Keg purging with active fermentation

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk:

DuncB

Well-Known Member
Joined
Jun 5, 2019
Messages
2,203
Reaction score
1,237
Location
Paremata New Zealand
I do purge the starsan into a bucket after the fermenting keg has purged it's own headspace. I do this starsan into a spare container.
Yes dilution works but I'm trying to limit the oxygen as much as possible so I have 4 litres of oxygen less to dilute. I suppose the keg being flushed doesn't have to be in the keg ferment ( chest freezer).
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
Lifetime Supporter
Joined
May 14, 2014
Messages
12,941
Reaction score
9,562
Location
Renton
... It seems that the common process is to just connect the kegs at time of pitching yeast...but then all the air in the headspace of the fermenter will be the first to fill the "purged" keg. Then from there you have to rely on some dilution to bring the oxygen levels to an acceptable level.
The analysis in post # 3 assumes that both the fermenter headspace and keg to be purged start out with an air atmosphere. If you start with StarSan in the keg, the final O2 content will be even lower than if you start with air.

And yes, it is all about dilution, since the CO2 flow rate is not enough to safely assume that we get a "sweeping" effect.

Brew on :mug:
 

CascadesBrewer

Supporting Member
HBT Supporter
Joined
Mar 24, 2013
Messages
2,031
Reaction score
1,830
Location
VA, USA
And yes, it is all about dilution, since the CO2 flow rate is not enough to safely assume that we get a "sweeping" effect.

I tried to sort through the math and see if it applies to my setup, but could not. I see there was a response earlier saying that fermenting 5 gals was enough to purge out two 5-gal kegs + 1.5 gals of headspace. So I was hoping that my 2.6 gals of 1.053 wort would be enough to purge a 5-gal keg plus the 2.7 gals of headspace.

Any idea if the spunding valve has a positive or negative impact on the mixing of CO2 and air?

Nope. Due to the design of the lid, when liquid comes out of the PRV, there is a 3 fl oz air bubble trapped in the lid. I personally measured this (as have others.)

I wonder if this varies from keg to keg. When I look at my ball-lock kegs, the rim of the lid is higher than the posts. When I fill my kegs with StarSan I do so through the open lid, and I cannot see where there would be any pockets of air. If I filled the keg through the liquid post, I also don't see where there would be any sizeable pockets if liquid was coming out of the PRV on the lid. Maybe if liquid was coming out of the gas post. Or am I not understanding what is being said here.
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
Lifetime Supporter
Joined
May 14, 2014
Messages
12,941
Reaction score
9,562
Location
Renton
I tried to sort through the math and see if it applies to my setup, but could not. I see there was a response earlier saying that fermenting 5 gals was enough to purge out two 5-gal kegs + 1.5 gals of headspace. So I was hoping that my 2.6 gals of 1.053 wort would be enough to purge a 5-gal keg plus the 2.7 gals of headspace.

Any idea if the spunding valve has a positive or negative impact on the mixing of CO2 and air?



I wonder if this varies from keg to keg. When I look at my ball-lock kegs, the rim of the lid is higher than the posts. When I fill my kegs with StarSan I do so through the open lid, and I cannot see where there would be any pockets of air. If I filled the keg through the liquid post, I also don't see where there would be any sizeable pockets if liquid was coming out of the PRV on the lid. Maybe if liquid was coming out of the gas post. Or am I not understanding what is being said here.
Look closely at the lid. If you unscrew the PRV and turn the lid upside down, you can fill the lid cavity with water until it runs out the PRV opening. Dump the water into a measuring cup - it should be about 3 fl oz.

As far as the effect of a spunding valve, I'd have to rerun the analysis at a higher pressure to see what really happens. A lot of what goes on is counter intuitive.

Brew on :mug:
 

DuncB

Well-Known Member
Joined
Jun 5, 2019
Messages
2,203
Reaction score
1,237
Location
Paremata New Zealand
I have never looked at corney keg so not sure about them. Sankey I fill to the brim, back fill the dip tube via the post and then put it on it overflows a bit as lid attached. Then purged with the ferment gas. I vent the gas from fermenter and thru the jumper tube for a while before the keg purging.
There is a maths way of working out how much theoretical gas is made for the sugar to alcohol reaction. But friend has a plaato and reckons that a 6 gallon ferment makes about 2000 litres of CO2. This could be right having looked at the bubble rate thru an airlock, assuming each bubble about 1 ml theres a lot of bubbles per hour and for days.

