Visit to DB breweries

Ken writes:

I’m back in New Zealand for a long overdue holiday and a visit to friends and relatives.

And since I’m here, I had to pay a visit to DB breweries’ Waitemata brewery to view the continuous fermentation system (cf). I wrote about cf earlier in this post.

The DB cf process was covered in brief in the earlier post. There are three main tanks. The tanks are all stirred to keep conditions homogeneous within each tank. The first tank, known as the hold-up tank, comprises 6% of the total system volume and blends vigorously oxygenated fresh wort (O2 25-45ppm) with actively fermenting beer from the second tank and recycled yeast to maintain a pH of <4.4 and ABV of >2.5%. Hold-up tank temperature is 9C. The cool, acidic, mildly alcoholic conditions help guard against bacterial infection.

Drawing of CF system at Waitemata Brewery


The second tank is 63% of the system volume and is where the bulk of the fermentation happens. The tail end of the fermentation happens in tank three, which is 31% of system volume. The second and third tanks are maintained at 15C. A fourth, yeast sedimentation, tank finishes off the CF system proper. The yeast is washed with cold sterile carbonated deaerated liquor and some is recycled into the hold-up tank. CO2 is collected from fermentation and used elsewhere in the brewery and packaging lines.

I was really struck by how small the cf system was. The total system volume was 2000hl it is capable of producing up to 50hl of green beer per hour, meaning a residence time of 40hours, although my guide on the day, brewery technical manager Doug Banks, said that the residence time is varied between 50 and 90 hours depending on demand. The green beer undergoes a continuous maturation stage for a further 40 hours to remove diacetyl and then is ready for preparing for packaging (I assume this means clarification, and possibly bringing CO2 levels into product specifications).  Traditional batch processes would need a huge tank farm to accommodate these volumes of beer (and in fact the Waitemata brewery does have a large tank farm because it also produces beer by the batch for different brands, notably Heineken). Before seeing it for real, I don’t think I’d really appreciated just what an enormous capital saving cf achieves. A quite physically small brewery could produce huge quantities of beer.

The other highlight of the tour for me was seeing the old brewhouse. It was built in art deco style and is a work of art as well as an impressive piece of engineering.

lauter tun with run off sample ports.
lauter tun with run off sample ports.
Steps up to upper level where the mashing and lautering happens.
Steps up to upper level where the mashing and lautering happens.
brewhouse control panel
brewhouse control panel

One interesting feature of the old brewhouse is a lactobacillus delbruecki reactor vessel for acidifying the mash. Lactic acid was produced in this vessel to bring the mashing conditions to optimum pH level.

Lactobacillus Delbruecki reactor
Lactobacillus Delbruecki reactor

It was a very impressive space. Beautifully uncluttered and beautifully designed and I can imagine it would have been a wonderful space to work. I don’t think the brewhouse is in regular use at the moment although it is still in working order.

I’m very grateful for the chance to view first hand a piece of New Zealand brewing history and happy to see one of New Zealand’s key innovations still going strong into the future. (Some people may be snooty and dismissive in a kind of reverse snobbery kind of way, but fermentation technology is in principle something every brewer can use provided they have the sales to justify producing beer in such large volumes. The details of the recipe matter for the earlier stages of wort production. Maybe in the future the likes of Brewdog or Sierra Nevada will use this technology for their flagship products).



Yoghurt and raw ale under the microscope

Ken writes:

The title is intended literally, I’m afraid. I got out my microscope today and had a little play around.

Yoghurt at 400x magnification
Yoghurt at 400x magnification

I first had a look at some of the boys’ yoghurt that they’d left lying around. I diluted with a little tap water and had a look. You can’t see too much from the photo. I think the debris you can see is probably some kind of protein. When I increased the magnification to 1000x I could definitely see motile particles, which I took to be bacteria (the rest of the field wasn’t moving so that eliminates the possibility that I just had too much liquid sloshing round on the slide). The little movers were pairs and chains of little circles (cocci?). I couldn’t get a decent picture unfortunately. The photos were taken with my phone, incidentally.

raw ale, iodine stain, 400x magnification
raw ale, iodine stain, 400x magnification

This one is interesting. I stained the slide with iodine, because I want to see if there was any unconverted starch in my beer. Unconverted starch stains black in the presence of iodine. You can see form this picture that there are indeed two lumps of it. The rest of the tattered looking stuff is protein, I think. A raw ale is one that hasn’t been boiled. It’s naturally cloudy because it still has a lot of protein in it. In the bottom of the picture, the circle with a thick dark outline is a tiny air-bubble. You can’t make out any actual yeast in this photo graph, but there were a couple of yeast cells visible in the live microscope version of this picture. Unfortunately, the iodine killed off any bacteria so I couldn’t see any little swimmers like in the previous picture. I think I will try to repeat this later without the iodine.

