We cool RF & microwave at every conceivable frequency

Yes. Cooled flanges for chamber interfaces are a typical case for press-fitted tubes: the flange stays a solid, machinable part with a defined sealing face, and the coolant runs inside the tube through the material without touching the sealing function. For RF applications it is decisive that the cooling does not disturb the electrical function: channels and tubes have to be positioned so that they do not form resonant cavities and do not interrupt the continuous metal wall. That is why we fix channel routing and wall thicknesses together with your RF geometry instead of adapting a standard plate. Limit: the RF design itself (field distribution, matching, shielding attenuation) stays with you. We assess the thermal and manufacturing side and tell you which channel routing is mechanically possible.

For RF contact faces the surface is not corrosion protection but part of the electrical function: the current runs at the surface, and a poorly conducting layer increases losses. Standard anodising is electrically insulating and therefore unsuitable for contact faces. Instead you work with electroless nickel, tin, silver plating or with conductive passivation on aluminium. In practice that means partial coating: we anodise the heat sink, then mask the contact faces and treat them separately. That is possible but noticeably increases effort and price. It is important to mark clearly in the drawing which face receives which treatment. Limit: there is no full-surface combination of thick anodising for corrosion protection and good conductivity. Having both at once requires partial coating or a compromise.

For RF output stages copper is chosen because it carries heat and current at the same time and is non-magnetic. Cooling is usually provided by press-fitted or soldered tubes, or by a milled channel structure with a soldered lid. The demanding part is not the heat but the combination: the area under the transistor has to be flat and clean, the channels must not weaken that area, and soldered joints have to stay tight even though copper expands strongly. For spare parts we work from a sample or drawing. Limit: copper is considerably more expensive and heavier than aluminium; where there is no electrical function and no extreme local power density, aluminium with a copper inlay is often the better choice, and we say so when we see it that way.

Mixed load cases are common: transistors of several hundred watts next to resistors that, depending on the operating case, carry no load at all or the main load. In that situation the plate is not designed for an average but for the worst operating case in each area. In practice that means we calculate several load cases separately and dimension the channel routing so that every case stays manageable; often the flow is deliberately routed over the most sensitive component first. An embedded copper heat spreader beneath particularly dense sources is a proven way to distribute local peaks without making the whole plate from copper. Limit: without the operating cases stated, the design ends up either oversized or uncertain, so name all relevant load cases, not just the nominal one.

Heat pipes and two-phase cooling are asked about regularly, and the honest answer separates two cases. Integrating heat pipes as a heat transport element into a heat sink is feasible: the heat pipe itself is a bought-in part that we embed, solder or press in. Developing and manufacturing two-phase systems such as thermosiphons or vapour chambers ourselves, however, is not our core business. Where heat pipes genuinely help is transporting heat over a distance or around a corner, for example out of a tight enclosure to an external cooling surface; where they are often overrated is power density directly at the source. Limit, honestly stated: if your project aims at a true two-phase solution, we will tell you and recommend a specialist rather than selling a compromise.

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 test in thermal simulation against variants of the channel routing before design work begins. In our thermal simulation (FloEFD) we include the variants you release to us. If it has to be a cold plate with a press-fitted tube, we optimise the design within those possibilities. If you give us design freedom, the best variant may well turn out to be a friction-stir-welded, internally structured cold plate.

With a fine channel structure in the liquid cold plate directly beneath the heat source, 3D-printed cooling channels, or turbulators used selectively in drawn-in or press-fitted tubes in cold plates. Where necessary with a change of material via heat spreaders or inlays at the interface. The key is to spread the heat flow within the cold plate as well as possible before it enters the coolant, and then remove it from there as fast as possible.

We work with copper, copper-nickel, stainless steel and aluminium. The choice follows three criteria: thermal conductivity, media compatibility and mechanical requirement. We frequently combine the two: conductive material at the heat source, resistant material at the medium. We also follow our customers' applicable design and release guidelines. In the high-performance range we rarely get past copper in cold plates and heat spreaders, but today we also rely far more on hybrid forms, right up to stainless steel cooling channels in high-performance applications.

Pressure testing of every single liquid cold plate is our standard; for demanding applications we additionally test with a helium leak test or a pressure-drop test and document the leak rate per part. We set the target leak rate together during the design process or the manufacturability assessment.

From a 30 mm microchannel cooler to a 3 m (30,000 mm) cold plate, and 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.

It can, via grounding, earthing and the mechanical interface. That is why we design in dialogue with your RF development, not after it. In many places our coolers also take on far-reaching housing or conductor functions. Whether for mounting a transformer or an ohmic adapter: we make possible what you expect from our heat sink.

Yes. In RF systems a considerable share of the waste heat arises outside the output stage: at circulators, terminating loads and in the structure. We cool the entire path, ideally fully integrated in a single cooler.

Depending on the manufacturing technology we deliver samples in around 6-8 weeks as standard, 2-4 weeks by express. For series production we manufacture in a highly automated plant in Germany; cost-sensitive standard parts come from our joint ventures in Asia. We supply quantities from 1 to 100,000 per year, depending on part complexity and size.

Yes, of course. That is the norm for us. We sign your non-disclosure agreement or provide ours. We never pass on customer projects or application details. Feel free to send us your NDA straight away, then we can start faster.