If you’re a wire harness supplier like me, you know that shielding effectiveness (SE) isn’t just a buzzword—it’s the difference between a harness that makes a client’s ECU glitch and one that keeps their entire system running smooth. I’ve been in this game for 12 years, and I can’t tell you how many times a client has come to us panicking because their current harness isn’t blocking enough interference, or they need to meet a strict spec for a new electric vehicle. The problem is, a lot of suppliers act like SE testing is some mysterious, lab-only thing, but it’s actually something you can get right in-house if you know what you’re doing. No fancy university degree required—just some basic gear, a clear plan, and an eye for detail. Let me walk you through how we do it here, the mistakes we used to make, and the wins that’ve kept our clients coming back for years. Wire Harness
First, let’s cut through the jargon. Shielding effectiveness is basically how well your wire harness blocks unwanted electromagnetic radiation (EMI) or keeps internal signals from leaking out (that’s called electromagnetic compatibility, or EMC, if you want to throw the fancy term around). The number you get is in decibels (dB)—higher is better. A SE of 40 dB means you’re blocking 99.99% of that interference; 60 dB is 99.9999%, which is what most auto and industrial clients demand these days, especially with all the ADAS and EV systems popping up.
Before you even touch a tester, though, you’ve got to set the stage. We used to skip this part and get garbage data because we were testing in the middle of our warehouse next to a forklift charging. Bad move. First, find a spot with zero external EMI—either an anechoic chamber (that’s the room with the foam wedges that absorb all radio waves) or a shielded box (called a Faraday cage, basically a metal box with no gaps big enough for waves to sneak through). If you’re testing a finished harness, you also need to make sure it’s terminated correctly—no stripped wires left hanging, connectors clipped down, because gaps are the #1 killer of SE. A shield that’s 100% perfect will still fail if you’ve got a 1mm gap around a connector. We learned that the hard way last year when we sent a batch to a truck manufacturer, and their test failed because one connector clip wasn’t crimped tight enough. Now we do a visual check for gaps before we even run a single test.
Next, pick the right test method, because one size does NOT fit all. Most small to medium wire harnesses (the ones for car doors, control panels, home appliances) use what’s called a transverse electromagnetic (TEM) cell. That’s a little metal chamber that looks like a long tube with two metal plates inside—you slide your harness through it, and it creates a uniform E-field (electric field) around the wires. It’s cheap, portable, and perfect for mid-frequency stuff (30 MHz to 1 GHz, which is where most auto signals live). For higher frequencies (like 1 GHz to 18 GHz, which is for 5G or industrial radar), we use a reverberation chamber, which is a bigger shielded room with metal stirrers that bounce waves around to simulate real-world conditions. We don’t mess with free-space testing (just antennae out in the open) because our shop is in a busy industrial park—we’d pick up cell phone signals, radio towers, even our own welding machines, and that would throw off the numbers. Trust me, the extra $500 for a TEM cell is worth it to avoid messy, inaccurate data.
Now the actual testing steps—this is where it gets real, and where people cut corners. Let’s start with pre-test calibration. Before you plug in your harness, you need to run a “baseline test” without anything in the TEM cell. That tells you how much external EMI is already getting through the cell walls, and you can subtract that from your final number. If you skip this, your SE will be 10-20 dB lower than it actually is, and you’ll be chasing phantom problems for weeks. We do a baseline test every single day before we start testing batches—even if we just used the cell 2 hours ago. Temperatures in the shop change, and that can mess with the cell’s metal, so consistency is key.
Once baseline is done, you prep your harness for testing. For a typical door harness, we strip back a 10cm section of the outer shielding at both ends, because that’s where we’ll connect our signal and receive antennae. If it’s a braided shield (the most common for wire harnesses, since it’s flexible and cheap), make sure the braid is evenly folded over the end of the harness and secured with a crimped metal ferrule—no loose threads sticking out. Those loose threads are tiny gaps, and they’ll let EMI leak right through. We use a continuity tester to double-check that the entire shielding is connected to the ferrule before we put it in the cell—if there’s a break in the braid, the test is useless.
Then you plug in the antennae. The TEM cell has two ports: one for the signal generator, which pumps in a known, controlled EMI signal at a specific frequency, and one for the spectrum analyzer, which measures how much of that signal comes out the other side. The SE formula is simple: SE (dB) = Signal In (dB) minus Signal Out (dB). So if you pump in a 100 MHz signal at 50 dB, and only 10 dB comes out, your SE is 40 dB. Easy, right? Wait, no—you have to test across all the frequencies your client cares about. If they need SE from 30 MHz to 1 GHz, you test every 10 MHz, not just one spot. A shield might do great at 100 MHz but terrible at 500 MHz, and if you only test one frequency, you’ll miss it. That’s another mistake we made early on—we tested at the middle of the range, and had a whole batch fail because the SE dropped off at the high end. Now we test at 20% intervals across the full frequency range, and log every single number.
