A headlamp is a promise. It has to keep water out for ten years or more, through heat, cold, vibration, and pressure washing — and it has to keep doing it after the car has been built, shipped, sold, and driven. When a lamp leaks, the customer does not see a welding defect. They see condensation inside a lamp, a failed inspection, and a brand they trust a little less.
That is why automotive lighting is the most demanding application in hot plate welding, and the most honest test of a machine builder. Anyone can weld a lamp housing once in a laboratory. Doing it ten thousand times, on a production line, with a leak test after every part, is a different job entirely.
I have spent twenty years building hot plate welding machines for automotive parts, including lamp housings, and this article is the advice I give to lighting engineers and their procurement colleagues before they buy: what makes a lamp leak, what the leak test is really measuring, and how to set up the process so the test passes the first time — not after three rounds of rework.
Why a lamp is harder to weld than it looks
A lamp housing is not a simple shape. It is a large, thin-walled, often curved assembly, usually in PP or ABS or a blend, with a long, continuous weld path that follows the contour of the lens. The joint has to be airtight along its whole length. A leak the width of a hair, anywhere on that path, is a failure.
Three things make that hard. First, the weld path is long, so the hot plate has to heat evenly from one end to the other — including the corners, where heat is lost fastest. Second, the housing is thin and flexible, so it moves when it is heated and when it cools, and any movement during cooling can open the joint. Third, the lens and the housing are often different materials or different grades, and they do not melt at the same rate. The process has to bring both to a bonding condition at the same moment.
None of this is a reason to avoid hot plate welding for lamps. Hot plate welding is, in fact, the right process for most lamp housings: it handles large weld planes and complex contours, it produces a strong, sealed joint, and it does so without vibration or the part-marking that other processes can leave. But it is the right process only when the machine, the tooling, and the joint are designed for the lamp — not adapted from a generic machine.
What the leak test is really measuring
Most lighting customers test with air: the lamp is pressurised, and the pressure drop over a set time must stay inside a limit. Some use a water bath and watch for bubbles. Either way, the test is not measuring “weld strength.” It is measuring whether there is a continuous path for air or water through the joint. A weld can be mechanically strong and still leak, because strength is about the average of the joint and a leak is about the weakest point.
That distinction changes how you should think about the process. You do not need a weld that is strong on average. You need a weld that is complete everywhere, including the places that are hardest to heat and hardest to hold. The corners. The ends of the weld path. The spot furthest from the fixture clamps. When a lamp fails the leak test, mark the exact position of the leak and keep a record. In my experience, leaks that always appear in the same place are a tooling or heating problem. Leaks that appear in different places are a process-stability problem. The fix is different, and guessing wrong costs weeks.
The five decisions that decide the leak test
Before a single parameter is set, five decisions have already decided whether the lamp will pass. Get these right and the process window is wide. Get them wrong and no parameter will save you.
1. The joint design. The joint is the weld. A well-designed lamp joint gives the molten plastic somewhere to go, keeps the two halves aligned during closing, and provides enough melt area for a continuous seal. A joint designed for appearance, or copied from another process, will fight you for the life of the programme. Bring the welding engineer into the joint design early — before the mould is cut, not after the first leak.
2. The hot plate and its heating zones. A lamp tool is large, and a large tool does not heat evenly by itself. The centre stays hot; the corners cool. That is why our servo machines divide the hot plate into eight heating zones, each with its own set temperature, actual temperature, and alarm band. If the corner zone drifts cold, the machine tells you before the leak tester does. If your supplier offers a single thermostat for a lamp-sized tool, ask them how they know the corners are at temperature. “We checked it once” is not an answer that survives production.
3. The fixture. The fixture has to hold a flexible, thin-walled housing in exactly the right shape while it is heated, welded, and cooled. That means supporting the part close to the joint, not just at the far edges, and controlling its position through the whole cycle. A good fixture is designed for that lamp. A fixture adapted from another lamp will hold the part approximately — and approximately is where leaks live.
4. The closing sequence. The time between the hot plate leaving and the tools closing is the most underrated number in lamp welding. The melt layer cools in air, fast. A slow or inconsistent close means the two surfaces meet after the outer layer has skinned, and the bond is weak at exactly the moment it needs to be complete. Servo-driven closing — programmable speed, programmable position, the same every cycle — exists for this reason. On a machine where the closing speed depends on factory air pressure, the leak rate will change with the time of day.
5. Holding through cooling. The joint must be held under controlled force while it cools and the plastic shrinks. Release the tools early to save two seconds of cycle time and the joint can open microscopically as it shrinks. Two seconds saved, one lamp scrapped. It is the worst trade in the factory, and it is made every day under production pressure. Write the holding time into the process sheet, and do not let anyone shorten it without a new validation.
How we validate a lamp before the machine ships
A lamp programme should never discover its leak problem at the customer’s factory. That is what the factory acceptance test is for — and it has to be a real test, not a demonstration.
Bring production parts, including parts at the edge of your tolerance, not hand-picked samples. Run a batch large enough to see variation, not five pieces. Leak-test every part, and record the leak position on every failure. Ask the builder to show you the live position and temperature screens during the run, and to demonstrate what happens when a heating zone is forced out of its alarm band. Check the safety systems in the same visit: the light curtain must stop the machine when interrupted, the safety door must close automatically, and the emergency stop must halt all movement immediately. The operator who will run this machine is standing in front of a hot plate; those systems are not paperwork.
