Inquiring Winemaker


Wine From Sludge: Lees Filtration

March 2010
by Tim Patterson
Recovering good wine from gooey lees seems like such a good idea. The sludge at the bottom of all those tanks and barrels is mostly made up of the same wine (or juice) that got racked off, and it could be put to better use than compost, couldn’t it? After all, a lot of time and work and money went into that stuff, so surely…

In the real world, the reputation of lees recovery—especially of fermented wine—isn’t so great, and it certainly isn’t glamorous. Lees filtration is mainly the province of the larger, more industrial wine operations, which are able to extract large volumes of liquid from large amounts of lees and turn it into generic, low-end wine, or wine for the bulk market, or wine-like products destined for toaster tarts and salad dressings. Lees filtration may have a bigger fan base among corporate bean counters than winemakers.

It’s a dirty job, but a lot of wineries have to do it. Several service providers in California keep busy shuttling mobile lines hither and yon at harvest time. And new technology coming on the market promises better quality and greater suitability for at least mid-sized producers. Chances are good that in an increasingly competitive worldwide wine bazaar, lees filtration is going to become a bigger concern for more and more wineries.

Sludge management methods
There are of course lees and lees: red and white, clean and funky, freshly minted and beyond-shelf-life stale. The most important distinction is between sweet or juice lees—settled from juice, mostly whites, during clarification and before fermentation—and dry or post-fermentation lees. Not surprisingly, wineries of different sizes and aiming for different market segments handle the various types of lees in several distinct ways.

Method No. 1 is the use of gravity, the self-cleaning approach to lees removal. This method is in fact pretty effective, but only if you have the time to wait for nature to do the job. In many production situations, one tankful of clarified juice or fermented wine has to be moved out to make room for another, well before liquids and solids have had a chance to find their separate levels. Smaller wineries will sometimes take lees from a post-fermentation settling or from a first barrel racking and barrel them down once more, getting a further portion of the wine back. German winemakers may do a white wine fermentation in a large puncheon and top it off when the fermentation is completed, retaining all the lees, then wait six months before racking and bottling—at which point the lees are a very solid cake, thanks simply to gravity.

Next step up is plate and frame filtration. In an era of membranes and crossflow, plate and frame filtration may seem a bit old-fashioned, but in the context of lees filtration, this method has some definite advantages. Plate and frame filtration can be done on a very small scale, reclaiming wine or juice from 50 or 100 gallons of lees pudding—maybe enough to stretch to another barrel, enough to matter to a very small operation. And it can be done using equipment the winery probably already has.

Perhaps more importantly, plate and frame technology runs less risk of oxidation than the rotary drum vacuum systems (up next) widely used in the industry. Minimizing oxidation can have a significant impact on resulting wine quality.

Rotary drum vacuum technology is the standard for large wineries and for the major mobile services. This approach utilizes a stainless steel drum with a screen on the outside and a partial vacuum on the inside; the drum is coated with a layer of diatomaceous earth (DE) that provides the filter medium. The drum is positioned in a trough through which the juice or wine flows, and the pressure differential pulls the liquid through the DE, trapping the solids. As the outside surface of the DE layer accumulates lees and starts to clog, it is continually shaved off, making a fresh surface available.

This method has several advantages, starting with relatively fast throughput. Juan Puentes, service manager for Napa-based Winetech, says its mobile equipment can handle up to 1,000 gallons of lees per hour. Since lees volume is normally somewhere between 5% and 10% of wine volume, that computes to cleaning up the residue from 10,000 to 20,000 gallons of wine. Settling that out by the gravity method would take a very, very long time. The rotary vacuum also can handle very sludgy lees without clogging, thanks to the automatic scrape-and-refresh mechanism that renews the coating of DE. Puentes’ motto is, “If I can pump it, I can filter it.”

There are potential disadvantages, however. The DE medium can have a subtle but perhaps noticeable effect on wine flavor, which may require further corrective action. Some exposure to air is inevitable, resulting in elevated levels of dissolved oxygen. There are various ways to minimize this, but some level of exposure is built into the technology.

The rotary vacuum also makes sense only for larger volumes of wine or juice. Winetech’s Puentes says its minimum mobile job is 400 gallons of lees; Bryan Tudhope of VA Filtration, which also provides mobile lees filtration, puts the minimum at 350 gallons. Translating this back into wine volume, this means the lees from 7,000 to 8,000 gallons—something like 3,000 cases. Since most of the bonded wineries in the United States produce far fewer than 3,000 total cases—let alone 3,000 cases of a single wine—this technology isn’t for everybody.

The fourth method, just now becoming ready for prime time, is the use of crossflow membrane filtration for lees filtration (more on that in a moment).

