by Clark Smith

All her pretty dreams argon.-Bruce Springsteen

Oxygen is not the enemy of wine. Yet the most outspoken proponents of O2's role in wine development will still scrupulously try to exclude it from partial tank headspaces. We all gotta gas. But in reality, few of us do it well. And in an imperfect world, it is not enough to shrug and say, AWe just try to keep topped tanks.@

Doing it right is a tricky business. To begin with, the two popular gases, nitrogen and carbon dioxide, both have insidious difficulties which make them terrible choices for excluding air. Vinovation receives dozens of bulk wine shipments weekly from our 500 clients, a large cross-section of California wineries. Dissolved oxygen readings on incoming wines indicate that even the most savvy wineries have not managed to establish proper inert gas procedures. The predictable response to our phone call: ABut we gassed the heck out of it!@ typically turns out to mean fifteen or twenty minutes of nitrogen from a high pressure cylinder through 1/4 line at 40 psi, or CO2 until freeze-up.

Although trucking is an easy time for procedures to slip, I believe that many winemakers have simply not thought the problem through. As a result, proliferation of myth and copying of poor procedures have resulted in a state full of wineries that do not use inert gas wisely. Fallacies include undersized delivery systems, failure to measure volumes (with a flowmeter) and results (with an O2 meter), and unenlightened cost analysis. In an effort to reduce D.O. post-shipment, several large wineries routinely nitrogen-sparge all incoming wines upon receipt! This practice is guaranteed to strip flavor, and treats the symptom rather than attacking the real problem on the shipping end.

THE MYTH: Nitrogen is the Cadillac of inert gases.

THE FACTS: Nitrogen is air without oxygen. As such, it is slightly lighter than air, and has no blanketing effect. If x = number of headspace volumes delivered, the fraction air = 1/ex. This means that after one volume, the headspace (and the wine beneath it) will have = 1/e, or 40% saturation; after two volumes, 1/e2 - 16%; for three volumes, 1/e3 = 5%; and so forth. Correct headspace gassing with nitrogen requires a large effort of will: huge supplies, delivery systems geared for high volumes, trained personnel, and headspace measurement instrumentation.

THE MYTH: Carbon dioxide blankets wine.

THE FACTS: Although substantially heavier than air (44 vs 29 MW), the turbulence with which this gas is introduced into a headspace through a 1/4" line results in substantially mixing. Dry ice works much better, but sets up a worse problem: headspace CO2 dissolves rapidly into wine, imploding the tank unless a vacuum relief valve (almost always) invisibly allows air to be sucked in.

THE MYTH: Pressure regulators can meter gas volume. Copied from a prestigious Napa winery=s cellar procedure: AHeadspaces should be blanketed with 12" of inert gas, which is discharged at the wine surface, 20 minutes at 15 psi.

THE FACTS: Temperature, hose length, hose diameter, regulator size, and a host of other factors contribute to the relationship between pressure at the regulator and gas volume delivered. In practice, there is no useful direct relationship. Flow meters measure flow.

THE MYTH: Argon is expensive.

THE FACTS: Argon by volume costs three times more than nitrogen by volume. But it is so much more effective than in actual use it can do a better job for less money. In many cases, a blanket that stays put will do the job better than several complete headspace volumes of nitrogen. Argon is the cheapest and most effective means for most inert gassing. In truth, gas cost is chump change. At 6.5 cents a cubic foot, an entire 1000 gallon headspace full of argon costs nine bucks. A barrel of argon costs fifty cents. Figuring which gas to use in a specific application will usually blow the savings as brain labor. Vinovation has nothing but argon on premise, because it=s never the wrong choice.


The key to economical gas usage is laminar delivery. If argon exits its delivery hose at low velocity, it can layer under air rather than mixing. A perfect blanket would register 100% air at the top of the vessel until one volume is introduced, then drop to zero. At this delivery velocity, a 2" line with a tee at the exit is required for acceptable laminarity. A 1 2" hose with tee is adequate at 200 SCFH but not for 500 SCFH.

