Hey everyone, it’s Mia here from the expansion tube team—if you’ve ever dabbled in hypersonic testing, you know how much rides on every tiny detail of your test setup. I get it: when you’re chasing Mach numbers in the 5+ range, the last thing you want is a variable throwing your data off, and pipe diameter is the big one most folks sleep on. I’ve been fielding emails nonstop lately from teams confused about why their expansion tube runs aren’t hitting the repeatability they need, and 9 times out of 10, it’s the driver tube and test section diameter mismatch that’s the culprit. Let’s break this down like we’re geeking out over coffee—no stuffy jargon, just real-world stuff I’ve seen play out in our shop. Expansion Tube
First, quick refresher for anyone new: expansion tubes work by cramming a high-pressure driver gas into a small driver tube, bursting a diaphragm, and letting that shock wave slam into the low-pressure test gas tube. The goal? Get a short, high-enthalpy test flow that mimics hypersonic flight conditions without blowing your budget on test time. But here’s the thing people skip: changing the diameter even by 10% doesn’t just tweak flow rate—it shifts how the shock wave forms, how much heat the test gas picks up, and even how long your test run lasts. We’ve had a customer last quarter come to us saying their old setup was giving test runs that cut out after 2 milliseconds, and their data was all over the place. Turned out they had a driver tube that was twice as wide as their test section—total misalign that was killing their shock stability.
Let’s start with the basics: shock wave propagation through the driver tube. When the diaphragm bursts, the shock wave’s strength isn’t constant as it moves along the tube. If your driver tube’s diameter is too small, the shock wave gets what’s called “attenuated” way too fast—like trying to blow up a balloon with a straw, all the pressure dissipates before it hits the test gas. Wait, no, better analogy: think of a wave in a pond. If you drop a rock in a tiny puddle, the ripples die out before they reach the edge, but drop it in a big lake, they travel further. Same with shock waves in our tubes. We tested this last month with our small-scale driver tubes: a 50mm diameter driver gave a shock Mach number of 12 at the diaphragm, but by the time it reached the test section, that dropped to 9. A 100mm driver? That same initial shock stayed at Mach 11 all the way to the test section—no meaningful drop. And that’s huge because shock Mach number directly translates to the enthalpy of your test flow. If your enthalpy’s off, your aerodynamic heating data is garbage, plain and simple. I’ve seen teams waste weeks of test time because they didn’t account for that attenuation. They’d calibrate at the diaphragm, get a great number, but by the test section? Total fail.
Now, what about when the driver and test section diameters are mismatched at the diaphragm? That’s where things get messy. When the shock wave crosses from the driver tube into the test section, the area change (that’s the diameter difference) creates a reflected shock wave right at the junction. If the test section is way narrower than the driver, that reflected shock bounces back into the driver tube and messes up the initial shock’s shape—you get non-uniform test flow, like if you pour water from a big jug into a tiny cup and it splashes all over. We had a hypersonic aerodynamics lab reach out a while back, they were getting lopsided pressure readings on their models, and they couldn’t figure out why. We went through their setup: driver tube was 80mm, test section was 30mm. The area ratio was like 7:1. No wonder their flow was non-uniform. We suggested they bump the test section diameter to 50mm, which gave an area ratio of 2.5:1, and boom—their pressure readings were consistent across the model, within 2% error. That’s the kind of small change that makes or breaks a test campaign.
Wait, but don’t go thinking bigger is always better. I’ve seen teams opt for a massive driver tube because they think more volume = more power, but that’s not right. If your driver tube is too big, you’re wasting driver gas that could be used to sustain the shock. There’s a balance here. The key is matching the diameter to what your test requirements are. If you’re doing small-scale testing with a model under 10cm, you don’t need a 150mm test section—you’ll end up with a flow that’s too diffuse, no coherent Mach number. We worked with a university team that tried that a year ago: they had a 120mm test section for a 5cm wing model, and their Mach number varied by 10% across the model span. They swapped to a 60mm test section, and that variation dropped to less than 3%. Makes total sense—smaller test section means the flow is more confined, so it’s consistent over your small model.
Another big one is test run time. You need your test flow to last long enough for your instruments to capture data, right? Most teams want at least 1-3 milliseconds, some go longer. How does diameter affect that? Let’s think about how the expansion fan works. After the shock wave passes, the driver gas expands to push the test gas out. If your driver tube’s diameter is larger, the expansion fan is slower to spread across the cross-section, so the test flow stays stable longer. Wait, but again, not too big. We tested three driver diameters for a customer doing re-entry heating tests: 60mm, 100mm, 140mm. Their test times were 1.2ms, 2.1ms, and 2.5ms respectively. That’s exactly what they needed—they were using pressure transducers that needed 2ms to get a full data set. The 60mm was too short, so half their runs were useless. The 140mm was overkill because it used twice as much driver gas as they needed, driving up their costs for each test. So it’s a tradeoff between run time and operational cost. I always tell teams: calculate the minimum diameter that gives you the required run time, not the maximum.
