Working at Kamil Juices II

In a previous post I described getting hired at Kamil Juices, and I said that I’d like to get into the details of working there. So that’s today’s long-delayed update.

As I already mentioned, during my 4 months at Kamil Juices I made something like 8000 bottles of wine. It’s difficult to put an exact number on this, because work was very collaborative in the sense that I worked with my coworkers/supervisors for pretty much every non-trivial task. Everybody did everything, to put it another way: no task was the sole preserve of one person, except payroll of course, which was something that only Jerry did. She did that very well, actually – one of the little things that made working there so good was that Jerry was very conscious of my (and Allison’s) pay schedule, and I was always given a cheque a few days before the 15th and the 1st of the month. This made things a little easier, especially regarding first-of-month major expenses like rent.

I think the easiest way to talk about working at Kamil Juices is to describe the in-house winemaking process. I’ve made a few batches of wine at home, and there are lots of similarities, but having a single nearly-uniform supply of starting material and a dedicated, professional set of equipment and space to use it makes a big difference in how long certain tasks take to accomplish. Each batch of wine takes 6 weeks in the store. Of that 6 weeks, a single batch requires attention from somebody like me for about 45 minutes, spread out thusly:

10 minutes to start
30 minutes to rack, stabilize, and move to the cooler
10 minutes to set up and filter
5 minutes of miscellaneous bookkeeping and running around.
Bottling takes about 15 minutes, maybe 20 if we’re not in any particular hurry, but that’s primarily the responsibility of the customer.

Starting happens the day the customer purchases the batch. The customer provides some legally-mandated details (“Can I get your address, please? Don’t worry, we’re not going to visit you or anything, we need this because of provincial law”), and pours a packet of yeast into a cup that contains a little warm water and a little fresh grape juice. Then the customer leaves, and I do something else (mop the floor) for 10 or 20 minutes, allowing the yeast time to hydrate. I know a start takes 10 minutes, because I timed several I did on a couple of different days, and even including set-up and cleaning afterwards, a start averages 10 minutes. This involves sterilizing a carboy, collecting the juice from the back room, pouring the juice into the carboy, adding the yeast and yeast nutrient, and placing the carboy onto a shelf. And mopping up the inevitable splatters – even with much practice, I’m unlikely to pour 11L of grape juice out of a canister without splashing a few millilitres onto the floor.

After a couple of days, a proper airlock is fitted to the carboy (replacing the paper towel we’d shove into the top) and the carboy is given a bit of a shake. This takes a trivial amount of time – if I was bored, I could kill 5 minutes by setting up airlocks in all the 2-day-old batches. Fermentation takes about 10 days or so, and we’d monitor progress by means of a hydrometer. Batches were integrated over a weekly schedule – we’d start measuring specific gravity for all batches started two weeks previously, on a given Tuesday. Thus some batches I’d be measuring would have been started 15 days ago, others only 10. Wine is pretty flexible in its schedule, so running fermentation longer doesn’t really do much – all of the sugar has already been converted to alcohol, so it’s just wine sitting in an airlocked vessel.

Batches deemed complete (specific gravities lower than 0.998) were racked and stabilized. This involves siphoning the batch into a bucket, cleaning the carboy, adding stabilizer (potassium metabisulfate), and moving it back into the carboy. Then the wine was moved into the large cold-room, to cold-stabilize for 3 weeks. Fermentation is a biological process; living cells convert sugars into alcohol, and poison themselves to death or dormancy. Stabilization, on the other hand, is entirely chemical, there should be no living yeast cells in the wine when it goes into the cooler. This sterility is achieved by the stabilizer: “meta” does very bad things to the enzymes inside living cells, thus ensuring that dormant cells don’t wake up. We used meta as a stabilizer mixed with 20% ascorbic acid, and as a sterilizer on its own at a higher concentration in water. Additionally, the racking process pulls the wine off of the layer of sediment built up during fermentation (the "lees", if you're pretentious), and this sediment contains a fair amount of viable yeast. Restarting fermentation later is a bad thing, yeast metabolism in the presence of 12% ethanol tends to result in lots of secondary metabolites that taste bad.

