Purging Oxygen from a Keg by Pressure Only - the calculations to determine time

Homebrew Talk

Help Support Homebrew Talk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

TheHygienist

Member
Joined
Feb 17, 2018
Messages
23
Reaction score
13
So I don't want residual sanitizer in my corny keg, I just want to purge the oxygen with CO2 at a low psi. Sure I could use fermentation CO2 to do this, but using my CO2 bottle, these are the calculations. You can use 2 formulas. One is from https://www.gigacalculator.com/calculators/pipe-flow-rate-calculator.php whereas you use a pipe pressure to obtain flow rate. Easy enough.

PSI initial is 5 psi. Final PSI is near zero, say 0.00001 PSI because you will vent gases in the the keg into the atmosphere. I chose a line inside diameter of 4 millimeters. Line (pipe) length of 914 mm, or 3 feet. CO2 has a dynamic viscosity of 0.01502 cP (centiPoise).

Plugging in these parameters, we can see that applying 5 PSI to that line, will flow 0.56 cubic feet CO2 per second, or 33.6 cubic feet per minute, cfm.

So, the ambient air has at the most, 23% oxygen. At 10,000 ppm per 1%, that is 230,000 ppm that needs to be purged. Lets say we want to shoot for diluting that oxygen with CO2 down to 0.1 ppm.

Now we need to use a basic purge equation https://en.wikipedia.org/wiki/Dilution_(equation). We need to solve this equation for time ending, where time intial is of course, zero.

The volume of a 5 gallon corny keg is measured by height 23" and radius 4.25", or 1,305 cubic inches or 0.76 cubic feet. A 3 gallon keg is 0.52 ft3.

We now divide ppm by 1,000,000 to get a decimal for the concentrations. C-initial is 0.23 (230,000 / 1,000,000) and C-final is 0.1 / 1,000,000.

The dilution formula is used to solve for time, t:

Screenshot 2024-08-10 193247.png


Notice that the right side of the equation is negative.

So, solving this for time = 20 seconds (or 11 ft3 CO2).

So, applying 5 PSI of CO2 pressure through a 4 mm line into a 5 gallon corny keg while holding the PRV open to vent, should purge the oxygen down to 0.1 ppm, if you let it flow for 20 seconds. If you want to purge a 3 gallon keg, the time is 13.5 seconds.

You can of course, play with the line internal diameter, pressure, keg volume, and desired final concentration of oxygen to determine the purge time for your specific setup. And I'm just considering the maximum concentration of oxygen as 23% as an enriched atmosphere. You can get away with 21% if you like. For my IPA's, I'll want the final ppm to be lower than 0.1 ppm.

I think this saves a lot of trouble with filling kegs with sanitizer and then trying to get that junk out. Of course if you don't want to use 11 ft3 of CO2 or similar volume, simply purge your kegs with the CO2 blowing off your fermenter. To each his own.
 
Last edited by a moderator:
I do this with fermentation blow off. I attach the fermenter to the liquid post of a sanitized keg and the gas post out to a jar with sanitizer for the escape. Then when I cold crash, I remove the blow off and change the fermenter to the gas post and add 10 psi to both fermenter and keg. This keeps sanitizer from sucking back and it can’t suck in any oxygen. I add a spunding valve set at 10psi just in case. This has worked great for me over the last 3 years. I do like your work on this and will use it if I don’t have an empty keg during fermentation. Thanks for sharing!!
 
I could probably engineer a table of >5 PSI pressures, but for that small of a line diameter for a 5 gallon keg, as a rough rule of thumb, CO2 at 5 PSI for 15 seconds will get you down to about 5 ppm oxygen, 20 seconds will get you to 0.1 ppm, 25 seconds to roughly 0.001 ppm and 30 seconds will completely remove the oxygen from the keg. To completely remove the oxygen that's about 17 cubic feet or roughly 2 lbs. of CO2, which for me, costs about $6.00. The benefit of doing a complete flush is questionable, but the cost is so low for 1 or 2 lbs of CO2, it's a minor expense for beers prone to the negative effects of oxidation.
 