Also confirm the handover documents: the operation manual, electrical and pneumatic drawings, spare parts list, and PLC backup. A lamp line that stops for a missing drawing is a lamp line that misses its delivery. Our manuals include the maintenance schedule for the machine — pneumatic components checked weekly, hoses replaced within their maximum service life even when they look fine — because a lamp machine that is not maintained will not stay leak-tight.
The mistakes I see most often
After twenty years, the mistakes repeat. Here are the four I see most, so you can avoid them.
Buying on machine price and paying in scrap. A lamp machine that is cheaper by ten percent and leaks three percent of lamps is not cheaper. Count the scrap, the retest, the rework, and the customer visits. Then compare prices.
Validating with perfect parts. Validation parts are usually the best parts the mould ever made. Production parts are the average parts. Validate with average parts, and with the worst parts your tolerance allows, or your validation means nothing.
Changing the material without revalidating. A new grade, a new supplier, a new filler level — any of these changes the melting behaviour, and the old parameters are no longer right. PP, PE, ABS, PMMA, ASA, and PA all weld well by hot plate, but two grades of the “same” plastic can behave very differently. Revalidate when the material changes, even if the name on the bag looks the same.
Treating the leak test as quality control. The leak test is a detector, not a fix. If you are finding leaks at the tester, the process has already failed upstream. The fix is in the joint, the heating, the fixture, or the closing sequence — not in testing harder.
What to ask a machine builder
If you are evaluating a supplier for a lamp programme, these are the questions that separate a builder from a reseller.
- Show me a lamp you have welded, and the leak-test record for a production batch, not a single sample.
- How many heating zones does the hot plate have, and can I see each zone’s actual temperature during a run?
- How is the closing speed controlled, and what happens to it when the factory air pressure drops? (On our servo machines, the critical movements are electric, so an air dip does not change the weld. The standard air supply is 6 bar, running the auxiliary functions.)
- How does the fixture support this specific lamp near the joint?
- What is the positioning accuracy report for a real machine, not a brochure figure?
- What happens at the factory acceptance test if a lamp leaks — do we stop and fix it, or do we sign and ship?
- What documents, spare parts, and training come with the machine, and how fast is the first response when we need help? (For reference: Jfortune responds within 12 hours on business days, and supports customers in Germany, Hungary, Mexico, and India.)
A builder who answers these questions directly, with evidence, is a builder who has done this before. A builder who changes the subject to price has told you everything you need to know.
The bottom line
A lamp that passes the leak test the first time is not lucky. It is the result of five decisions made before production starts: a joint designed for hot plate welding, a hot plate with zone-by-zone control, a fixture built for that lamp, a fast and repeatable closing sequence, and holding force maintained through cooling. Around those five, you need a machine that can prove, cycle after cycle, that it did the same thing — with live positions, live temperatures, and alarms that speak before the leak tester does.
Hot plate welding has earned its place in automotive lighting because it seals large, complex, thin-walled assemblies that other processes struggle with. It keeps that place only when the machine is built and validated for the lamp in front of it. Buy the repeatability, validate with real parts, and the leak test becomes what it should be: a confirmation, not a discovery.
Frequently asked questions
Can hot plate welding produce leak-tight automotive lamps?
Yes — it is one of the main processes used for lamp housings, because it handles large weld planes and complex contours and produces a strong, sealed joint. Leak-tightness depends on joint design, even heating across the whole weld path (including corners), a fixture that supports the part during cooling, and a repeatable closing and holding sequence.
Why do headlamps fail the leak test after hot plate welding?
The usual causes are an unmelted spot (often at a corner or the end of the weld path), a joint that opened during cooling because holding force was released too early, or trapped air from a closing speed that was too fast. Mark the exact leak position: leaks in the same place point to tooling or heating; random leaks point to process variation.
Which plastics can be hot plate welded for automotive lighting?
Common weldable thermoplastics include PP, PE, ABS, PMMA, ASA, PA, and compatible blends. Final suitability depends on the material grade, fillers, joint design, and appearance requirements — and any change of grade or supplier should be revalidated.
How do I choose a hot plate welding machine for lamp housings?
Look for servo-driven motion with programmable position and speed, zone-by-zone hot plate temperature control with alarms, a fixture designed for your specific lamp, and a factory acceptance test run with your own production parts and a leak test on every piece. Ask for a positioning accuracy report from a real machine.
Keep reading
- Hot Plate Welding Machine for Automotive Lighting
- Servo Motor Hot Plate Welding Machine
- Comprehensive Guide to Hot Plate Welding Quality Control
About Jfortune
Suzhou Jfortune Precision Machinery Co., Ltd. (Jfortune) designs and builds custom hot plate welding machines — servo, hydraulic, pneumatic, horizontal, and pallet systems — for automotive, appliance, and industrial plastic parts. Our core engineering team has worked in plastic welding for more than 18 years, and around 70% of our team works in design, project management, engineering, and technical support. We serve customers including Mercedes-Benz, BMW, Audi, Jaguar, Volkswagen, and General Motors, with CE-ready machines and service support in Germany, Hungary, Mexico, and India. Standard hot plate welding machines are typically delivered in 30–40 days; tooling in 40–45 days.
Next step: Send us your lamp drawing, material, and leak-test requirement. Our engineers will review the joint design and recommend the machine platform and validation plan before you commit to a quotation. Contact us at info@jfortunemachine.com or visit jfortunemachine.com and hotplateweldingmachine.com.