Good Angry

Ken writes:

A few weeks ago while I was driving back to Dublin on a Friday I missed about five or six calls from my boss. I was driving so I couldn’t take the calls. Eventually, he switched and rang my understudy Chris who took the call and got the bollicking. I called and left a message on my bosses answer machine returning the call but he never called me back about it. I had to find out from Chris that he was upset because we had finished early on the Friday. He was out of order for many reasons. The first is that we had told his wife, Boss #2, that we were finished and leaving early and she didn’t object at the time. We can only assume they speak with one voice or else having two bosses is impossible. But secondly, he was out of order because we had in fact done all the work there was there to do. Boss #1 had said in the past that he didn’t want to be involved in the day to day running of the floor, because he works most of the week in Dublin, for starters. He has given that responsibility to me. And there was no more work to do. I don’t believe in engaging in pointless make-work projects just to make up the hours. Finally, no way should he be taking the matter up with Chris who was the person with the least input into the decision to finish early.

Anyway, there is a deeper reason why we were finished early and I wanted to address it so I called a meeting for the next available time (the following Wednesday) when all interested parties were available. At root the problem is that we are making more beer than we are selling. Sales and production need to be coordinated or else you get steadily increasing stockpile of beer that is slowly going off (You can’t keep beer from ageing and staling no matter how carefully you prepare it). The blame for the overproduction falls squarely with the bosses because they have been riding me and Chris hard all year to get us to produce more beer more efficiently. They also are the only ones with access to data on sales. So they should know how much they are selling. In recent weeks over production has meant that we have been relying on other breweries kegs to package our product which is akin to squatting in someone’s property rent free. We deprive them of the use of their kegs for weeks and weeks while our beer sits unsold in it. To top it all off, the sales/production thing could have been managed perfectly if they had been willing to brew on contract for other microbreweries (most Irish microbreweries don’t actually have breweries as such but exist as brands only, produced and packaged by someone else). We could have had a piece of that action, it would have allowed us to run the brewery at full capacity and most efficiently and selling the beer would have been someone else’s problem.

It was because we had been over-producing that there was nothing more to be done on Friday afternoon. I called the meeting because I wanted to put on record and let the bosses know that sales was the problem for the brewery at the moment. You can tell boss #1 and boss #2 these things but the is no evidence that they listen. I wanted to make a more formal representation and also to clear the air after the nasty business that had been the Friday afternoon telephone call.

Despite the fact that I’d requested a meeting on Monday morning, no body mentioned the Friday call or the meeting all morning and so after lunch I went to the office to ask for the meeting to take place. Boss #2 said boss #1 had gone off making deliveries so we couldn’t have the meeting. I wasn’t happy with that so I asked her to deputise for her husband. She said she would but must have called boss #1 because when we arrived in the meeting room fifteen minutes later she said boss #1 would be back at 4:30pm and had insisted that we wait until then. While she was saying this, boss #1 called me and launched in to a tirade. He was incensed that I had called a meeting and had insisted it take place. He was the boss. It was his company and we would be doing things his way. As it happened, a 4:30 meeting would mean that I would miss an appointment later that day, but apparently that was not important. In fact it didn’t matter because while he was on the phone, boss #1 became so overcome with anger that he had to turn the car around and return to the brewery to have the meeting.

Now there has been a general souring of relations between the bosses and the workers over the last few months (probably because they’re worried about sales and revenue). But I didn’t want to get into that in the meeting. I really just wanted to clear the air and make as plain and clear as possible that production cannot outstrip sales (it’s hard because you have to anticipate sales so you can have beer available for them. They have never managed to give me any sales forecasts and yet I am supposed to provide them with packaged sellable beer of every kind when they need it). Anyway Boss #1 opened the meeting by reading me sections of my contract where is says no premium will be paid for anti-social hours or weekend work. Why? because he wanted us to brew late one night the next week so that we would still be brewing in the evening when he invited some local area publicans and tradesmen to the brewery. As it happened, only one publican showed up, and none from the little village where the brewery is located (Irish publicans are a truly inspiring and imaginative group). I said my reading of that passage was that it meant I would not get paid extra for working any antisocial hours, not that I had to work them. The contract states I will work a 40 hour week as well, but it doesn’t say I have to work any period of hours that the boss decides. As far as I am concerned that is all subject to negotiation and I told him that calmly and slowly. Boss #1 ranted and raved and tried his best to derail the meeting and talk about everything except the production issues that we had put on the agenda. Throughout he was talking about it being his business and him being the boss throughout, which although true, doesn’t give him licence to play the tyrant.

