News – Separator Shrinkage
Separator Shrinkage: The Silent 20% Error
01 October 2025
In unconventionals, everybody loves to talk about EURs, decline curves, RTA/PTA, yada yada. But there’s a quiet math problem screwing up forecasts (and royalty checks) every single day: separator oil shrinkage.
Here’s the setup:
· Wells in shale basins usually measure oil and gas at separator conditions.
· Those volumes are then (incorrectly) treated as stock-tank volumes.
· Result? Forecasts and allocations look rosier than reality—sometimes by more than 20%.
What the heck is “shrinkage”?
When separator oil is stabilized to stock-tank conditions (1 atm, 60°F), it shrinks. The light ends flash off as gas.
Here’s the setup:
· Wells in shale basins usually measure oil and gas at separator conditions.
· Those volumes are then (incorrectly) treated as stock-tank volumes.
· Result? Forecasts and allocations look rosier than reality—sometimes by more than 20%.
What the heck is “shrinkage”?
When separator oil is stabilized to stock-tank conditions (1 atm, 60°F), it shrinks. The light ends flash off as gas.
- Shrinkage Factor (SF): fraction of separator oil that makes it to the stock tank. Typical range? 0.65–0.99.
- Flash Factor (FF): the gas liberated when that shrinkage happens. Could be 5–1000 scf/STB.
- Separator pressure & temp (higher pressure = more shrinkage, lower temp = more shrinkage).
- Wellstream composition (changes over time as BHP declines).
- Shut-ins (hello “CGR kicks” and transient fluid changes).
Why it matters
Most operators slap on a single shrinkage factor for a well or even a whole basin. Problem:
This isn’t just academic. Oil is worth more than gas. Overstating oil rates → bad decisions on capex, royalties, reserves.
A better way
The paper lays out a three-step fix:
Boom: now you’ve got daily SF and FF. Not some “one-size-fits-all” fudge factor.
Don’t have an EOS model? Fine. Build a field-specific shrinkage correlation using test data. At minimum, use separator pressure, temperature, and liquid API.
Key takeaways
The bottom line: ignoring separator shrinkage is like not counting calories when you’re on a diet—it feels good in the short term, but you’re lying to yourself.
Want to know what your wells are really making? Stop trusting the separator. Start accounting for shrinkage.
Most operators slap on a single shrinkage factor for a well or even a whole basin. Problem:
- That shortcut is wrong.
- In volatile/near-critical fluids, it’s very wrong.
- It skews oil vs. gas split, screws up allocation, and can make a well look 20% better than it is.
This isn’t just academic. Oil is worth more than gas. Overstating oil rates → bad decisions on capex, royalties, reserves.
A better way
The paper lays out a three-step fix:
- Build a field-wide EOS model (tuned to separator shrinkage data).
- Predict daily wellstream compositions (from well tests, GOR, API, etc.).
- Run those daily wellstreams through the actual surface process (#stages, separator p & T)
Boom: now you’ve got daily SF and FF. Not some “one-size-fits-all” fudge factor.
Don’t have an EOS model? Fine. Build a field-specific shrinkage correlation using test data. At minimum, use separator pressure, temperature, and liquid API.
Key takeaways
- Shrinkage changes a lot over time—especially in volatile and near-critical systems.
- Ignore it, and you can be off by >20% in stock-tank oil rates.
- If GOR is swinging, if separator conditions shift, or if you’ve got shut-ins? Update shrinkage daily.
- The lazy industry habit of applying one shrinkage factor for a whole basin? Outdated.
The bottom line: ignoring separator shrinkage is like not counting calories when you’re on a diet—it feels good in the short term, but you’re lying to yourself.
Want to know what your wells are really making? Stop trusting the separator. Start accounting for shrinkage.