Spunding valve won't have any effect on the gas mixing except in the valve.
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
Lifetime Supporter
Joined
May 14, 2014
Messages
12,941
Reaction score
9,562
Location
Renton
I have never looked at corney keg so not sure about them. Sankey I fill to the brim, back fill the dip tube via the post and then put it on it overflows a bit as lid attached. Then purged with the ferment gas. I vent the gas from fermenter and thru the jumper tube for a while before the keg purging.
There is a maths way of working out how much theoretical gas is made for the sugar to alcohol reaction. But friend has a plaato and reckons that a 6 gallon ferment makes about 2000 litres of CO2. This could be right having looked at the bubble rate thru an airlock, assuming each bubble about 1 ml theres a lot of bubbles per hour and for days.

Spunding valve won't have any effect on the gas mixing except in the valve.
2000 litres is too high by about a factor of 4. In the analysis in post #3 of this thread, I go thru the math on CO2 generation. The net was about 440 litres from a 20 litre ferment.

Brew on :mug:
 

renstyle

Supporting Member
HBT Supporter
Joined
Aug 20, 2020
Messages
350
Reaction score
378
Location
Central Iowa
@doug293cz, I nominate entry #3 of this thread for "sticky" status, or something of similar stature.

Recently I cleaned, then sanitized a serving keg, then used the sanitized water during a brew day (dark mild).

After I seated the lid on the (now empty) serving keg with a burst of initial tank CO2, I used fermentation gas to (more or less) fully purge my serving keg of non-CO2 gases down to the very-low-PPB-I-am-OK-with ranges. The calculations were fantastic especially since they generally low-balled the amount of CO2 produced.

Appreciate the effort. Thanks doug, and everybody who played a part in the conversation.
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
Lifetime Supporter
Joined
May 14, 2014
Messages
12,941
Reaction score
9,562
Location
Renton
@doug293cz, I nominate entry #3 of this thread for "sticky" status, or something of similar stature.

Recently I cleaned, then sanitized a serving keg, then used the sanitized water during a brew day (dark mild).

After I seated the lid on the (now empty) serving keg with a burst of initial tank CO2, I used fermentation gas to (more or less) fully purge my serving keg of non-CO2 gases down to the very-low-PPB-I-am-OK-with ranges. The calculations were fantastic especially since they generally low-balled the amount of CO2 produced.

Appreciate the effort. Thanks doug, and everybody who played a part in the conversation.
Now stickied.

Brew on :mug:
 

Spivey24

Well-Known Member
Joined
Sep 2, 2020
Messages
165
Reaction score
112
Now stickied.

Brew on :mug:

Ha, I was looking at your final calculations, and was convinced this had to be an iterative calculation, and not a power to (^) calculation. I was dead wrong. :) the software developer in me made me.

I wrote a small program to calculate the decreasing O2 volume and the final O2 concentration and it matched your calculation exactly. I used your assumptions of perfect mixing rather than trying to factor in the diffusion front. I did end up changing the starting parameter to 400L of CO2 produced to be more conservative. But still quite surprised at how low that final concentration is. Thanks for your work.

O2 concentration levels | C# Online Compiler | .NET Fiddle
 

doug293cz

BIABer, Beer Math Nerd, ePanel Designer, Pilot
Staff member
Mod
Lifetime Supporter
Joined
May 14, 2014
Messages
12,941
Reaction score
9,562
Location
Renton
Ha, I was looking at your final calculations, and was convinced this had to be an iterative calculation, and not a power to (^) calculation. I was dead wrong. :) the software developer in me made me.

I wrote a small program to calculate the decreasing O2 volume and the final O2 concentration and it matched your calculation exactly. I used your assumptions of perfect mixing rather than trying to factor in the diffusion front. I did end up changing the starting parameter to 400L of CO2 produced to be more conservative. But still quite surprised at how low that final concentration is. Thanks for your work.

O2 concentration levels | C# Online Compiler | .NET Fiddle
Thanks for checking my math. I don't know if anyone has ever done it before, but no one every pointed out an error.

Brew on :mug:
 

Adam Zerwick

Well-Known Member
Joined
Apr 13, 2018
Messages
51
Reaction score
14
Here's a spreadsheet I made based on the post by @doug293cz. I decided to explore different purging schemes for my smaller batch sizes as well as using a separate vessel for dry hopping. Calculations are made for the following schemes:

- Purge headspace using only fermentation CO2. Assumes entire headspace is not filled with water/starsan.
- Purge unfilled headspace with fermentation CO2, then purge water/starsan filled headspace with fermentation CO2. Gives you the gravity when to start the water purge given a target O2 ppb
- Purge unfilled/filled headspace using fermentation CO2

Hopefully it helps somebody else. It certainly opened my eyes to potentially how much O2 was left in my kegs using my normal purge process. Also one thing to note, the example in post #59 seems to off by a factor of 1000 based on my math... so please let me know if any of you spot a problem with my spreadsheet!

Keg Purge Final O2
 
Top