I never raised my voice and I steered the meeting back to the agenda and I didn’t permit myself to saying that I would regret. I am not happy in the job, but I recognise that it has given me opportunities I wouldn’t otherwise have, and it also pays my bills. I didn’t just want to chuck that away because boss #1 is a prize twit. During the meeting, it felt great being angry. But not angry in a hot, can’t think or see the wood for the trees way, but in a cold way. I knew why I was there and what I wanted from the meeting, and what my goals and vulnerabilities were. Above all it felt like I was in control and I liked that. I’ve got to go back to work on the 5th. There will be new meetings in the new year. and although it is tiring and effortful to have to do so, I will be going in to bat for me and for the production crew on the floor and I will do it in the cold cold way I handled myself in that last meeting.

Messing about with malt, or the problems of ensuring consistency in a small brewery

Ken writes:

When thinking about the malted barley we use in our beer, there are three fixed points to consider. We have to keep the amount of potential alcohol the same from batch to batch. We have to keep the colour the same, since people tend to notice things like changes in colour. And we have to keep the flavour the same. The amount of potential alcohol is determined by the yeast we use and the amount of sugar we get out of the malt. Maltsters supply spec sheets listing the LDK (litre degrees per kilo), essentially how many ‘sugar points’ yielded by each kilo of malt. A kilo of malt with an LDK value of 300˚, for example, would yield 1 litre of wort with a specific gravity of 1.300 or 10 litres of wort with a specific gravity of 1.030 (i.e. 1 litre with an excess gravity of 300, what I’m calling ‘sugar points’, or 10 litres with an excess gravity of 30 (30 sugar points). If the LDK values change, then it is a simple matter to increase or reduce the weight of malt accordingly to keep the original gravity, and there the potential alcohol content, constant.

Maltsters also provide values for the colour of wort made from the different malts. As with strength, colour adjustments can be made by increasing or reducing the proportions of the different malts as necessary to accommodate changes in the colour from batch to batch.

Unfortunately, there is no objective reckoning for taste. I don’t even know if taste follows strength or colour. If the recipe calls for a certain proportion of the sugar points to come from a certain type of malt, then because the brewer just has to take the LDK and colour of the malt as it comes, the mass of malt and therefore its effect on the colour of the final beer will be determined. On the other hand, the recipe could be turned on its head, and we could specify what proportions of the overall colour of the beer come from what malts. Just as before given the colour specifications, the weight of malt will be determined, and therefore, how much sugar will be obtained and indirectly the strength. In other words, if the malt varies, then keeping colour fixed may mean a change of strength and vice versa.

I don’t really know how to solve this problem. I can see why large breweries would want to dictate to the maltster what malt specifications they will accept. It would be so much easier to keep a consistent product if you could rely on colour and potential strength not changing.

I think I’m going to sort of assume flavour follows colour. I don’t know how safe an assumption this is, but at least some of the compounds responsible for the colour of darker coloured malts are also responsible for some of the malt flavours (Maillard reaction products). I’ll try to have the coloured malts contribute as much colour as they previously did, but I’m not going to try to keep the colour contribution of the base malt the same. Instead, I’ll just adjust the base malt however I have to adjust it once the coloured malts are calculated to keep the original strength the same. There’s no real principles behind this; it’s a compromise to meet competing desiderata.

crystal 150
(image of crystal malt nicked off the internet from somewhere and used without attribution or permission)

Brewing vs Yeast Husbandry

Ken writes:

I’ve come to realise a big difference between working in a microbrewery and homebrewing, which is in the amount of shepherding and monitoring of fermentation that goes on after the actual brew day.

At home, my fermenting vessels were buckets made from (food grade) high density polyethylene (HDPE) or polypropylene (plastic types 2 and 5 respectively). There was very little I could do to control the conditions of fermentation. I had a warming belt I could put around a bucket to keep it warm on cold winter nights, but I had no way of chilling the fermenters or of regulating temperature variation over the course of a twenty four hour period. So I had to give up brewing during the hottest part of the year or try brewing beer styles, like a Belgian saison, with yeasts that can tolerate high temperatures without spoiling the beer. In other words, there was a limit to what I could do and the course of fermentation was more or less in the hands of the gods.

