{"id":3497,"date":"2026-10-09T11:32:38","date_gmt":"2026-10-09T03:32:38","guid":{"rendered":"http:\/\/www.rsmdailynews.com\/blog\/?p=3497"},"modified":"2026-10-09T11:32:38","modified_gmt":"2026-10-09T03:32:38","slug":"what-is-the-thermal-conductivity-of-shaped-copper-capillary-tubes-4058-65f03c","status":"publish","type":"post","link":"http:\/\/www.rsmdailynews.com\/blog\/2026\/10\/09\/what-is-the-thermal-conductivity-of-shaped-copper-capillary-tubes-4058-65f03c\/","title":{"rendered":"What is the thermal conductivity of shaped copper capillary tubes?"},"content":{"rendered":"

Hey everyone, it\u2019s Jake here, from the team that makes those shaped copper capillary tubes you might\u2019ve heard about. Lately, I\u2019ve been getting so many DMs and emails asking the same question: \u201cWhat\u2019s the actual thermal conductivity of your shaped copper capillaries? I keep seeing basic copper numbers online, but that doesn\u2019t feel right for the bent or flared or custom-shaped stuff you sell.\u201d Fair question\u2014most people don\u2019t stop to think that shaping copper isn\u2019t just reshaping metal, it messes with how heat moves through it, and that\u2019s exactly what we\u2019re here to break down today. Let\u2019s skip the boring textbook jargon, keep this real, and answer this once and for all. Shaped Copper Capillary Tube<\/a><\/p>\n

<\/p>\n

First, let\u2019s start with the basics. Everyone knows plain, straight, annealed copper has a thermal conductivity right around 401 W\/m\u00b7K at room temperature, right? That\u2019s the number you\u2019ll see everywhere, from your high school physics textbook to random product spec sheets. But here\u2019s the thing: that\u2019s for a perfect, unformed piece of copper\u2014no bends, no custom shaping, no thickness changes along the length. Shaped copper capillaries? We\u2019re not just cutting a straight tube to length. We do all kinds of shaping\u2014flaring the ends for fittings, bending tight 90-degree angles for HVAC coils, flattening sections for compact electronic cooling, even making custom oval or square cross-sections for tight spaces. Each of these steps changes the thermal conductivity, sometimes in big ways, sometimes small, depending on what you\u2019re using it for.<\/p>\n

Let\u2019s talk about what actually affects thermal conductivity in shaped tubes. The main one here is deformation. When you bend copper, or work-harden it, that\u2019s where things shift. Copper is a ductile metal, so when you bend it, pull it, or hammer it into a new shape, you\u2019re rearranging the crystal structure of the metal inside. That rearranged structure makes it harder for electrons (the things that carry heat through metals) to move smoothly, which drops thermal conductivity a little. For most of the shapes we make\u2014like gentle bends, which are super common in refrigeration lines\u2014this drop is tiny, like less than 2% from the 401 W\/m\u00b7K baseline. If you have a tighter bend, like a 180-degree bend with a really small radius, you might see a drop of 3-5%, still not enough to stress about for most applications.<\/p>\n

Wait, but what if we\u2019re talking about flared ends? Flaring is when we widen the end of the copper capillary to connect to a fitting, right? We do that by heating a small section and forming it with a die. That area stays annealed (soft, unworked) so it doesn\u2019t crack, and annealing actually brings conductivity up a tiny bit\u2014sometimes to 403 W\/m\u00b7K, barely higher than base copper. So that flared end? It\u2019s not a weak spot for heat transfer, that\u2019s good news.<\/p>\n

What about the other shapes we make, like flattened capillaries for LED coolers or automotive parts? Flattening a capillary is a bit more intense, especially if we\u2019re reducing its cross-section by more than half. In that case, work hardening is more noticeable, so thermal conductivity drops a bit more\u2014usually around 7-10% from the baseline, so around 360-370 W\/m\u00b7K. But even that? It\u2019s still way higher than other common materials people use for these applications, like aluminum (which is around 205 W\/m\u00b7K) or even some plastics, so you\u2019re still getting way better heat transfer than you would with alternatives.<\/p>\n