I could probably engineer a table of >5 PSI pressures, but for that small of a line diameter for a 5 gallon keg, as a rough rule of thumb, CO2 at 5 PSI for 15 seconds will get you down to about 5 ppm oxygen, 20 seconds will get you to 0.1 ppm, 25 seconds to roughly 0.001 ppm and 30 seconds will completely remove the oxygen from the keg. To completely remove the oxygen that's about 17 cubic feet or roughly 2 lbs. of CO2, which for me, costs about $6.00. The benefit of doing a complete flush is questionable, but the cost is so low for 1 or 2 lbs of CO2, it's a minor expense for beers prone to the negative effects of oxidation.
My gut says there is something off with the flow rates here. I find it hard to believe that running a CO2 tank at 5 PSI for 30 seconds is going to use up 40% of a 5 lb tank.
 
My gut says there is something off with the flow rates here. I find it hard to believe that running a CO2 tank at 5 PSI for 30 seconds is going to use up 40% of a 5 lb tank.
Something does seem off.
The site for the flow rate calculation says that a requirement of the formula is that the fluid not be compressible. Co2 can be easily compressed, so it may not be valid.
 
My gut says there is something off with the flow rates here. I find it hard to believe that running a CO2 tank at 5 PSI for 30 seconds is going to use up 40% of a 5 lb tank.
This is all based on formulas, I'll see what I get in about a 12 days when I try this with my tank. There are tables that convert lbs. CO2 to volume and I think that yes, if the flow rate is correct then I can expect the tank to lose that much weight.
 
Last edited:
I estimate that my sankey kegs purged of sanitiser with later stage closed ferment gas, then inverted to vent the remaining sanitiser out through the prv have less than 20ml of sanitiser remaining.
I've visually checked this as well on some clear sankey connector type kegs.
I'd rather have this sanitiser than 5 ppm of oxygen in the keg. There is no doubt it's a complicated problem but @doug293cz has done the maths very well on this in the past.
 
So I don't want residual sanitizer in my corny keg, I just want to purge the oxygen with CO2 at a low psi. Sure I could use fermentation CO2 to do this, but using my CO2 bottle, these are the calculations. You can use 2 formulas. One is from https://www.gigacalculator.com/calculators/pipe-flow-rate-calculator.php whereas you use a pipe pressure to obtain flow rate. Easy enough.

PSI initial is 5 psi. Final PSI is near zero, say 0.00001 PSI because you will vent gases in the the keg into the atmosphere. I chose a line inside diameter of 4 millimeters. Line (pipe) length of 914 mm, or 3 feet. CO2 has a dynamic viscosity of 0.01502 cP (centiPoise).

Plugging in these parameters, we can see that applying 5 PSI to that line, will flow 0.56 cubic feet CO2 per second, or 33.6 cubic feet per minute, cfm.

So, the ambient air has at the most, 23% oxygen. At 10,000 ppm per 1%, that is 230,000 ppm that needs to be purged. Lets say we want to shoot for diluting that oxygen with CO2 down to 0.1 ppm.

Now we need to use a basic purge equation https://en.wikipedia.org/wiki/Dilution_(equation). We need to solve this equation for time ending, where time intial is of course, zero.

The volume of a 5 gallon corny keg is measured by height 23" and radius 4.25", or 1,305 cubic inches or 0.76 cubic feet. A 3 gallon keg is 0.52 ft3.

We now divide ppm by 1,000,000 to get a decimal for the concentrations. C-initial is 0.23 (230,000 / 1,000,000) and C-final is 0.1 / 1,000,000.

The dilution formula is used to solve for time, t:

View attachment 855259

Notice that the right side of the equation is negative.

So, solving this for time = 20 seconds (or 11 ft3 CO2).

So, applying 5 PSI of CO2 pressure through a 4 mm line into a 5 gallon corny keg while holding the PRV open to vent, should purge the oxygen down to 0.1 ppm, if you let it flow for 20 seconds. If you want to purge a 3 gallon keg, the time is 13.5 seconds.

You can of course, play with the line internal diameter, pressure, keg volume, and desired final concentration of oxygen to determine the purge time for your specific setup. And I'm just considering the maximum concentration of oxygen as 23% as an enriched atmosphere. You can get away with 21% if you like. For my IPA's, I'll want the final ppm to be lower than 0.1 ppm.

I think this saves a lot of trouble with filling kegs with sanitizer and then trying to get that junk out. Of course if you don't want to use 11 ft3 of CO2 or similar volume, simply purge your kegs with the CO2 blowing off your fermenter. To each his own
Just a question, why the concern about residual sanitizer? Are you flushing the keg with something else after sanitizing? If you are filling with something else, then why wouldn't you just use CO2 to flush that out at the end and be within your required low O2 level? Maybe I am missing something.
 