Wait, what about common-mode vs. differential-mode signals? Oh, that’s a big one for wire harnesses. A lot of people forget that EMI can come from both inside the harness (differential mode, like data signals) and outside (common mode, which is the interference that wraps around the entire harness). Most TEM cells test common mode, which is what we care about for external shielding, but if your harness has high-speed data lines, you might also need to test differential mode with a different setup. We always ask clients which signals are most critical—if it’s a brake control harness, we test heavily at the low end (30-100 MHz), and if it’s an ADAS camera harness, we bump up the frequency range to 1 GHz because of the high-speed video signals. That’s the stuff that can make or break a client’s project.
Now, what about different shielding types? Because not all harness shields are the same, and testing needs to adjust for that. Braided copper shields are the workhorse, but they have tiny gaps between the copper threads. Foil shields are more continuous, but they’re stiff and can crack if the harness is bent a lot. Conductive polymer shields are newer, lighter, but more expensive. When we test a braided shield, we always test it in both a straight position and a slightly bent position, because braided shields can have lower SE when curved. We had a client last year who specified a braided shield for a steering column harness, and when we tested it straight, it hit 55 dB, but when bent 90 degrees (like it would be in a car), it dropped to 42 dB. We switched them to a hybrid braid+foil shield, and it held steady at 58 dB even when bent. That’s the kind of thing that makes clients trust you—you don’t just test in perfect lab conditions, you test like the harness will actually be used.
Post-test data is just as important as the test itself. We log every single test run: date, harness part number, batch number, frequency range, SE numbers, even who ran the test and the temperature in the shop. If a client calls us 6 months later with a problem, we can pull up their original test log and see exactly what their harness was rated for. We also run a second verification test on every 10th harness in a batch, just to make sure the testing equipment didn’t drift. Last quarter, our spectrum analyzer was off by 5 dB, and we caught it during a verification test before we shipped 500 harnesses. That saved us a huge headache and a lot of money.
Now, what if you don’t have access to a fancy TEM cell or lab gear? A lot of small wire harness shops can’t drop $20k on a test setup, so what do you do? We’ve helped some small startup clients with a cheaper method: the “parallel plate test.” You get two large metal plates (like 30cm x 30cm), mount them parallel to each other, put your harness between them, use a signal generator and a small antenna. It’s not as accurate as a TEM cell, but it works for mid-range frequencies and gives you a rough idea of SE. Just don’t use that for critical auto or aerospace parts—save it for prototyping.
The biggest mistake we see, though, is cutting corners on the physical harness before testing. I can’t stress this enough: a harness that’s crimped wrong, has a gap in the shielding, or has loose wires will give you terrible SE, no matter how good the material is. We have a 10-point pre-test checklist that every tech signs off on before a harness even goes near the tester: check all crimps with a microscope, check for gaps in shielding, make sure the connector shells are properly grounded, no stripped wires, no sharp bends in the shield. That step alone has cut our failed test batches by 75% in the last 5 years.
At the end of the day, measuring SE for wire harnesses isn’t about being perfect—it’s about being consistent. If you follow the same steps every time, test the right frequencies, account for how the harness is actually used, and don’t cut corners on preparation, you’ll get numbers that mean something. We’ve built our whole business on this: we don’t just ship a harness and cross our fingers that it works. We test every critical batch, share the full SE report with clients, and if they have questions, we walk them through the numbers. If you’re tired of dealing with harnesses that fail EMC tests, or you need a supplier that actually knows how to back up their shielding claims with data, hit us up. We’re here to help you get the SE right the first time, no guesswork.
Stamping Parts REFERENCES
- MIL-STD-285, Shielding Effectiveness of Enclosures Against Electromagnetic Radiation, Department of Defense, 2018.
- IEC 62153-4-1, Metallic Materials for Telecommunication and Other Electronic Cable and Wire – Test Methods for Electromagnetic Shielding – Part 4-1: Shielding Effectiveness at Transverse Electromagnetic Wave (TEM) Cell, International Electrotechnical Commission, 2020.
- White, J., “Practical Testing of Shielded Wire Harnesses,” Wire Journal International, Vol. 47, No. 3, 2022, pp. 45-50.
JDE Automotive Technology Co., Ltd.
JDE Automotive Technology Co., Ltd. is one of the most professional wire harness manufacturers and suppliers in China. As we have world-leading production equipment and strong manufacturing capabilities, we warmly welcome you to buy customized wire harness from our factory. Contact us for more discount information.
Address: No.38 Jiufei’e Road, Dalingshan Town, Dongguan City, China
E-mail: Kellyzhong@jdeauto.com
WebSite: https://www.jdecontact.com/