After 3 weeks in the ‘fridge, a batch comes out for filtering. We let wine rest at room temperature (in the filter room) for about a week, guaranteeing everything going through the filter was not cold. Filter day was Monday, in the event of a Monday holiday, we filtered on Tuesday. All of the wine to be filtered was racked into buckets, and laid out in the filter room in a logical order. Then the filter was assembled (16 pieces of cut filter paper inserted), and we’d start filtering. Filtering is something that can be done by 1 person (stressfully), but is much easier with 2, and relaxing and fun with 3. The easiest way to do it is have one person on the filter machine, controlling the valve, and one person on the each hose, inflow and outflow. Wine enters the inflow hose, gets pumped through the filters, and comes out the outflow hose into a carboy. To keep bubbles out and to avoid spilling wine between batches, the person on the pump shuts the valve every time a new batch goes in and every time a carboy fills. White wines go through the filter first, then reds, following the general rule that you can contaminate your red wine with a little white and it doesn’t matter, but you can’t get any red into your white – only a few drops of red wine are required to turn a nice white wine into a crappy-looking rosé. It’ll taste fine, but it looks like somebody has been bleeding into your bottles.

Filtered wine hangs out in the filter room until the customer shows up to bottle. Allison would call all the customers who were getting their wine filtered on Monday, and set up appointments. When a customer came in for their appointment, we’d bring out the wine, siphon it into one of our bottling machines, and help the customer sterilize, fill, and cork their bottles. Some customers, for whatever reason, needed much more help with this than others. Some of my favourite customers could do pretty much the entire process themselves, with me providing just enough help to make things go faster, not more smoothly. Other customers required me to basically do everything, but I generally didn’t mind this because the majority of these people were friendly and really old and feeble, so they’d tell me stories about their grandkids or whatever, and I’d blast through bottling in 15 minutes. If you handle hundreds of bottles every day, bottling an entire batch single-handedly is not something that takes a long time. If you make 3 batches of wine a year, and you’re 70 years old with mild memory-loss, you’re going to take an hour to do the same thing.

White wines take a little more work than reds, as they get clarifying agents (gelatine and kieselsol) added when they’re racked, and they get filtered twice (coarse and fine) where reds only get coarse-filtered. Filtering is mostly cosmetic – multiple rackings remove the vast majority of sediment, and you don’t expect to see through a red wine anyways. It does make a difference to the taste, and helps with aging – filtered wines are less likely to go weird in the bottle after a year or two.

We’d do many batches simultaneously for each step of the process, which is why it’s difficult to estimate how long it takes to rack, stabilize, and filter a single batch – a big chunk of the time involved is communal for 4 or 6 or 20 batches. For example, racking and filtering both require lots of clean and sterile buckets. It takes me about 90 seconds to sterilize one bucket, but only about 2 minutes to sterilize two, and maybe 10 minutes to sterilize a dozen. Similarly, it might take 30 minutes to set up and prime the filter, but only 2 or 3 minutes to filter each batch laid out in rows.

Doing the same process at home would take about twice the attention and time, because home isn’t specifically set up to handle large volumes of wine. Having a giant stainless-steel sink (ours was about 6 feet wide, counting the drain area immediately adjacent) with the overhead one-hand sprayer and a carboy-washer on the industrial faucet, and plenty of floor space and flat-bed carts around cuts the time to wash and sterilize anything by more than half. We didn’t bother running meta solution through the siphons, for example, because we were blasting hot water through them several times a day.

Day-to-day work at Kamil Juices was pretty good. There was usually something that needed to be done even when customers were not around, such as mopping, or checking hydrometer readings, or restocking stabilizer or oak chips for customer purchase. And my coworkers/supervisors treated me quite well. I only had one real argument with anyone, and all was forgiven the next day. Four months was about perfect for me, as it gave me a much-needed break from academia, but wasn’t so long that I settled into any particular ruts.

Thanks are due, I think to everyone there: Allison, Adrian, Jerry, and Bob. Cheers, guys!

---

While I was working at Kamil Juices, I was also working on Part II of my end-of-PhD plan. This involved the search for a “real” job. Working at Kamil Juices was great, but I knew and so did everybody else that this was never intended to be long-term. I had sent out emails when I quit my PhD, to people I’d worked with or had planned to work with, to let them know I was no longer a PhD student and thanking them for their help. I heard back from most people within a couple of days, but one contact I’d made didn’t reply for 3 weeks. His reply started with “I’m sorry to hear that [you’ve quit your PhD], but I think I can offer you a job.” I’ll save that story for another post.