What can go wrong

The core

A core that tested healthy can, very rarely, not survive another heat cycle. The silicon, its substrate and the thousands of microscopic joints underneath all see the same temperature swing. The risk is small. It is never zero, and we will not pretend otherwise.

The memory

Your original modules make the same journey as the core, and each one carries the same small reflow risk. Every module on the finished board is tested before the card ships.

Gradual heating and cooling

Thermal shock comes from heating and cooling too fast. We stretch the preheat and cool-down phases as long as the job allows, so the board and the silicon move through temperature gradually in both directions instead of being snapped across it.

Live temperature readings

Thermocouples sit at the board and the component during every cycle, so the profile follows the real temperature of your silicon rather than the machine's setpoint.

Lower-temperature solder

We seat your core and memory with leaded solder, which melts at 183°C against the 217°C and up of lead-free alloys, so the most delicate step runs roughly thirty degrees cooler. The alloy is also more ductile, absorbing mechanical stress that would crack a more brittle joint. The same give helps during the rework, in transit on the way back to you, and against the slow strain of a heavy card sagging unsupported in its slot.

Dry before heat

Boards and components are baked dry before any reflow. Trapped moisture turning to steam inside a package is a classic rework failure, which is exactly why every board and component is baked first.

Tested before and after

Nothing proceeds on a card that fails pre-build validation. Every finished build then runs the full test protocol across a 48-hour window, which is where early-life failures tend to show themselves, before the card ships.