Estimate PCB temperature in two stages: the air temperature around the board, then the additional temperature rise inside each critical component. A sealed enclosure can have acceptable average air temperature while its converter runs too hot. Start with actual electrical losses, not the power supply’s nameplate rating. Then check the converter, capacitors, connectors and other temperature-sensitive parts against their own operating limits.
Separate board, air and junction temperatures
A PCB does not have one temperature. Copper beneath a switching regulator can be much hotter than an unused board corner. Air near the enclosure roof may also be warmer than air entering the converter’s immediate surroundings.
Keep these temperatures separate:
- Outside ambient: the air surrounding the enclosure at its installed location.
- Local internal air: the air immediately around the component being assessed.
- Board temperature: the temperature at a defined copper or laminate location.
- Case temperature: the temperature at a specified point on a component package.
- Junction temperature: the semiconductor temperature inside the package.
Compare each measurement with the limit defined for that same location. A package surface measurement cannot be compared directly with a junction limit without an appropriate thermal relationship.
Use recommended operating limits and temperature-dependent derating curves for acceptance. An absolute maximum junction temperature is a damage boundary, not a normal operating target. Thermal shutdown is protection against an abnormal condition, not evidence that continuous operation is acceptable.
Capacitor life, connector current capacity and converter output capability can deteriorate before any semiconductor reaches thermal shutdown. The control cabinet temperature limits article explains how to identify the component that sets the system limit.
Count converter losses and downstream heat
A supply marked 12 V, 5 A can provide up to 60 W at those ratings. That does not mean a connected PCB consumes or dissipates 60 W. Determine the actual current under the operating condition you need to assess.
For a converter delivering 3.3 V at 2 A:
Output power = 3.3 V × 2 A = 6.6 W
If its efficiency is 88% at that input voltage, load and temperature:
Input power = 6.6 W / 0.88 = 7.5 W
Converter loss = 7.5 W − 6.6 W = 0.9 W
The corresponding input current at 12 V is approximately 0.625 A. Use efficiency data at the actual operating point, rather than a headline peak efficiency.
The enclosure heat load depends on where the 6.6 W output goes:
| Energy destination | Heat to include inside the enclosure |
|---|---|
| Converter feeds electronics entirely inside the enclosure | Approximately 7.5 W, if nearly all supplied energy becomes internal heat |
| Converter feeds a load entirely outside the enclosure | Approximately 0.9 W, plus losses in internal wiring and other internal devices |
| Converter feeds both internal and external loads | Converter loss plus the portion dissipated by internal loads |
Do not count the 6.6 W internal load and then add the full 7.5 W converter input. That counts the downstream power twice.
Also separate converter loss from chip loss. The calculated 0.9 W may be shared among the switching IC, inductor, rectifier and other components. Applying the entire converter loss to the IC’s thermal resistance can give a misleading junction estimate.
The PCB heat in a sealed enclosure article covers enclosure-wide heat accounting. For component assessment, preserve the loss breakdown by device wherever the design information allows it.
Estimate air rise and junction rise separately
For an initial steady-state estimate, use the enclosure’s effective thermal resistance under conditions representative of the installation:
Internal air rise = total internal heat × enclosure thermal resistance
For illustration, assume the completed installation has an effective enclosure thermal resistance of 4 °C/W. With 7.5 W dissipated internally, the estimated average internal air rise is 30 °C. At 40 °C outside ambient, that gives approximately 70 °C average internal air.
The 4 °C/W value is an assumed example, not a rating for a catalog enclosure. Obtain the actual value from applicable engineering data or a representative steady-state test. It depends on enclosure dimensions, mounting, surface conditions, airflow outside and heat distribution inside.
Next, assess the component above its local air temperature:
Junction temperature ≈ local air temperature + component loss × applicable junction-to-air thermal resistance
Suppose a validated assembly model gives 40 °C/W from the converter IC junction to local air. If that IC dissipates 0.5 W, its estimated junction rise is 20 °C. With 70 °C local air, the estimated junction temperature becomes 90 °C.
Those numbers illustrate the calculation only. Local air may differ from the enclosure average, and the IC loss must come from an appropriate loss estimate.
Do not substitute a generic data-sheet junction-to-ambient value without checking its reference board and airflow conditions. Copper area, thermal vias, layer count and nearby heat sources change the result. Do not apply junction-to-case resistance to a case-to-air temperature difference.
A deliberate conduction path needs its own resistance calculation. The heat sink thermal resistance article explains the separate interfaces between a component and its eventual heat-rejection surface.
Understand what ratings and data sheets establish
An ingress rating does not establish allowable PCB dissipation or predict semiconductor temperature. It describes protection under specified environmental test conditions, not the cooling performance of your completed circuit.
Our stainless wall-mount enclosures are rated NEMA 4X / IP66 by the manufacturer; not UL listed. That rating does not provide a thermal resistance for your installed assembly.
Likewise, a converter’s current rating does not guarantee continuous operation at that current in every enclosure. Its usable output depends on input voltage, switching conditions, PCB layout, local temperature and the manufacturer’s derating requirements.
Read the temperature labels carefully. An operating ambient range, recommended junction range and absolute maximum junction rating answer different questions. A data-sheet efficiency curve may also represent typical conditions rather than a guaranteed worst-case loss.
For outdoor equipment, include solar heating separately from electrical losses. A shaded workshop test does not represent an enclosure installed in direct sun. The outdoor enclosure heat-gain article explains how to distinguish internal dissipation from external heating.
