To Filter or Not to Filter


Dear Clark:
I own a small winery in Washington State. I produce small lots of Bordeaux wines. My annual production is 500 to 700 cases a year. Merlot, Cabernet Franc, and Cabernet Sauvignon are my main focus. I also produce about 100 cases of Syrah. Due to my small size I have chosen to bottle my own wine rather than hire a mobile bottling line. My question to you concerns filtration of my wine prior to bottling. I originally intended to use a plate and frame filter (1 micron) to remove yeast and other large particles when transferring wine to tank for blending followed by a pes cartridge membrane filter (.45 micron) prior to the filler during bottling. Since researching a number of articles, it is a bit confusing whether or not my original plan would achieve my goal of maintaining high quality stable wine in the bottle. What would your thoughts be?

Jim Myers
Medicine Creek Winery
Olympia WA
Dear Jim:

You’ve picked a thorny issue, one that strongly divides the industry. Many winemakers wouldn’t consider bottling anything without sterile filtration – they consider that the risk of microbial activity in the bottle is too great. This could be of three types: Saccharomyces (i.e. normal wine yeast), malolactic bacteria, or Brettanomyces.

On the other side (my side) are an increasing number of winemakers who find that this type of filtration disrupts wine structure so that texture and aromatic integration suffer.

I wrote about my thinking in Wine Business Monthly awhile back. Relevant blogs on this site include Good Structure is the Key to Soulfulness and Skill in the Cellar vs Benign Neglect

The key to living without sterile filtration is a planned program to establish microbial equilibrium. The extreme case of sulfite-free wine is considered in my discussion of the making of WineSmith Roman Syrah.

There are some guidelines to assess your risk. I presume you are aware that fermenting to dryness and completing ML assure against the first two, so the issue comes down to Brettanomyces, which is much less well understood. This yeast can ferment pretty small amounts of sugar, so we’d want to be below 100 mg / 100 ml of glucose + fructose by enzymatic method, and also to bottle with fairly low dissolved oxygen or at least have good reactivity so after a month or two, you’re down to zero (this last should not be a problem if you bottle big reds which still have some reductive strength, i.e. not too overripe and not over-aged and tired).

An important element of microbial equilibrium is a warm cellar. If your wine never gets over 59oF in the cellar, it’s much more likely to play out its microbial endgame in the bottle. I age at 62 - 65 oF over the summer. It is also critical to refine a good aromatically integrative structure in the wine, so any Brett character is not perceived as a distinct element. To acheive this, we use Phase 1 (pre-ML) micro-oxygenation to stabilize a rich, light integrative structure. There are other paths to refined structure -- the result is the critical point.

You might want to try experiments with and without sterile filtration to make up your own mind. A lot of this decision gets down to personal stylistic preference. I’m personally not too Brett-averse, and I really want my wines to be energetic, multidimensional and soulful. Of course the final decision may come down to the power of your sense of adventure vs whether you like to sleep at night.

My point of view as a consultant is more experimental than most commercial operations. I’m interested in learning more about what wine is and how it behaves. Exploration necessarily involves risk. I’m more willing than most of my clients to explore possibilities outside the mainstream along with my brand's hardcore customers who are willing to strap in for an interesting journey which they realize may be bumpy, but never boring.