematically illustrated in Fig. 5. The 70 K heat sink will be maintained by a GM cooler. The cylindrical sorption compressor cell is tightly clamped by two copper blocks that are mounted on the 1st stage of the GM. To establish a uniform temperature in longitudinal direction, the cross-sectional area of the copper block is optimized and additional thermal traps link the top and middle part of the heat-sink to the 2nd stage and 1st stage of the GM cooler respectively as shown in Fig. 5. With a proper design of the dimensions of the thermal traps and the cooling temperatures of the GM cooler, the longitudinal temperature gradient of the heat-sink can be limited to 1 K, whereas the temperature swing in an operating cycle is less than 0.2 K. Buffers are applied in the setup to store the working gas when the setup is idle at 300 K, and also to dampen the pressure fluctuations resulting from the small number of cells. A 2 liter buffer is applied at the high-pressure side and a 0.5 liter at the low-pressure side. This combination reduces the storage pressure to 58 bar and the pressure swing down to about 1.3 bar (peak-to-peak) both at the high- and low-pressure sides. The pressurized gas flows through a CFHX, and is warmed up to room temperature for measuring the pressures and the mass-flow rate. The aim of this demo setup is to validate the design method and the performance of the helium sorption compressor cell for the METIS cooler.

5. Acknowledgments

This research is enabled through the Netherlands Research School for Astronomy (NOVA) by financial support from the Netherlands Organization for Scientific Research (NWO) under contract 184.021.006.

References

Gilmozzi, R., et al., The European Extremely Large Telescope (E-ELT), The Messenger 127: 11-19 (2007).

Brandl, B.R., et al., Instrument concept and science case for the mid-IR E-ELT imager and spectrograph METIS, Proc. SPIE 7735, (2010): 77352G.

Brandl, B.R., et al., METIS: the thermal infrared instrument for the E-ELT, Proc. SPIE 8446, (2012): 84461M.

Burger, J.F., et al., Long-life vibration-free 4.5 K sorption cooler for space applications, Review of Scientific Instruments 78 (6): 065102-065110 (2007).

ter Brake, H.J.M., et al., 14.5 K Hydrogen Sorption Cooler: Design and Breadboard Tests, Cryocoolers 16, (2011): 445-454.

Wu, Y., et al., Sorption-based vibration-free cooler for the METIS instrument on E-ELT, Advances in Cryogenic Engineering 59A, (2014): 142-147.

ter Brake, H.J.M., et al., Sorption-based vibration-free cooler for the METIS instrument on E-ELT, Proc. SPIE 8446, (2012): 84467O. Wu, Y., et al., Switchless sorption-compresor design, International Cryocooler Conference 18, Syracuse, NY, USA, 2014.

Referenties

GERELATEERDE DOCUMENTEN

Daar komt bij dat het centrale object van deze voordrachtenserie, het Kalman filter. ook buiten de regelteehniek toepassingen kan vinden, bijv. in de

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:.. • A submitted manuscript is

The focus will be on the three agricultural annual festivals: the Feast of Unleavened Bread; the Feast of the Harvest and the Feast of Booths (Exod 23:14-17). These annual

concentrate especially on multi blade rotors, which are typical for water pumping windmills. For the project as a ' whole, slow running rotors are of specific

The two meth- ods naturally complement each other: the novelty detection provides interesting regions in the latent space to explore, at the same time helps the generative mechanism

Hence, it may happen that the profiles of the seed genes match better with the ribosome biogenesis expres- sion pattern than with the background pattern over a sufficiently

BSVMM cannot be covered by existing convergence results, and in order to understand its convergence, it is necessary to exploit the special structure of the problem; thirdly, BSVMM

Table 3: Leave-one-out cross validation set performances PCC and AUROC of LDA, LOGIT and LS-SVM obtained with the full (F) candidate input set (40 inputs) and with the