In this question defining cold crashing, there is some discussion of ramping the temperature down slowly over a long time, but then says quicker cold crashing is possible if the beer has already completed fermentation.

In this question, I'd like to explore an appropriate, quick, and effective cold crash technique on an ale that's had plenty of time to ferment and "clean-up". And for this example, the purpose of cold crashing is primarily for settling.

Given the carboy is in a refrigerator which is turned on normally (no external temperature control), it will take some time for the heat to be removed from the volume of beer. If the beer started at, say, 68F, with the refrigerator "on", there would be a ramp down, and within 6 hours, the beer would be at, say, 40F. Then you'd wait Y hours as settling continued.

What range of Y would you expect 50% of the eventual settling to be done? What range of Z would you expect 90% of the settling to be done?

Note that in this diagram, I have assumed that settling happens faster earlier, and slower later. This may or may not be your experience (could be linear). The blue line is the temperature of the beer. The black line is the amount of settling over time, starting at no cold crash settling and ending when 90% of the eventual possible settling is done.

temperature and setting fraction graph

  • 3
    I don't think there is enough evidence to directly answer the question you're asking with any precision. Gravity and composition of the beer would play a huge part in the sedimentation rate. The # of variables is probably too big for it to simply scale to temp vs. time.
    – brewchez
    Jun 3, 2019 at 12:11
  • 2
    I will say that the completeness fo the ferment has nothing to do with the reason for slower temp drops. If the temp is dropped too quickly the yeast tend to release esters and other flavor compounds due to the shock of the temp drop. Slowing the temp drop prevents the shock. This happens regardless of whether final gravity has been achieved or the mythical clean-up step has occurred post ferment.
    – brewchez
    Jun 3, 2019 at 12:11
  • 1
    I like the thought process though so I upvoted the question.
    – brewchez
    Jun 3, 2019 at 12:12

2 Answers 2


For cold crashing we go from tank temp to 4C in about a day, that is due to the limitations of our cooling plates and chillers. If we could drop a tank from 18-20C to 1C in 12H I would take that. ~1.5C/H gets the yeast to go to sleep and drop out.

Anything below 4C and even using a lager yeast it will be effectively stopped, and starts sedimenting out.

Y and Z are going to depend on a number of factors such as flocculation rate of your yeast, are you at FG or crashing 1 or 2 points above it so you don't have to prime back for packaging, how deep is your FV, a 10HL 3.5m FV takes a bit longer to drop clear than a 50l/10 gal FV, but likewise a tall 50l vs a squat 25l will have different values.

But, for arguments sake in a 50l FV at FG, with a standard ale yeast you are crashing down to 4C in a day, I would expect for most brews Y <1day, and Z to be 2 days.

2 days in a small FV at <=4C will drop out most of your yeast, heck most should be out already if you are at FG.


Since this question involves estimation and I haven't actually run the proposed experiment myself or recorded specific data, I'll take my guess based on what I think I might know based on 20 years homebrewing experience. To be "clear", what I personally would be aiming for is not to settle out obvious chunks of yeast and hops, but rather the finest sediment that changes the beer from a slight haze to the point of brilliant clarity. As I bottle my beer but do not have an enormous refrigerator and thus can only refrigerate a few bottles at a time, I experience this after a bottle has been moved from the basement at about 60 F (16 C) to the refrigerator at about 40 F (4 C) for a period of weeks. Brilliant clarity is a rarity for me except when the bottle has been in the refrigerator for a long time. So, my estimates for 90% brilliance vs. 50% "clear":

Y is approximately 1 week?

Z is approximately 3 weeks?

Of course, this is just an estimate. Anyone interested should run an experiment to support or refute.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.