Oil-free flow of CO2 benefits research centre

The first CO2 cooling system equipped with Lewa pumps was developed and produced by the National Institute for Subatomic Physics Nikhef in Amsterdam for the Hadron Collider beauty (LHCb) experiment. This approach to cooling removes heat by exploiting the phase change of CO2 from liquid to vapor.

The pumps were supplied by Lewa GmbH, a German manufacturer of metering pumps and process diaphragm pumps. They ensure a uniform flow rate, which is essential for continuous and uniform cooling and disturbance-free operation. The Large Hadron Collider beauty (LHCb) experiment is just one of several experiments currently installed at LHC, which also include the A Toroidal LHC ApparatuS (ATLAS) experiment, and the Compact Muon Solenoid (CMS) experiment.

The first CO2 cooling system equipped with Lewa pumps was developed and produced by the National Institute for Subatomic Physics Nikhef in Amsterdam for the Hadron Collider beauty (LHCb) experiment. This approach to cooling removes heat by exploiting the phase change of CO2 from liquid to vapor.

Source: CERN

In order to achieve precise measurements, silicon detectors are built in close vicinity to the interaction point of all experiments. Carbon dioxide cooling plants cool the innermost layers of the silicon detectors down to temperatures as low as −40°C. This is where the diaphragm metering pumps come in. They ensure a uniform flow rate, which is required for continuous uniform cooling and problem-free operation.

For the CMS pixel detector upgrade to be installed in 2016, a new CO2 cooling system featuring a remote head Lewa metering pump has been built and commissioned. The remote-head design prevents the fluid from absorbing the heat produced by the pump motor and gearbox. Unlike standard pumps, remote-head pumps can convey the highly compressed CO2 without heat input.

The first cooling system equipped with Lewa pumps was developed and produced by the National Institute for Subatomic Physics (NIKHEF) in Amsterdam for the Large Hadron Collider beauty (LHCb) experiment. The aim of the experiment is to answer the question why the universe is comprised primarily of matter and not anti-matter. One of the things researchers will look at is the B meson, which contains an elementary particle known as a b quark, also known as a beauty quark, from which the name LHCb is derived. In order to obtain such particles, the LHC must accelerate protons to near the speed of light and induce them into collision. The particles obtained in this way are recorded using special instruments and then analyzed with the assistance of computer programs.

Three remote-head pumps were delivered. One was installed in the prototype of a 15 kWh system and has already been tested successfully. The two others will be used redundantly in the two final systems involved in the CMS experiment where they will pump the fluid without adding heat.

Oil-free CO2 cooling for more precise results.

The LHCb detectors are unlike other recording systems at the LHC, since detection occurs in only one direction. The first sub-detector, called the VeLo (Vertex Locator) is located directly at the collision point. Others are arranged one after the other along a distance of 20 metres. Among other things, the VeLo is used to precisely determine the location of decays and for particle reconstruction. In order to reach the highest possible precision, the entire system must be under vacuum.

Furthermore, to prevent severe radiation damage on silicon sensors, two carbon dioxide loops cool each half of the VeLo detector to about −25°C. Silicon detectors subject to the strong radiation levels of the LHC are vulnerable to two kinds of damage: displacements in the crystalline structure due to non-ionizing energy loss and accumulation of positive charge in superficial layers due to ionizing energy loss. The most relevant effects of these combined radiation-induced damages are a sharp increase of the voltage required for the sensor depletion, an increase of the leakage current (hence of the signal-to-noise ratio), and a sensitive decrease of the breakdown voltage. While huge R&D efforts are dedicated to new generations of ever more radiation-resistant detectors, it is well known that operation at temperatures well below 0°C greatly mitigates these damaging effects.

CERN chose diaphragm metering pumps from Lewa for a number of reasons, but particularly because no oil can be tolerated in the detector cooling circuit, because oil can start to solidify under the influence of radiation and may then cause a blockage in the thin cooling lines. Prior to the use of the Lewa pumps, a compression cycle for CO2 was impossible as no oil-free compressor for CO2 existed on the market. It was only possible to adopt a pumped loop operated by an oil-free pump. Rotary oil-free pumps require lubrication by the circulating refrigerant fluid, but CO2 is a very poor lubricant.

More powerful Lewa LDE-1 diaphragm metering pumps were needed in order to effectively cool the Compact Muon Solenoid (CMS) detectors. These pumps have a remote head design in order to prevent warming of the CO2 or cooling of the oil, which could lead to formation of gas bubbles and require pumping activity to stop.