Just a question, why the concern about residual sanitizer? Are you flushing the keg with something else after sanitizing? If you are filling with something else, then why wouldn't you just use CO2 to flush that out at the end and be within your required low O2 level? Maybe I am missing something.
That issue is kinda exclusive to sankes purely for the dimensions, volumes and reach of the spear...With a corny keg, at the end of the purge of Star San, I turn mine upside downwith the gas still flowing so that that last bit of Star San in the diptube will exit. With a sanke, the bottom of the spear has a greater volume beneath it and since the gas enters the keg at the very top, by inverting the keg you'll get at least a couple ounces of Star San available for removal at the PRV on the coupler...significantly more than a corny. Commercial keg washers with the sanitizing and purge routine leave the keg inverted for the entire process...just SOP with sankes.
 
Just a question, why the concern about residual sanitizer? Are you flushing the keg with something else after sanitizing? If you are filling with something else, then why wouldn't you just use CO2 to flush that out at the end and be within your required low O2 level? Maybe I am missing something.
When using the push sanitizer out method with CO2, I inverted my corny keg and tried to get the rest of the sanitzer out through the PRV. As evident my the amount of fluid left behind, there was quite a bit of it, and used quite of bit of CO2 to try to get it out. The bottom of the PRV port is a raised nipple, that makes it even harder to get sanitizer out. So I thought, why bother. Just sanitize, drain it with the keg upside down, add the hardware, then purge the air using a flowrate and time formula. Simple, easy, and hopefully, effective.

I just finished using that method; applied 5 psi of pressure to the tank and kept the PRV valve open for 30 seconds. Didn't drop the tank guage at all, added another 15 seconds for good measure and that's that. If the purge and flow calculations are correct (I have no instrumentation to make any flow or final O2 concentration measurements) then the keg should have zero oxygen. At this point, if some has the correct instruments, please attempt to confirm that this works.
 
Getting it out thru a cut off gas dip tube works a lot better.
^^^THIS^^^

I’ve started combining StarSan displacement with CO2 (either from fermenter or CO2 bottle) in a modified Corny with a cut-off dip tube on the Gas-In post. I have a gas QD with a short length of 4mm beer line to drain the last few milliliters of sanitizer into the sink by inverting the keg. There’s nothing left in the keg except a few psi volume of gas when I’m done.

After the StarSan is evacuated, I follow up by injecting a small amount of solution (<50ml) of ¼ tsp ascorbic acid and ¼ tsp NaMeta in distilled water. I swirl it around, invert the keg several times a day for a few days to scavenge O2, then drain (under 10 psig CO2) in the same manor as the sanitizer purge. Before filling a keg (under pressure of course) I burp the keg to get rid of residual sulfur.

I don’t have the capability to measure O2 in the empty keg before filling, but intuitively it has to be in the “very few parts per billion” category, hopefully scavenging any DO in the beer or any O2 pickup in the transfer.
 
As a student of these things over the past 5-6 years, I am suspect of the approach. Mainly because most everything O2 & gas related is more difficult than it appears!

1) Opening the PRV is a red flag to me as oxygen will get in if you leave the thing open for 30 seconds.
2) Without a trustworthy DO meter, it is just speculation and hope.
3) Gas mixing is what complicates things. Fermentation purge happens over days with a smaller amount put in at a time. 20-30 seconds just sounds like a lot of O2 & CO2 are going to mix together and some of both will stay behind.

I may be wrong but I would like to see the test done with measurement equipment.
 
When using the push sanitizer out method with CO2, I inverted my corny keg and tried to get the rest of the sanitzer out through the PRV. As evident my the amount of fluid left behind, there was quite a bit of it, and used quite of bit of CO2 to try to get it out. The bottom of the PRV port is a raised nipple, that makes it even harder to get sanitizer out. So I thought, why bother. Just sanitize, drain it with the keg upside down, add the hardware, then purge the air using a flowrate and time formula. Simple, easy, and hopefully, effective.