Measure the temperatures that decide acceptance
Use a representative enclosure, mounting arrangement, cable routing and closed-door configuration. An open enclosure changes the heat paths and can conceal the temperature rise you need to assess.
Install insulated measurement leads with power isolated. Prevent probe wires from contacting terminals, obstructing clearances or holding the door gasket open. Conduct energized measurements under the site’s electrical safety procedure.
Record these quantities together:
- Input voltage and current at the converter’s input boundary.
- Output voltage and current at the converter’s output boundary.
- Outside ambient, shielded from direct sun and nearby hot exhaust.
- Local air beside the converter, without touching the board or enclosure wall.
- Converter package and board temperatures at documented locations.
- Temperatures of nearby capacitors, inductors and heavily loaded connectors.
For a switching converter, input and output measurements need suitable instruments and consistent operating conditions. Converter loss is the difference between two larger powers, so measurement uncertainty can be significant compared with the loss itself.
A thermal camera helps locate hot spots, but shiny metal and small packages can produce misleading readings. Confirm critical locations using a suitable contact method or the component manufacturer’s documented temperature-estimation method.
Test the combinations that could produce the highest temperature. These can include maximum sustained load, input-voltage extremes, simultaneous loads and the highest installation ambient. Maximum output current is not automatically the worst efficiency point.
Continue until relevant temperatures stabilize. A brief successful run only establishes short-duration behavior. Use the sealed-enclosure thermal soak test to structure the steady-state verification and record the operating conditions.
Change the limiting heat path
Choose the correction from the measured temperature pattern, rather than adding cooling hardware indiscriminately.
If local air and most components are too hot, reduce total losses or improve heat rejection from the enclosure. A larger enclosure may help, but its installed thermal performance still needs verification.
If local air is acceptable but the converter package is hot, address the component-to-board and board-to-surroundings paths. Follow the device manufacturer’s layout guidance for copper, thermal vias and exposed-pad attachment. A higher-current converter is not automatically cooler; compare losses at your actual operating point.
If one board region is hot while another remains cool, improve spacing or move temperature-sensitive components away from the heat source. The component layout article covers those placement decisions.
An internal fan can reduce local gradients, but it does not remove heat from a sealed enclosure by itself. A thermal pad to the enclosure can help only when its contact area, compression, electrical insulation and complete conduction path are properly engineered.
After any change, repeat the closed-enclosure test. Record the highest temperature, the applicable operating limit and the remaining margin. Include measurement uncertainty and any installation conditions the test did not fully reproduce.
Specify the thermal basis with the enclosure
Choose enclosure dimensions from both the component layout and the required thermal performance. The stainless enclosure size finder filters outside and nominal inside dimensions, but dimensions alone do not establish temperature acceptance.
Put the following information in the enclosure specification:
| Specification item | Information to provide |
|---|---|
| Installation conditions | Maximum ambient, indoor or outdoor location, solar exposure and nearby heat sources |
| Electrical heat load | Converter losses, downstream internal dissipation and simultaneous operating loads |
| Component limits | Recommended temperature limits and required derating for critical devices |
| Mechanical arrangement | Board positions, mounting distances, cable routes and available conduction surfaces |
| Thermal interfaces | Contact locations, insulation requirements and any engineered heat-transfer interfaces |
| Acceptance evidence | Sensor locations, steady-state test conditions and required temperature margin |
State whether each temperature estimate is calculated, measured or supplied by the component manufacturer. Keep those distinctions on the drawing and test record. That gives enclosure sizing and component selection a traceable basis instead of relying on an ingress rating or a successful open-bench run.
Enclosures for this job
Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.
| Enclosure | Material | Outside H x W x D / display | Rating |
|---|---|---|---|
| 16 × 12 × 8 in | 304 stainless | 16 x 12 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 24 × 20 × 8 in | 304 stainless | 24 x 20 x 8 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 30 × 24 × 10 in | 304 stainless | 30 x 24 x 10 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
| 12 × 10 × 6 in | 304 stainless | 12 x 10 x 6 in | Rated NEMA 4X / IP66 by the manufacturer; not UL listed |
Questions
Does a 12 V, 5 A supply put 60 W of heat into the enclosure?
No. The supply rating describes available output capacity, not the connected load’s actual consumption. Measure input voltage and current under the relevant operating condition. A converter delivering 6.6 W at 88% efficiency takes approximately 7.5 W from its source and dissipates approximately 0.9 W in the conversion stage.
Is acceptable enclosure air temperature enough to protect the PCB?
Not necessarily. Average internal air temperature can hide a hotter local region, and a semiconductor junction operates above its surroundings. Check local air, relevant package or board temperatures, and the component manufacturer’s temperature limits. Use an applicable thermal model to estimate junction temperature when direct measurement is unavailable.
Can I use the converter data sheet’s junction-to-ambient thermal resistance?
Only as an initial estimate when its test conditions are understood. The reference PCB, copper area, airflow and mounting can differ substantially from your assembly. Use manufacturer-supported methods or a validated assembly model, then verify the completed enclosure under representative operating conditions.
Will a higher-current converter always run cooler?
A higher current rating alone does not establish lower temperature. Compare conversion efficiency and component losses at the actual input voltage and output load. Package construction, PCB layout and thermal interfaces also matter. Verify the replacement in the closed enclosure rather than relying on its current rating.
