Here is my repost from a similar story from June 15, 2023:
"As power systems designer, it would cause me great concern to see the number of failures that are happening if this were my design. The typical warranty calls for most products will reflect a small number of actual field failures, as there are probably very many more and people are not reporting them. I have seen this many times over my career.
I took a look at the Molex Mini-Fit connectors that are similar to the actual one being used and found the spec for the 12 pin part. For 600W on a 12VDC system, you will need 8.33A per pin - if they all share perfectly. They won't, and a good designer plans for some redundancy to account for that. For instance, even in the NEC electrical code, they derate wires if operated in parallel. Connectors are done the same way, but with even more derating needed.
Connector vendors also play spec games and will list the maximum current per pin for a single or minimum pin housing for good marketing exposure. If you plan to use more than that 1 or 2 pins for high current, you should always derate depending upon the vendor recommendations - check the specs. This is nearly universal on power connectors and where many design engineers get into trouble when using power connectors.
In the spec chart on the Molex part, they recommend derating to 5.5A per pin with the 12 pin configuration. The derating has many factors including contact pin resistance spread (how well they will share), as well as the thermal profile of the connector housing. The pins toward the center of the housing get the most thermal stress since they get heat from both sides.
There are many design factors external to the connector that effect the reliability. One is the PCB design: did they put agressive thermal reliefs into the copper artwork? This can generate added local heating (just like using too small of a wire into a connector). Thermal reliefs are used to help with soldering and usually look like a wheel spoke pattern. Another factor is the typical humidity in the operating environment, which effects the speed of metal oxidation for the contact metals. If the connector environment also has high vibration, then the contacts can undergo fretting failure from the contacts rubbing which slowly eats the outer plating and exposes the metal underneath (usually copper). Depending upon the contact and housing designs, this might not need much vibration to dramatically accelerate this. Think about cooling fans, water pumps, and other rotating devices in a computer case.
On my past designs, I have performed thermal tests on critical power connectors like this to check the inside pins are not getting uncomfortably hot. Before the crimped pin is inserted into the housing, I would attach a thermocouple as close as I could to the contact area on the crimp pin. And choose a contact that is by the center of the connector. Bake the system until it reaches thermal steady state. The connector housing should not be anywhere close to the maximum rated temperature.
The 8.33A rating per pin needed in this application needs some inspection. And I don't believe blaming the user is a reasonable action. People are not responsible and criticized for the plug on their hairdryer for instance. It is expected to work, and safely."
Facebook
Twitter
Instagram