No replies to this topic

[Arsalan_1992]

Hello everyone,
I’m currently working on a lean glycol cartridge filter that contains about 98% TEG and 2% water. The client is insisting on a PSV for the fire case, and I’m struggling with some practical aspects of this requirement.
When I model the fire case, the relieving temperature comes out to be around 900°F, which is extremely high compared to the vessel design temperature of 250°F. The set pressure is 1330 psig, while normal operating pressure is about 895 psig.
I do understand from API 521 that the relieving temperature can exceed the vessel design temperature under fire exposure. However, this value still seems unrealistically high, and I believe it’s because I’m forcing vaporization of a very stable, high-boiling liquid (TEG) under high pressure.
This raises a few concerns:
Thermal degradation: TEG starts to degrade well below the calculated relief temperature. So should decomposition or combustion be considered before actual vaporization? If glycol is already degraded at such temperature, then what phase (liquid/vapor/decomposed product) will the PSV actually relieve?
Liquid-filled vessel: The client expects the vessel to remain completely liquid-filled, meaning the relief under fire should primarily be liquid expansion, not vapor generation.
API 521 approach: Can I simply calculate the heat absorbed (Q = C × F × A⁰·⁸²) and use it only to account for thermal expansion of liquid from operating temperature up to the boiling point, without extending into the vaporization zone?
I’m trying to determine a realistic and code-compliant approach for sizing the PSV in such a non-flashing, high-boiling liquid system.