How to correctly place a thermocouple and get the best measure
Thermal testing is the definitive “trial by fire” for any electrical product seeking certification. It serves as one of the most critical safety tests. It confirms that the insulation systems inside your device will not degrade or fail catastrophically over time. However, a thermal test is only as reliable as the sensors you employ. Thermocouple placement Guide will help you avoid passing a dangerous product or failing a perfectly safe one.
This guide simplifies the complex requirements of IEC OD-5012. This document governs thermocouple use in the IECEE CB Scheme. The guide provides a practical roadmap for gluing, taping, and measuring. It helps you avoid being burned by the regulations.
What to measure?
In my years working across different test labs, I’ve often seen technicians place thermocouples with surgical precision, only to realize they weren’t sure what the actual “passing” grade was. It is one thing to know how to stick a sensor to a PCB. However, understanding which components are the most critical is an entirely different challenge. Understanding what the IEC 60335-1 standard for example, actually demands from them is also challenging.
Temperature rise testing is not a “one size fits all” process where you aim for a single number. Instead, the standard provides a roadmap of limits based on the material type, the function of the component, and how likely a user is to touch it. If you are preparing your device for certification, you need to know exactly which parts of your Bill of Materials (BOM) are the thermal “weak links.”
Understanding Thermal tests: What to Monitor
When you start instrumenting a product for an IEC 60335-1 test, your first instinct might be to cover every single resistor and capacitor. But and this is a common mistake, over-instrumenting can actually interfere with the airflow of the device. You should focus your efforts on components that provide “protective” functions or those where the insulation system is the only thing standing between the user and a 230V shock.
Typical critical components to measure
The list below outlines the primary areas where you should focus your thermocouple placement to ensure a robust compliance report:
- Internal Wiring and Cable Insulation: Any wire that passes near a heat source (like a heating element or a high-power resistor) must be monitored to ensure its PVC or XLPE jacket doesn’t melt.
- Capacitors and Power Semiconductors: X and Y capacitors, along with MOSFETs and diodes in the power stage, are sensitive to thermal runaway and must be kept within their rated operating temperatures.
- Plastic Enclosures and Support Materials: The “non-metallic” parts that support live connections are critical because if they soften, live parts could move and bridge creepage distances.
- User-Accessible Surfaces: Knobs, handles, and the outer casing itself must be monitored to prevent “burn hazards” to the end user during normal operation.
- Every winding: and i mean all of the windings. Inductors, motors, relays, every single winding can create a fire hazard and must be measured.
- Protection devices: every safety device must me measured. If your product have a bimetal switch, an overload protector or any other type of hardware protection, it must me monitored to certify that is working in its normal working conditions.
Always double check:
The list above is a starting point, remember that the aim of the test is to check that the most unfavorable conditions and hot points are moritored and verified.
Always double check, prior to starting the official test, where the hottest points in the product and in the specific components sits.
Focusing on these specific zones allows you to create a thermal map that mirrors the safety priorities of a certification auditor. By concentrating on these critical nodes, you ensure that the most vulnerable safety barriers are the most scrutinized during the trial.
How to measure temperatures?
Choosing the Right Weapon: Wire Type and Size
Before you even touch the product, for the correct thermocouple placement, selecting the correct is type is crucial. While various types exist (J, K, T, E), Type K (Chromel-Alumel) and Type T (Copper-Constantan) remain the undisputed workhorses of compliance testing.
Ideally, you must select a thermocouple wire diameter that strikes a perfect balance between durability and thermal accuracy. According to IEC OD-5012, the “heat sink effect” is a primary source of error.
- Thin is in: For compliance testing (IEC 60335,IEC 62368-1, etc.), use fine-gauge wire. The standard recommendation is 0.2 mm to 0.3 mm (approx. 32 to 28 AWG). This allow a more precise thermocouple placement.
- The Physics: Thick wires act as thermal drains. If you attach a heavy-gauge wire to a small component (like a surface-mount diode), the wire itself draws heat away from the junction. This results in a reading that is artificially low. Read more about thermocouples placement in this article.
Tech Tip: Watch out for color codes! In the ANSI standard (USA), the Red wire is Negative. This is the opposite of almost every other DC electrical standard. In IEC color codes, the Negative is usually White. Always double-check your polarity before starting a 4-hour test.
Thermocouple placement: Glue vs. Tape
The IEC standards generally require that the thermocouple be in “good thermal contact.” It should cause a “negligible effect” on the temperature of the part. This requirement fuels the eternal lab debate: Glue or Tape?
While tape is convenient for quick engineering scans, adhesive (gluing) is the preferred method for final compliance testing under IEC OD-5012 for several specific reasons:
- Reliability: Tape relaxes under heat, leading to air gaps and inconsistent readings.
- Thermal Mass: A microscopic drop of adhesive has significantly less thermal mass than multiple layers of Kapton or glass tape.
- Conductivity: Using specialized adhesives ensures the bead is in intimate contact with the component surface.
