Precision cooling that makes nanometres possible

Deionised water is standard in semiconductor manufacturing, but it attacks aluminium: it lacks ions, so it dissolves them out of the material and the protective oxide layer does not stay stable. The result is pitting corrosion and particle release into the circuit, and neither is acceptable in a fab. For DI water circuits we therefore work with stainless steel, copper with a suitable internal surface, or aluminium plates with copper or stainless steel tubes press-fitted into them, so that the medium only touches the tube material. This separation of load-bearing structure and media-wetted surface is the real advantage of press-fit technology in this environment. Limit: copper is not permitted everywhere either. In areas with a copper contamination ban, the media-wetted surface has to be stainless steel.

Vacuum compatibility is not a property of the cooler alone but a question of design, surface and cleaning. Trapped volumes are critical: blind holes, overlapping joints and threads hold residual air and moisture that outgas for hours under vacuum. We solve this by design with through-holes, relief grooves and joining processes without cavities. For low-particle environments, post-treatment is added: defined surface roughness, burr-free edges and a documented final cleaning, packed in cleanroom film. What we need for this is the target environment: vacuum class, cleanroom class, permissible outgassing rate. Limit, honestly stated: we do not carry out outgassing testing to specification (e.g. residual gas analysis) or cleanroom final assembly ourselves; we design and manufacture to vacuum requirements and agree the verification with your test laboratory.

The temperature stability achievable at the component surface depends almost entirely on the chiller and its control, not on the cooler. The cooler does determine how strongly a fluctuation in the supply line reaches the component. Two design levers are decisive: a high thermal capacity of the plate damps rapid fluctuations, and uniform channel routing prevents different areas of the plate from reaching different temperatures. For stability-critical applications we therefore design the channels not for minimum pressure drop but for uniform temperature distribution, and verify this by simulation. Limit: if your specification demands stability in the range of a few hundredths of a kelvin, that is a control and chiller task. We supply the plate that does not spoil it, but not the control quality.

An NDA before the first technical exchange is the norm in the semiconductor industry and routine for us. In practice it works like this: you send your NDA template to your sales contact, we review it and come back with approval or comments; only then do we exchange geometries, performance data and simulation results. If you do not have your own template, we provide one. Please state in your enquiry which address the agreement should go to. The most common delay arises when the NDA lands in a general inbox and the responsible engineer never hears about it. Limit: we do not use drawings or design results from your project for other customers; conversely, we cannot show existing designs of other customers as references.

Aluminium and ceramic can be joined, but not with the same processes as metal to metal: the very different thermal expansion creates stress in the joint zone with every temperature change. Common routes are bonding with a thermally conductive, strain-tolerant adhesive, soldering via a metallised ceramic surface, or a clamped joint with a thin thermal interface foil. Which route fits depends on three inputs: the required thermal resistance, the temperature cycling range, and whether the joint has to be leak-tight. For assemblies in which the ceramic is both insulator and lid of a cooling channel, leak tightness over temperature cycles is the critical point. Limit, honestly stated: active brazing of ceramics is not a standard process for us. We assess feasibility but may bring in a specialist rather than improvise it in-house.

We are not the right choice if you need a catalogue product in high volume at market price and no customer-specific geometry: a catalogue or mass manufacturer is cheaper there, and rightly so. We are equally not a fit if the requirement includes complete cleanroom final assembly with residual gas analysis and a particle certificate per part. That is the job of a manufacturer specialising in vacuum components. Our strength lies where a geometry has to be newly developed, where materials and media are combined in unusual ways, and where several loops between simulation, prototype and series are needed. If your project sounds more like catalogue than development, we will tell you so in the first call. An enquiry that belongs with another supplier only costs us both time.

We work with defined, documented cleaning processes: multi-stage flushing of the internal channels, controlled drying, residual contamination analysis to an agreed test procedure, and cleanroom-compliant packaging with sealing caps. We design the channel geometry so that no dead volumes arise in which particles or media residues could collect. We define the cleanliness class we guarantee together with you at the start of the project, including how it is tested and documented.

Depending on the design and joining process, we achieve leak tightness down to the range of helium leak testing. Pressure testing of every part is our standard; for demanding applications we test with a helium leak test and document the leak rate per part. The target leak rate is set in the specification. It directly influences material, joining process and testing effort, which is why we clarify it before design begins.

Yes. For vacuum applications we select low-outgassing materials and avoid problematic auxiliary substances in manufacturing and cleaning. The decisive factors are surface treatment, the joining process and the avoidance of trapped volumes. If your application requires outgassing rates to a particular standard, we align material and process selection with it and document accordingly.

We process copper, copper-nickel, stainless steel and aluminium. The choice follows three criteria: thermal conductivity, media compatibility and mechanical requirement. Copper offers the best conductivity, copper-nickel considerably higher corrosion resistance against aggressive media, stainless steel the highest chemical resistance and cleanliness, aluminium the best ratio of weight to cost. In many semiconductor projects we combine materials: conductive material at the heat source, resistant material at the medium.

We start from your load case: heat load, distribution across the area, permissible component temperature, inlet temperature, volume flow and installation space. This yields the Rth budget, which we split across contact resistance, material conduction and the convective share. In thermal simulation we test variants of the channel routing until peak temperature and pressure drop both fit. Only then does design work begin, which usually saves a complete sample loop.

Turbulators significantly increase convective heat transfer, in our applications up to 40 % more cooling capacity. The pressure drop rises with it, which is why we use them selectively: only in the sections beneath the hotspots, not along the entire channel length. That keeps the system pressure drop within your pump's budget while the critical spot gets the additional performance. Available in drawn-in tubes and, more recently, in press-fitted tubes as well.

From a 30 mm microchannel cooler to a 3 m (30,000 mm) cold plate, right up to a fully cooled large precision part. What is possible is not decided by size alone but by material, design and tolerance: give us your dimensions and we will tell you in which design we can produce them.

In semiconductor applications that is actually the norm. We cool housings and structural parts, optics and mirror mounts, chucks, vacuum chamber components, stators of positioning drives, and process gas. With mechanical parts it is rarely about maximum heat removal alone but about the smallest possible temperature gradient, because the gradient creates the deformation that costs you precision.

In positioning and handling systems, linear and torque motors generate waste heat right next to the precision mechanics. If it is not removed, it migrates into guides and structure and creates drift. Stator cooling removes the heat directly at the winding before it spreads. For this we design heat sinks that sit closely against the motor geometry, absorb high heat flux densities and still respect the tolerances of the assembly.

Process gas coolers have to achieve two things at once: hold the gas temperature within a narrow band and not impair gas purity. We design the gas-wetted surfaces accordingly (resistant material, clean surface, no dead volumes) and size the heat exchanger area to your volume flow and temperature window. Pressure drop on the gas side is usually the critical secondary constraint; we optimise that in the simulation as well.

From a few centimetres to around 3 m in length, from a single piece for a prototype build to series production. Semiconductor projects typically run with us in small to medium quantities with a high variety of variants, which is what our manufacturing is set up for. We deliver samples in around 6-8 weeks, 2-4 weeks by express, depending on product and complexity.

We supply material certificates, dimensional reports, pressure and leak test records, plus cleaning and packaging records, to the agreed scope per lot or per part. For initial sample inspections we prepare the documentation to your specification. Exactly what you need is clarified before series start, so that the test steps are planned into the process from the outset.

Yes, that is standard in the semiconductor industry and the norm for us. We sign your non-disclosure agreement or provide ours. We never pass on customer projects or application details. Even the references on this page are named only with explicit approval.