In a microbrewery it’s a little different. We have large insulated fermenters equipped with cooling jackets which let me set a temperature for fermentation which will be maintained. I’ve got to decide what temperature best suits the yeast and the beer style. Ale yeasts prefer to work at warmer temperatures 18˚-24˚C and lager yeasts prefer cooler temperatures 12˚-14˚C. Warmer temperatures lead to an increase in metabolic byproducts of fermentation such as fruity tasting esters, so if you want a clean, neutral tasting beer, you need to choose a temperature at the lower end of a yeast’s range. On the other hand, the time it takes for the fermentation to be completed is also temperature dependent.

Judging the end of fermentation is a bit of a guessing game because you don’t know how fermentable the wort is. It might ferment out dry or it might stay comparatively sweet. In theory, by following the recipe closely and following the same procedure we shouldn’t have much variation from batch to batch, but you only know you are doing everything the same way when you get the same results and your wort is exactly as fermentable, dry or sweet as it’s supposed to be. Just thinking you’ve followed the same procedure doesn’t mean you have followed the same procedure, so you’re still in the same state of ignorance about when the fermentation is complete. There’s a test you can do to find the fermentability of wort, but it takes a couple of days and we’d need more lab equipment than we actually have to carry it out.

Judging the end of fermentation correctly is important because we need to move the beer into conditioning tanks. If the beer is moved to soon, there will be a lot of beer in suspension, which will be carried over into conditioning. This is problematic for two reasons. The first is that high concentrations of yeast combined with cold, nutrient depleted, mildly alcoholic conditions can result in yeast cell death which can give an off-flavour to the beer. The second is that high concentrations of yeast can make filtering the beer difficult.

Also, lighter flavoured beers, especially lagers, benefit from a diacetyl rest at the end of fermentation. This means keeping the beer on the yeast for 24 hours after fermentation has finished to allow the yeast to absorb diacetyl from the surrounding beer and metabolise it to a less flavour active compound. Diacetyl is a distinctly buttery or butterscotch off-flavour and is a normal byproduct of fermentation.

Add to all this the complication of managing the stocks and the tank space in the cold room and ensuring we always have a fermenter free to brew each week means the part of my job concerned with managing fermentation is much larger and more important than I realised. So far I seem to be doing alright. I’ve had one diacetyl-related disaster and one yeast cell death related disaster but I’m hopeful that lessons have been learned.

Job agony

Ken writes:

My work life is almost unbearable at the moment. I will bear it, of course, because I can bear almost anything. It’s one of the most important skills I learned studying philosophy, namely How to Take Beating. I feel like my self-esteem and credibility is taking a savage beating on an almost daily basis at work at the moment. I’m on a six month probationary contract at the moment, which I am convinced will not be made permanent. I want to write this post to blow off a bit of steam, but also to think about some of the positives I can take away from this if I really do get a P45 in a couple of months. (NB: Gosh! how many times can I say ‘at the moment’ in one paragraph?)

Things seem to be going wrong a lot at work at the moment. For example, we’re having difficulties with filtering beer leading to heavy losses. We’re also having difficulties with beer superattenuating, that is, becoming more alcoholic than we planned. This only happens at the expense of residual sweetness and mouthfeel in the beer so the beer ends up tasting crisp and dry (which can be a good thing, but only if that is what you want). Super attenuation happens because the yeast is still active in the beer after it is transferred to the conditioning tanks. We have a bit of a backlog in the conditioning tanks because it took so long to get our bottling/kegging equipment. We started brewing at the end of January, and the bottling/kegging equipment arrived the second weekend in March. We’ve got thousands of litres of beer in tanks waiting to be bottled or legged.

Bottling is itself a bit of a problem because the bottle filler only fills two bottles at a time, the capper only caps a single bottle at a time, the labeller only labels individually and there is a neck label that must be individually applied by hand as well. The bottles must be rinsed before labelling, to remove beer foam, and allowed to dry. Boxes need to be made up and labelled as well. Bottles need boxing up and stacking onto pallets. All told, producing a single bottle of beer takes about a minute. An 8 hour day contains 480 minutes. So we’re only capable of bottling, labelling, packaging etc 480 bottles per day or 240 litres. That’s 1200 litres per week, which is a problem, because we brew about 2000 litres per week. And they want us to produce more!

I’m only supposed to work 40 hours per week (9-5 M-F). I arrive before 9. I leave after 5. I don’t take a lunch hour but fit the sandwiches and fruit in my lunchbox around the available time. Does anyone recognise this? No. I’m being judged on results. So far the results are… meh. I’m no better than I should be.