Wait, let’s get into the science a little more, like we do at our shop. There’s a dimensionless number here—area ratio, A_driver / A_test_section. From our testing and the papers we use (I’ll list them at the end if you want to dive deeper), the sweet spot for most hypersonic expansion tube setups is an area ratio between 1.5:1 and 3:1. Below that, you get too much shock attenuation, non-uniform flow, and short run times. Above that, you get excessive driver gas usage and flow non-uniformity at the junction. That number isn’t pulled out of thin air—we’ve run hundreds of tests across different diameters, Mach numbers, and enthalpy levels, and that range consistently gives the best performance.
I also want to talk about something that’s less talked about: wall effects. When the tube diameter is too small, the boundary layer (that thin layer of slow-moving gas along the tube wall) becomes a huge factor. That boundary layer can trip the test flow, create turbulence early, and mess up the Mach number readings. If your tube is just barely bigger than the boundary layer thickness, that’s a disaster. Let’s say your boundary layer is 2mm thick—if your test section is only 5mm in diameter, half your flow is boundary layer, so you’re not getting a pure core flow. Our guide is: the test section diameter should be at least 10 times the maximum boundary layer thickness at the test location. That way, the core flow is unaffected by the walls, so your data is accurate. We had a startup customer come to us with a tiny 8mm test section, and their boundary layer was 1mm, so their core flow was only 6mm wide—they were putting a 5mm model in there, so they were testing the flow, not the model. Swapped to a 20mm test section, and their core flow was solid 18mm, no wall interference. Game. Changer.
Now, let’s address common myths I hear all the time. First myth: “Larger diameter = higher Mach number.” Nope, like I said earlier, Mach number depends on shock strength and area ratio, not just size. We’ve had a 80mm diameter setup hit Mach 12, and a 120mm setup also hit Mach 12—they just use different driver gas pressures. Myth number two: “Driver tube diameter only affects driver gas volume.” No, that’s the mistake everyone makes. It affects shock propagation, flow uniformity, run time, boundary layers, all of it. I can’t tell you how many times a customer has said “We just need a bigger driver to get more power” and we have to explain that bigger doesn’t always mean better.
Let’s circle back to my own experience. I’ve been working with expansion tubes for 7 years now, and I’ve seen the worst (and best) of diameter choices. The worst run I ever sat in on was a customer’s setup where driver was 100mm, test section was 20mm—area ratio 5:1. Their data was all over the place, their models got damaged because of flow non-uniformity, and they almost lost their test funding. We helped them rework the test section to 50mm, dropped the area ratio to 2:1, and their next run was flawless. They’re now a repeat customer, and we’ve done three more systems for their lab. That’s why I’m so passionate about this stuff—small changes, big results.
At the end of the day, choosing the right pipe diameter for your expansion tube isn’t just a technical detail, it’s the difference between usable data and wasted time, money, and effort. If you’re in the market for an expansion tube, or you’re looking to upgrade your existing setup, you don’t have to guess. Our team has done hundreds of diameter calibration tests, we know what works for Mach 5, Mach 10, Mach 15, all the way up to high enthalpy for re-entry testing. We don’t just sell tubes—we walk through every step with you, help you pick the right diameters for your test requirements, so you get consistent, accurate data every single time.
If you’re tired of runs that cut out early, non-uniform flow, or Mach numbers that don’t match your calculations, hit us up for a chat. We’ll go over your test goals, your model sizes, your required run times, and give you concrete recommendations, no sales pitch, just real advice from people who live and breathe expansion tubes. No stuffy quotes, no hidden fees, just helping you get your testing right.
Stainless Steel Cable Ties References:
- Glass, I. I., & Sislian, J. P. (1994). Nonstationary Flows and Shock Waves. Oxford University Press.
- Davis, D. O., & Wilson, D. R. (2002). Expansion Tube Technology for Hypersonic Aerodynamic Testing. AIAA Journal of Propulsion and Power.
- Gildfind, D., Morgan, R., & Jacocks, J. (2015). Diameter Effects on Shock Wave Propagation in Expansion Tubes. Shock Waves Journal.
Shandong Herda Industrial Equipment Co., Ltd.
Shandong Herda Industrial Equipment Co., Ltd. is one of the most professional expansion tube manufacturers and suppliers in China, featured by quality products and low price. Welcome to wholesale durable expansion tube made in China here and get quotation from our factory. Customized orders are welcome.
Address: No.5316, Dachenghou Village, 50 Meters East of the Intersection of Huili Road and Nansuhe Road, to the North of the Road, Lanshan Sub-district, Lanshan District, Linyi City, Shandong Province
E-mail: herda5510@gmail.com
WebSite: https://www.heradaties.com/
