This edition change is not a decimal point. ISO 16750-4:2023 deletes temperature codes A and B outright, consolidates the minimum temperature at -40 °C, and adds four high-temperature classes U, V, W and X reaching all the way to 180 °C. If you are still building a DV outline from the 2006 code table, the problem is not adjusting a few parameters — the whole table has to be re-laid.
1. What moved in the code system
The old edition paired a minimum with a maximum temperature: A started at -20 °C, B at -30 °C, and both capped the high end at 65 °C. That pairing logic was adequate a decade ago, when most electronics sat in the cabin.
Today everything outside the cabin is electronic too — e-drive, domain controllers, on-board chargers, screens. The 2023 edition simply cuts the two low-threshold classes, sets the minimum uniformly at -40 °C, starts the high end at 65 °C and adds four new codes at the top.
| Item | ISO 16750-4:2006 | ISO 16750-4:2023 |
|---|---|---|
| Low-threshold codes A / B (from -20 °C, -30 °C) | Retained | Deleted |
| Minimum temperature | Banded, including -20 °C and -30 °C | Consolidated at -40 °C |
| Upper high-temperature limit | Previous top class reached 160 °C | Raised to 180 °C (code X) |
| New high-temperature codes | None | U / V / W / X (165 °C, 170 °C, 175 °C, 180 °C) |
The table above is a version comparison compiled from public sources; individual limits and applicability are governed by the official standard text.
In one line: the floor tightened and the ceiling opened up. The tightened floor affects low-temperature assessment for every part; the opened ceiling affects only a small group of high-temperature parts — but that small group is precisely where growth has been fastest these two years.
2. A and B are gone, and old designs have nowhere to sit
Classes A and B existed to leave room for parts mounted in the cabin and used in cold regions. With them deleted, -20 °C and -30 °C low-temperature assessment has no landing point in the 2023 edition.
There are three practical consequences.
The first is the test condition. Low-temperature operation and storage previously run to class B (-30 °C to 65 °C) now have to be re-run at -40 °C; dropping the temperature point by 10 °C is a different order of severity for plastic embrittlement, compression set of seals and thermal stress in solder joints.
The second is the acceptance items. Functioning normally at low temperature is not the same as passing; parameter drift, cracking and seal failure still have to be checked afterwards. Change the condition and the acceptance items follow.
The third is the one most easily overlooked: historical reports are no longer comparable. A report issued in 2022 to class B for a given part, used in 2026 to support a new project, no longer matches the new table. The report has not expired — the referenced edition has changed.
3. Who U/V/W/X are for
The four high-temperature codes land in a narrow place: 165 °C to 180 °C, somewhere an ordinary cabin part never reaches.
In practical terms that means motor controllers and the area around the e-drive assembly, parts close to heat sources in the engine compartment, and high-power display modules. Since silicon carbide devices entered production, power density has risen and local heat flux with it, pushing internal housing temperatures well above the past — which is the real reason the standard needed these four classes.
| High-temp code | Maximum operating temperature | What to watch after the test on such parts |
|---|---|---|
| U | 165 °C | Softening and deformation of housing and brackets, connector boot recovery |
| V | 170 °C | Compression set of seals, cracking of potting compound |
| W | 175 °C | Solder joints and bond interfaces, parameter drift in electrolytic capacitors |
| X | 180 °C | Whether the long-term material temperature rating is sufficient, insulation resistance decay |
Note that the "what to watch" column is not an acceptance criterion, it is an experienced view of where risk concentrates. Passing is still judged against the acceptance items written in the technical agreement.
Which leads to a common error in the other direction: treating a higher temperature class as automatically better. Put a cabin display module on class X and the test itself may pass, but material selection is forced up to a 180 °C grade and cost and lead time rise with it, for an in-vehicle environment that never gets close to that temperature. The class follows the mounting position, not a desire for reassurance.
4. GB/T 28046.4 has not caught up, and this is the easiest trap
The current edition of GB/T 28046.4 corresponds to the 2006 edition of ISO 16750-4, so codes A and B are still in it.
That leaves two tables in circulation: domestic projects and customers who recognise only the national standard still work from the 2011 national code table, while joint-venture, foreign-funded and export projects have already re-laid their outlines to the 2023 edition.
The situation itself is not complicated; the trouble is that the same report gets cited by both sides. A supplier submits a report made to the national standard against a project requiring the 2023 edition, the conditions do not line up, and extra testing is the only remedy at review.
The fix is crude but effective: state the standard number and edition year on the first page of the test outline, and repeat it on the request form. When TMC LAB accepts climatic load work, the first step is to confirm which code table the customer is working from, then set the test conditions and acceptance items — done up front, this saves the re-test later.
To compare the clause relationships between the two systems, start with the breakdown on the ISO 16750 standard page and the GB/T 28046 standard page, then check back against the technical agreement.
5. A few honest words
Confirm the edition before discussing conditions. When a customer says "test to 16750", asking whether that is the 2006 or the 2023 edition is far cheaper than rework afterwards.
Set the temperature class from the mounting position, not by copying the last project. The previous vehicle used class C, this one sits next to the e-drive and still uses class C — a very common starting point for trouble. Where measured temperature data exists, check it against the table; where it does not, look the position up and keep margin.
Once the high-temperature class goes up, materials and sealing need re-assessment, not just re-testing. Electronic function may pass while housing deformation, adhesive cracking and seal failure only appear after the test. Environmental testing and thermal stress work such as thermal shock are best scheduled in the same round rather than one after the other.
If you need to re-lay a test matrix to the new temperature classes, or run a condition check against an existing outline, submit the samples and the outline through the environmental reliability testing route and we will go through it clause by clause.
Turnaround time is arranged according to the actual project. You are welcome to state your target date when submitting samples; we will check it against the current schedule, confirm a feasible plan with you, and update you at each milestone.