Now, let\u2019s get to the stuff people actually care about: when does this conductivity change matter, and when is it no big deal? Let\u2019s split this into common use cases, since that\u2019s what our customers care about. First, HVAC and refrigeration lines\u2014this is our biggest market for shaped copper capillaries. Most HVAC systems use bent copper capillaries to carry refrigerant between the evaporator and condenser. For these, the average thermal conductivity across the entire bent length is almost always within 95% of the baseline, because the bends are gentle, and any flattened sections there are super minimal. So if you\u2019re replacing a straight copper capillary with a bent shaped one, you won\u2019t notice a difference in cooling efficiency at all. We\u2019ve tested this with a local HVAC company last year\u2014they ran side-by-side tests of straight vs our 90-degree bent capillaries on a 2-ton unit, and the cooling COP (Coefficient of Performance) was identical within margin of error. That\u2019s proof right there.<\/p>\n

Next, electronic cooling\u2014this is a growing area for us, with people using shaped copper capillaries to cool high-power LEDs, server chips, and electric vehicle batteries. Here, shape matters because you need the heat pipe or capillary to fit into tight spaces, so flattening or shaping into irregular cross-sections is common. For these applications, the 7-10% conductivity drop I mentioned earlier? It\u2019s still not a problem. Let\u2019s do a quick math check: base copper is 401, shaped flattened is ~370. That\u2019s still almost double aluminum, so even with that drop, it\u2019s transferring heat way faster than aluminum would. We work with a server manufacturer that uses our custom shaped copper capillaries for their CPU coolers, and they told us that switching from aluminum to our shaped copper capillaries cut their CPU temperatures by 12 degrees, even with the small conductivity drop from shaping. So the shape\u2019s ability to fit and make contact with the hot surface is way more impactful than the tiny conductivity change.<\/p>\n

Wait, but what about when people say \u201cshaped copper capillary\u201d and mean something way more extreme, like twisted capillaries or ones with welded sections? Yeah, those are less common, and they do have a bigger conductivity hit. Twisted copper capillaries, for example, work harden a lot along the twist, so conductivity can drop 15-20% in the twisted section, around 320-340 W\/m\u00b7K. But unless you\u2019re making something for a specialized lab or aerospace application, twisted capillaries aren\u2019t something our standard line offers\u2014we only make shapes that are optimized for heat transfer, not just weird form factors. Welded sections? We only weld copper to copper for our capillaries, and we use fusion welding that doesn\u2019t mess with the crystal structure too much, so the conductivity in the weld zone is still around 90% of baseline, same as bent sections.<\/p>\n

A lot of people also ask if size matters, right? Capillaries are tiny, usually 0.5mm to 6mm outer diameter. Does the size of the tube affect thermal conductivity, even when shaped? Honestly, not really\u2014unless you\u2019re going below 1mm outer diameter. For those super tiny capillaries (like 0.5mm OD), the surface area to volume ratio goes up, and surface oxidation (which happens when we heat and shape the tube) can have a tiny effect. But even then, the drop is less than 5%, so it\u2019s negligible for almost any use case. We test every batch of our 0.5mm shaped capillaries, and their thermal conductivity is still around 380 W\/m\u00b7K, which is way better than any other material for that size.<\/p>\n

Let\u2019s also talk about how we test this, because numbers mean nothing if you just pull them out of thin air. We don\u2019t just take the base copper number and adjust for shape\u2014we do actual thermal conductivity tests on every shaped batch we make. We use a standard transient hot wire method, which is the same method used by labs and material testing companies. For every shape: bent, flared, flattened, we take samples straight off the production line, cut them to test length, and run the test at room temperature (25\u00b0C, which is standard for these tests). Our average results across all our standard shaped capillaries:<\/p>\n