The “Superglue” Method
For most plastic and metal surfaces, a cyanoacrylate adhesive combined with an activator spray is standard industry practice.
The flow to follow
- Clean: Wipe the surface with isopropyl alcohol to remove oils.
- Place: Position the thermocouple bead firmly against the surface.
- Glue: Apply a tiny drop of adhesive (cyanoacrylate).
- Freeze: Spray the activator to cure it instantly.
- Relieve Strain: Secure the wire with tape a few centimeters away from the bead. This is the only correct use of tape. Preventing the weight of the wire from pulling the bead off the hot spot.
The “Winding” Trap: When Thermocouples Are Forbidden
This is where many newcomers fail their audit. If you are testing a motor, a transformer, or a solenoid, do not rely solely on a thermocouple. It might be attached to the outside of the winding.
Change of Resistance Method
IEC standards (such as IEC 60335-1 Clause 11) usually mandate the Change of Resistance Method for windings.
- The Physics: The hottest part of a winding is deep inside the copper layers, inaccessible to a surface probe. A surface thermocouple will inevitably measure significantly lower than the internal “hot spot,” potentially hiding a failure.
- The Rule: You must measure the cold resistance and cold temperature before the test. After the test, you measure the hot resistance and calculate the internal temperature rise using the standard formula for copper:
Where is the temperature coefficient of, in this example, copper (234.5).
Note: Only use a thermocouple on a winding if the standard explicitly allows it. This is usually permitted for very small coils. Use it if the winding is non-uniform and the resistance method is physically impossible. Even then, most standards impose a penalty (e.g., reducing the allowed temperature limit by 10°C) to account for the measurement error.
Troubleshooting Thermocouple Placement Guide FAQ
Even with the best placement guide in hand, the reality of the test lab often throws curveballs. A component might be too small, a voltage might be too high, or the data might just look “weird.” For many years, I tested products ranging from LED drivers to industrial machinery. I’ve collected a list of the most common questions technicians ask when setting up a temperature rise test.
I feel your pain, but be careful. While Kapton (polyimide) tape is excellent for electrical insulation, it is notoriously unreliable for mechanical fixation at high temperatures. As the device heats up, the adhesive softens, and the wire can lift the bead off the surface. That tiny air gap acts as an insulator, and your reading could drop by 5°C to 10°C instantly.
The Verdict: Use tape only for temporary scans during R&D. For the final certification test (the one that goes into the report), use the cyanoacrylate (glue) method. It is the only way to guarantee the “good thermal contact” required by IEC standards.
This is a critical safety issue. If you glue a thermocouple to the heatsink of a primary switching transistor (400V DC or mains potential), that voltage travels to your data logger.
Check your equipment: Ensure your Data Acquisition Unit (DAQ) has channel-to-channel isolation rated for the voltage you are testing. Standard handheld thermometers often do not have this protection.
The “Kapton Sandwich”: If your logger isn’t isolated, place a single layer of thin Kapton tape between the component and the thermocouple bead.
Warning: This adds thermal resistance. You must verify if this approach is acceptable for your specific standard.
This usually points to electrical interference (EMC) or a ground loop.
Ground Loops: If you weld your own thermocouples, the bead is conductive. If you attach it to a grounded metal chassis and your data logger is grounded, you’ve created a ground loop. Isolate the bead or use an ungrounded (battery-powered) logger.
EMC: If you are testing a high-frequency switching power supply, the thermocouple wires act as antennas. Twist the thermocouple wire pairs tightly all the way to the logger to cancel out interference. Use ferrite beads on the thermocouple wires near the logger inputs.
Generally, yes, but be strategic. You need to get the wires out without altering the airflow or thermal properties of the device.
Drill Small: Drill a hole just big enough for the bundle of wires.
Seal It: After routing the wires, seal the hole with duct seal putty or silicone. If you leave it open, you are creating an artificial “chimney” that lets heat escape. This will make your test result invalid (and likely too optimistic).
This is common with modern SMD (Surface Mount Device) components. If the wire is larger than the component, it acts as a massive heat sink. In these cases, you might need to use Type T thermocouples with even finer gauge wire (0.1 mm), or consider non-contact methods like IR Cameras if the standard allows it. Most IEC safety standards still prefer thermocouples for final verdicts, but IR is great for finding the hotspot first.
Correction Check: Always verify if your specific product standard (e.g., IEC 60335-2-xx) has specific clauses for “micro-components.”
Do not just hang a wire in the air!
Air conditioning cycles can cause the ambient to swing by ±3°C every few minutes, making your Delta T calculation useless.
The Fix: Improve thermal stability of the room as much as possible.
Conclusion
Compliance testing is fundamentally about consistency and worst-case verification. Whether you are testing a toaster or a server rack, the goal is to capture the thermal reality of the device. It’s important to assess it under stress. Use thin wires (0.2–0.3 mm), glue them tightly, relieve the strain with tape. Never trust a surface probe on a transformer winding unless absolutely necessary.