Source: Lewa Gmbh.

The Large Hadron Collider beauty (LHCb) experiment. The system they use weighs 5600 metric tons and contains a variety of detectors that are used to identify particles and examine their properties.

Source: CERN

Hans Postema, senior mechanical engineer at CERN who is developing CO2-based cooling systems, together with the cooling team of the Detector Technology group, explains: “That's why we use three LEWA ecoflow diaphragm pumps in the detector loop. Unlike other pump types, they have the benefit of not requiring lubrication by the fluid.”

So a membrane pump was the best choice for long term reliability. Furthermore the Lewa pumps have the ability to deliver the completely uniform flow rate that is necessary for continuous, stable, two-phase cooling. This approach to cooling removes heat by exploiting the phase change of CO2 from liquid to vapor. It has the benefit of using significantly less coolant and much smaller pipes than in single-phase cooling. In other respects, handling the coolant is not so easy. Mr. Postema added: “Liquid temperatures can be as low as −50°C, which is within the critical range because CO2 begins to solidify at −57°C. We are currently working within a range of +20 to −40°C for testing purposes. In most cases the temperature is around −30°C.”

However, for the specific case of the ATLAS IBL detector, an operational range extended down to −40°C was required: in this case a two-stage primary chiller was adopted, with carefully in-house designed controls in order to provide the pump with the correct sub-cooling level even at these low temperatures. By taking all this into consideration, the robust membrane pumps were the best choice for long term reliability.

Remote head design

A different detector at the LHC, known as the Compact Muon Solenoid (CMS) experiment, is involved in the discovery of the Higgs boson, the search for evidence of super symmetries, and the study of what happens when heavy ions collide. The tracker used in this experiment contains 25,000 silicon sensors, each of which must be cooled individually. A major advantage of CO2 cooling becomes apparent in this situation: due to the high level of compression, the volume of vaporized CO2 remains very low, allowing the use of very thin tubing with a diameter of just 2 mm. As a result, very little material is needed despite having several hundred cooling tubes.

From 2015, the collision energy in the LHC, operated at 7 TeV during its first run, was increased to 13 TeV and subsequently to as high as 14 TeV. With the increased number of collisions to be recorded, a more powerful silicon detector will be installed in 2016. For this, a new CO2 cooling system, recently commissioned, will be put in operation.

This system will have a total dissipated power of 15 kW, much higher than the LHCb and the ATLAS ones (of the order of 2 kW). For the new plant engineers have chosen Lewa LDE-1 diaphragm metering pumps with a remote head design. The pump head has a cooling jacket and is constructed of 1.4571 type stainless steel. The displacement movement is transferred by way of a liquid column, also known as the hydraulic rod, contained in the connection line. The plunger puts the rod into an oscillating motion, which is forwarded to the valve head. The check valve responds to pressure and alternates between open and closed, inducing a unidirectional, pulsating flow of the fluid in the valve head.

In this way, the remote head design ensures that the displacement system stays out of the critical range in order to protect the system and the surrounding environment. It also prevents warming of the CO2 or cooling of the oil, which would result in the formation of gas bubbles and cessation of pumping action, a common problem with standard pumps. According to Marc Geiselhart, managing director at the Swiss Lewa subsidiary: “In two-phase cooling, the CO2 must be close to its boiling point, as it tends to vaporize at warmer parts of the pump. That's why the low heat input into the fluid is important. This means that diaphragm metering pumps with remote head design can bring substantial advantages to the plant performance.”

Initial tests successful

Since the failure of a pump during an experiment would be very costly in terms of time and money, additional Lewa-specific features help ensure reliability. For example, the two-layer PTFE diaphragm prevents contamination of the CO2 in case one layer of the diaphragm becomes damaged. In addition, an integrated pressure switch triggers immediate shutdown of the pump in the event of leakage. If requirements change, the flow rate can be regulated by remote stroke adjustment via two check valves.

The prototype for the new 15 kW system has already been installed and has successfully passed its initial test. The system is 10 times larger than the one that uses Lewa standard pumps. The two CMS systems have also been fully assembled and have been commissioned. They will run redundantly in the actual experiment. In any further scale-up, even more powerful pumps will be required, but engineers are currently waiting for results from the most recent experiment.

When the LHC started operations, it marked a turning point in the field of particle physics, as it may help unlock answers to fundamental questions, such as the origin of matter. The diaphragm metering pumps used in these experiments are playing a small, but vital part in those momentous discoveries.