From these investigations I calculate Smith's Rule of Thumb for argon delivery: keep delivery velocity below 3 feet per second. To get gas velocity, just divide cubic feet per second by hose cross-sectional area in square feet. Table 1 shows maximum volumes for various hose sizes.

The high pressure cylinder with the 1/4" regulator and hose which one can find at every winery in the state is completely inadequate to gas the headspace of a tank or truck. Table 1 shows why. This set-up can only deliver 30 gal/hr of blanketing gas.

Most wineries need to deliver at least 1,000 gal/hr, which takes a 1 2" line, and more likely a 2" line with tee is preferred. This kind of flow will require an oversize 3/8" regulator, discharging into flowmeter recalibrated for argon.

Smith=s Rule of Thumb of Gas Delivery states that acceptable laminar delivery to a tank occurs at under 1 meter per second of velocity. Table 1 shows how much headspace you can gas per hour with a given hose diameter. [Gas velocity (ft/sec) = SCFH/3600 sec/hr) %r2 for hose radius expressed in feet.]


In the 1980's, a lot of progressive wineries tried to measure D.O. in their wines. The instrument most of them bought was a YSI bench model with a stand-mounted agitator to provide turbulent medium to the electrode. This equipment is a piece of junk. If you own one of these ancient relics, please throw it out! We now know that the agitation function is not important, and that it is much more critical to take the reading, whether in the wine or the headspace, in the tank rather than in the lab. At Vinovation we use a portable Orion Model 830, which for under a thousand dollars, gives reliable headspace and dissolved O2's at the tank. It's a robust instrument, but it does break when dropped from serious height. Cost of doing business. Once headspace gassing procedures are well established, the need to measure and corresponding danger to the instrument may diminish.


  1. Set up to stock the volumes of gas you really need.

  2. Set up a laminar delivery system.

  3. Measure gas flow with a gas flowmeter.

  4. Verify headspace O2 with a headspace O2 meter.


If you don't want to refit your set-up, figure one fresh 220 SCF nitrogen cylinder, (about 1600 gallons of gas) blasted in through a 1/4" line, for about 500 gallons of headspace. Switching to argon will give about the same results in about 1,000 gallons. A twenty-foot length of 2" hose with a tee at the end will get you near 1500 gallons, and will allow you to start blanketing for real.

Table 1: Maximum laminar delivery rate @ 3 ft/sec

Hose diam.

Cross section (sq.in.)

Gal. per Min.


















1 2"





1 2" w/tee










2" w/tee










3" w/tee






Gas Fundamentals

Gas is used by volume, be it molar volume or Standard Cubic Feet (measured at 1 atm and 250C. If we buy gas by the pound, we can=t compare use of different gases until we divide each gas by its Molecular Weight (44 for CO2, 40 for argon, 29 for air, or 28 for N2) or its specific gravity (1.0 for air, 0.97 for N2, 1.39 for Ar, 1.53 for CO2). A mole of any of these gases at standard temperature and pressure occupies one molar volume, which works out to about 0.8 SCF or 24.4 liters or 6.4 gallons.

The inert gas law, PV = nRT, describes how gases behave. Pressure x volume is in fixed proportion to moles x absolute temperature. This means:

gas molecules take up the same space regardless of how much they weigh.

An outside tank headspace vacillating between 150C and 350C, or 288 K and 208 K, will pump 7% of its headspace in and out every day.

An inside tank at constant temperature will pump 3% of its volume when a storm passes and the barometer swings from 30 to 29 inches Hg.  


Previous Page 

< Winemaking Fundamentals



Home  |  WineQuality Enhancement  |   The State-of-the-Art Estate Winery  |  Services at our Facility  |   Winemaking Fundamentals  |  CostSavings  |   Vineyard Enology  |  International Services  |   About Us  |  Contact Us  |   International Links

Vinovation, Inc
ph:(707) 824-7900  •  fax:(707) 824-7905  •  email: info@vinovation.com
office: 652 Petaluma Ave, Sebastopol, CA 95472
facility: 1365 Gravenstein Hwy. South, Sebastopol, CA 95472