I just finished using that method; applied 5 psi of pressure to the tank and kept the PRV valve open for 30 seconds. Didn't drop the tank guage at all, added another 15 seconds for good measure and that's that. If the purge and flow calculations are correct (I have no instrumentation to make any flow or final O2 concentration measurements) then the keg should have zero oxygen. At this point, if some has the correct instruments, please attempt to confirm that this works.

Why do you think that the calculations are correct? The website you linked says that the flow rate formula using pressure difference "does not work so well for gases where additional information is required for an accurate computation."

How much CO2 did you use by weight when purging? Weigh the tank before and after. That will tell you how much CO2 went through.
 
Why do you think that the calculations are correct? The website you linked says that the flow rate formula using pressure difference "does not work so well for gases where additional information is required for an accurate computation."

How much CO2 did you use by weight when purging? Weigh the tank before and after. That will tell you how much CO2 went through.
The purge calculation is correct. It's been in use by the Industrial Hygiene profession for decades. The flow calculation was used to ballpark the flow rate. I asked someone out there to confirm. I don't have access to a rotometer. The CO2 weight is irrelevant to me because CO2 is cheap.
 
The CO2 weight is irrelevant to me because CO2 is cheap.
Above you said that a 30 second purge would use up about 2lbs of CO2. A scale would let you know if you are using close to that.

My gut says that 2lbs of CO2 with a slow purge from the bottom would likely get oxygen levels down to an acceptable level. I just highly doubt that 30 seconds as 5 PSI will come close to pushing out that much CO2. I run CO2 for 10-15 seconds just to try an purge a short line of tubing. Plus, the turbulence of a very fast purge will likely cause a lot of mixing of the CO2 and air.

A while back I did some calculations on purging a filled keg. As I recall, it was about $1 worth of CO2 with my 5 lb tank which costs around $30 to swap. At that price, it would cost me $12 to purge with 2 lbs of CO2. With my 15# tank, 2 lbs of CO2 would be around $5, vs maybe $0.25 to push out StarSan solution.
 
Above you said that a 30 second purge would use up about 2lbs of CO2. A scale would let you know if you are using close to that.

My gut says that 2lbs of CO2 with a slow purge from the bottom would likely get oxygen levels down to an acceptable level. I just highly doubt that 30 seconds as 5 PSI will come close to pushing out that much CO2. I run CO2 for 10-15 seconds just to try an purge a short line of tubing. Plus, the turbulence of a very fast purge will likely cause a lot of mixing of the CO2 and air.

A while back I did some calculations on purging a filled keg. As I recall, it was about $1 worth of CO2 with my 5 lb tank which costs around $30 to swap. At that price, it would cost me $12 to purge with 2 lbs of CO2. With my 15# tank, 2 lbs of CO2 would be around $5, vs maybe $0.25 to push out StarSan solution.
The flow rate is the variable in the purge formula and can be directly measured so that is what I will do, I just ordered a cheap rotometer from Amazon. I will check the flow rate directly and not rely totally on a formula. I don't care about the small cost of the CO2. More than likely however, I will use fermentation CO2 to purge the tank most of the time, without purging sanitizer, just to be sure I've displaced all the O2. However, using a CO2 tank is a fall-back method that I would like to use occasionally. A prior post mentioned using a cut-off dip tube as a drainage aid, and using ascorbic acid in the corny keg. Seems that there are some useful comments here for future batches.
 
Which you said you can't measure. OTOH, if you know how much CO2 you used and how much time it took to use it, then you can very easily calculate what the flow rate was and confirm that it matches what you expected.

After I retired after 35 years and a BS and MS in Industrial Hygiene, I don't have my access to my lab. I'll get a rotometer and ballpark the flow rate. You can get one for about $15 on Amazon. Probably will do the job.
 
You put a lot of thought and work into this. It’s certainly impressive.

I’m going to sound like such a dick in this response and I do apologize in advance for it, but this seems like a waste of time for the op.

Hook up the fermentation to your keg to purge or just liquid purge it and call it a day. Most kegs can only leave 30-40 ml in the keg when fully empty. So the diluted sterilizer left in the keg has a ratio of 1 to 625/675 of remaining solution to the beer. If you’re using starSan, this will have nearly zero impact on beer or final ph.
 
For kicks, I set my regulator at 5 PSI and hooked up just short piece of gas line. It does run through a quick disconnect valve, but I am sure that add much less resistance than connecting to a ball valve quick disconnect. I then ran the gas on full for 10 seconds. I measured 0.020 lbs and 0.025 lbs of usage. My scale has a 0.005 lb resolution.