Winery strategies
I talked with the winemakers at a pair of medium-sized wineries who have experience with relatively large- and relatively small-batch production, to see where and how lees filtration fits into their programs. Larry Gomez at Lockwood in Monterey County, Calif., makes 80,000 cases per year under the Lockwood label, plus custom work for another three or four clients every year. In his spare time, Gomez makes much smaller volumes of wine for his family’s Via Vega label.

At Wente Vineyards in Livermore, Calif., Karl Wente is involved with the winery’s overall production—more than 300,000 cases per year—and also supervises its upscale “winery within a winery” label, Nth Degree.

Gomez emphasizes that every vintage is different, every lot of w ine is different, and so are the factors that go into deciding what to do with a given batch of lees. In short vintages, the value of recovered juice and wine goes up, and Gomez finds particularly that some of his custom clients will be fairly insistent on meeting a volume target.

The freshness of the lees in question is a critical issue: The longer lees sit before processing, the lower the quality of the result. And some grape varieties and wine styles lend themselves to using lees-recovered juice and wine more than others: New Zealand-style Sauvignon Blanc, he thinks, would not be a good candidate, given its emphasis on clean, angular flavors; some big hearty reds might work better. Fat, creamy style Chardonnay would be somewhere in the middle.

All these considerations add up to Gomez’ general orientation: “Ask me now how useful lees recovery is, and ask me three years from now, and ask me three years ago, and you’ll probably get three different answers.” Lockwood employs plate and frame technology for its lees filtration, and although a lot of juice and wine gets filtered, not much of it makes its way to the Lockwood label. “We usually crush more than we bottle,” he says, “so we can pick from what we’ve fermented and send the rest to the bulk market.”

And as for the use of lees filtration for the Via Vega label, Gomez just chuckles. “We do 11 grape varieties on 15 acres; there’s no point in worrying about the lees.”

Wente Vineyards uses rotary vacuum technology and DE for white juice lees, and by treating the lees right after a short period of post-press settling, it’s able to send the recovered juice back into the stream for fermentation. Post-fermentation white lees are plate-and-frame filtered and generally sent out to the bulk market.

Fermented red lees go through the rotary vacuum, and Karl Wente thinks the modest amount of oxygen exposure can actually be helpful here, reinvigorating the recovered wine and dampening reductive tendencies. Still, it is destined for the bulk market, either for generic “red table wine” or other purposes.

Wente thinks that if a winery had the capacity to process red lees right after pressing and a brief period of settling, the wine might be more useful and valuable. And he thinks that the prospect of crossflow lees filtration could have real potential—it’s just that a crossflow system hasn’t made it to the top of the budgetary wish list yet.

Enter crossflow
Until now, crossflow filtration hasn’t been part of the lees filtration options list. The technology has demonstrated good results in several other wine filtration applications, from simple clarification to removal of volatile acidity. The limiting factor has been fouling: Unlike the rotary vacuum setups, membrane systems tended to clog with an onslaught of lees.

Koch Membrane Systems (KMS) has announced a new twist on crossflow membrane technology that can handle sludgy lees and potentially deliver recovered juice and wine of higher quality and value. The Koch equipment utilizes a self-cleaning pre-filter to remove the matter found in some batches of lees. For the membranes themselves, Koch took its lead from technology developed for the fruit juice industry, which deals with large amounts of solids and demands high levels of recovery. The result is new, 10-foot-long cartridges filled with one-half- and 1-inch diameter membrane tubes, rather than the hollow fibers used in standard wine crossflow configurations.

Market manager David Akin says the new systems increase recovery from 85%-90% of the potential wine to 99%, and they deliver better quality recovered wine. Since the crossflow/membrane equipment is a closed loop, there is no oxygen exposure, nor is there need for DE. The combination, he says, makes the recovered wine more valuable, whether it is blended in-house or bulked out. For a winery that generates lots of lees, an increase in wine value of, say, a dollar per gallon can add up to a lot of dollars.

Akin says KMS is working on ways to adapt its new system for smaller scale wineries and for mobile service use. He suggests that a winery considering purchase of a $40,000-$50,000 crossflow system for other purposes (which starts to make some economic sense at about a 20,000-case production) should consider spending another $12,000-$15,000 to add lees filtration capacity to the mix.

Advances in lees filtration methods should continue to make the quality of recovered wine better and make the effort of doing lees filtration economically attractive to additional medium-sized producers. On the other hand, lees salvaging is never going to be part of the program for $200 cult Cabernets, or for 1,500-case wineries bottling five different vineyard-designate labels of old vine Zinfandel. For those folks, a certain portion of their wine is just going to go un-bottled and un-consumed.

Tim Patterson writes about wine and makes his own in Berkeley, Calif. Years of experience as a journalist, combined with a contrarian streak, make him interested in getting to the bottom of wine stories, casting a critical eye on conventional wisdom in the process. Contact him through


Currently no comments posted for this article.