My immediate boss seems to have lost all patience with me. To be fair, she is pretty stressed herself. Setting up the brewery involved a massive capital outlay and our early attempts (we’ve still only brewed twenty batches) haven’t really made any sort of dent in the loans. We’re still calculating the unit cost of making the beer (the real unit cost not the projections), and we’re still so inefficient that I doubt they make any money on it at all. Still, part of the problem would be the ridiculously labour-intensive bottling regime. But be that as it may, some of the problems stem from the boss’s insistence to sell every drop that comes of the new machinery. This has created problems for us. The first batch was out of specification and as a result had to be blended away which looked like a lot of jiggerypokery latterly when we tried to explain it to revenue. If we’d just dumped it as a trial brew, we’d have gained the same educational value, but not had the stress of squeezing it past the tax man. They started with, to be honest, ludicrous estimates of how much the brewery could produce in a week. This has meant that they want something to be brewed every week despite the fact that we aren’t actually selling much of it yet and we can’t efficiently package what we have made.

My boss has become really terse and unsympathetic with me. For example, I let her taste the rejigged version of a beer we have which I am not satisfied with. THe rejigged version is a marked improvement in my opinion. She didn’t like it, but instead of saying e.g. ‘hmm, I think I preferred the earlier version’ she said ‘it does nothing for me’. For example, we had a revenue inspection this week (more stress for everyone). The revenue agent asked how much the brewery cost. The boss gave a figure and the revenue agent said, ‘wow, well it had better work then’ and we all sort of laughed and my boss said, ‘yeah, Ken’ and looked at me. For example, she blamed me for the super attenuation problem, but it is really a scheduling problem as it would not have happened if the tank turn around times were faster.

Unfortunately, it is my duty to report any upsets and setbacks to my boss. So I am inevitably the bearer of bad news. If things go smoothly, there’s nothing to report.

OK. enough of that. what are the good things to come of all of this?
A. I know a lot more about day to day operations in a brewery than I did before.
B. It should be that much easier to find another job if this doesn’t work out.
C. If this doesn’t work out, I’ll probably be able to work nearer to home in the future.
D. They didn’t have to give me anything. They took a chance on me and that was itself a gesture of good faith.

What I want is for a natural diminution of stress levels as beer sales start to pick up and the problems get ironed out. I hope this is still the most likely outcome. I just need for there to be no more fuck ups. I’m going to operate a more stringent ‘need to know’ policy from now on. I have a tendency to be honest about everything, but when people are stressed, it doesn’t always benefit them to hear the details about everything.

Settling in at the brewery

Ken writes:

Just a quick post to record how things are going at the brewery. It has not been without a few teething problems, but we’re starting to get a good smooth flow to our processes now. Here is a picture of our leaky filter. We brew two beers and one of them filters nicely and the other one likes to leak out of the filter press. I’m not sure why. It could be that length of time the beer has conditioned before we filter it. It could be the type of yeast we use. It could be another aspect of the recipes. We do something called double filtering, which involves passing the beer through two sets of filters(coarse and fine) in the same filter press. It was leaking on the coarse side of the press, so we’ve added more filters to that side to spread the filtering duty over a wider total surface area in case that helps.

leaky filter

Here’s a picture of the apparatus of filling a keg of beer. We mostly use 30 litre kegs. The keg sits on a scales so we can accurately judge how much we’re putting in and how fast we’re filling it. This relies on the wonderous properties of the metric system that one litre of water weighs exactly one kilo. Beer is ever so slightly heavier than water, but we can still calculate precisely how heavy 30 litres will be. We can tell how fast we’re filling the keg from how quickly the weight is climbing. If you fill a keg too quickly, you get a lot of foam breaking out. This means that if we simply relied on foam coming out the tapping head to indicate whether the keg was full, we might underfill kegs (since a keg full of foam in the brewery would translate into a mostly empty keg in the pub when the foam had died back down). The tapping head is connected both to our carbonator and a separate supply of gas. The carbonator supplies the beer with a desired level of CO2 gas in it, and we use the separate gas supply to purge all the air out of the empty keg before filling (to avoid staling oxidation reactions), and to provide a small countervailing pressure in the keg to make sure that the gas dissolved in the beer doesn’t break out of solution when it is put in the keg. We’re aiming for what we call a black fill, where the beer is put in the container without any foam breaking out.

2014-03-28 12.17.13