Edit: Quick math says that to run 2 lbs of CO2 into my keg at the rate of 0.025 lbs every 10 seconds, I would need to purge the keg for 13.3 minutes.
 
Last edited:
For kicks, I set my regulator at 5 PSI and hooked up just short piece of gas line. It does run through a quick disconnect valve, but I am sure that add much less resistance than connecting to a ball valve quick disconnect. I then ran the gas on full for 10 seconds. I measured 0.020 lbs and 0.025 lbs of usage. My scale has a 0.005 lb resolution.

Edit: Quick math says that to run 2 lbs of CO2 into my keg at the rate of 0.025 lbs every 10 seconds, I would need to purge the keg for 13.3 minutes.
Did the pressure, as measured by the low pressure gauge on the regulator, drop when you turned on the flow? If so, then the flow rate is being limited by the needle valve in the regulator body, and not the tubing/QD.

Brew on :mug:
 
Getting it out thru a cut off gas dip tube works a lot better.
This is exactly what I do. I have a picnic tap with a gas QD i hook up to the keg and invert so the gas post is at the lowest point. All my gas diptubes are trimmed

Also, I add 3 grams of SMB to my saniclean solution in the keg before purging. This way any traces of residual sanitizer left won't have any DO
 
Also, I add 3 grams of SMB to my saniclean solution in the keg before purging. This way any traces of residual sanitizer left won't have any DO
Do you know where you came up with the 3g amount? I have done this a few time, but I was not sure how much I should add. At RiteBrew, a 1 lb pack of PMB is under $6, so cost is not much of an issue.

With most of my 5 gal kegs, the dip tube angles into a small indentation in the center of the keg. These kegs leave very little solution behind. My 10L Torpedo kegs have more room from the end of the dip tube to the bottom. The dip tubes on the Torpedo kegs are already pretty short, but maybe I should try to trim them up a little more.

My Hazy IPAs seem to stay fresh for 2+ months in the keg (though they seldom last that long).
 
I would be concerned about (dredging my memory for the correct term, could be wrong) flow channeling with this approach. Basically, depending on the geometry of the vessel, purge gas takes a preferential path from input to output, leaving dead volume that mixes with this stream relatively slowly and therefore isn't effectively purged. I would not expect this to be an issue when you're purging with fermentation gas, given the very low flow rate and the long purge time.

Also, I encourage folks to keep an eye on the ball, which for O2 exposure, is the final concentration of dissolved oxygen in the beer, not the keg. It might not sound great to have a concentration of 10 ppm O2 remaining in the keg, but that means there are 270 ug of O2 in the empty (5 gallon) keg. If you dissolved all of that O2 into 5 gallons of beer, you'd have 270 ug O2 in >18.9 kg of beer, which means 14 ug/kg, or 14 ppb. When you consider that filling the keg displaces most of that gas, this is negligible.

You also don't have to worry about the oxygen dissolving in the beer before the empty keg can be filled. Even assuming the oxygen in the keg instantly reaches equilibrium with the beer (it doesn't), at 10 ppm O2 in the gas, that's 0.4 ppb in solution. As long as the gas is displaced on a faster timescale than oxidation of flavor compounds in the beer (which would then allow more oxygen to dissolve), this is not an issue.
 
Do you know where you came up with the 3g amount? I have done this a few time, but I was not sure how much I should add. At RiteBrew, a 1 lb pack of PMB is under $6, so cost is not much of an issue.

With most of my 5 gal kegs, the dip tube angles into a small indentation in the center of the keg. These kegs leave very little solution behind. My 10L Torpedo kegs have more room from the end of the dip tube to the bottom. The dip tubes on the Torpedo kegs are already pretty short, but maybe I should try to trim them up a little more.

My Hazy IPAs seem to stay fresh for 2+ months in the keg (though they seldom last that long).
10ppm SMB eats up 2ppm O2 per my understanding

Bru'n Water has a SMB parameter and if I plug in 5.5 gallons for the volume with 150ppm that yields 3 grams

This would scrub 30ppm of DO which is plenty more than there would ever be in tap water.

If you are concerned about having more residual sanitizer than I do then you should probably cut the dose to 2 grams.

Mix the SMB in water first before adding too
 
Back
Top