Supremacy of the beer engine…

Sparkler in Action

Sparkler – being used incorrectly I believe, but it does nicely show how they function – from Flickr

There has been some discussion of sparklers recently. Yet again. The near-religious divide they create in the cask beer world is amusing. Personally I am usually happy to put it down as a matter of personal preference & taste and leave it like that. (Although it can be fun to carry on purely for the sake of a good argument.) Really it is another thing that goes in the same bag as preferences over temperature, carbonation, and actual styles of beer… I’d not profess that stout is the one true way to drink and enjoy beer, so why would I profess that some given fixed method of dispense is a “one true way”?

I, personally, prefer that my pint is not put through a sparkler. What they do is force carbonation out of beer. For the sake of appearances… that nice tight head.[1] There is also an impact on the flavour and aroma of the beer. What they cannot do is add condition that isn’t there in the first place. Neither can using a beer engine without a sparkler. Neither can gravity dispense.

Gravity dispense… is where I found my thoughts wandering. It is common knowledge, widely believed, that the humble beer engine is the best way to serve a cask ale. But why is it so? Is it so?

What a beer engine does is “suck” the beer from the cask along several feet of pipe, through a cylinder, and shoot it down a narrower pipe into your glass. In the process the beer is severely agitated – releasing CO2 from solution and helping to form a nice head on your beer. But this is a side-effect really – the purpose of the beer engine is simply to move the beer, against gravity, from the barrel to the glass. From the best place to keep the beer (nice cool cellar) to the best place to serve the beer (nice warm bar). It certainly saves on having to run up and down stairs a lot.

So what we have is a practical device that has the side-effect of helping to force CO2 out of solution and giving your beer a nice foamy head. The humble sparkler takes this one step further by adding to that agitation, forcing more CO2 out of solution more rapidly – and forms a tighter head on your beer. (Whether or not that is an improvement is entirely up to the individual.)

But the fact is that that same beer dispensed directly from cask into a glass is going to have more CO2 in solution. Beer dispensed via beer engine (with or without sparkler) simply cannot be less “flat” than the same beer dispensed directly from the cask. Yet is seems to be a commonly held belief that gravity dispense means flat beer. This, I suspect, comes down to two things:

  1. The “head” on a beer is seen as an important indicator of non-flatness. Yet is it really? The mechanism of a beer engine (especially with sparkler) can force what little CO2 remains in a beer out of it, it creates head against all odds. The result is your already pretty flat beer ends up even flatter for the sake of that head. Appearances can be deceiving?
  2. Most people’s experience of gravity dispense is beer festivals. And there are a whole host of issues with beer festivals, especially: insufficient conditioning time, insufficient cooling, insufficient cellaring skills. All three of these yield flatter beer.
    1. Conditioning: the tight schedule of festivals relies on beer coming from the brewery/supplier sufficiently conditioned. This is the norm these days, but sometimes a beer needs time that a festival simply cannot give it.
    2. Cooling: warmer beer has less CO2 in solution. If you vent your beer at 15C, say, it’ll loose a lot more CO2 – and if it then cools to 12C it will appear even flatter as a result. (Ensure your beer is cool before you vent it.)
    3. Cellar skills: knowledge of the previous two, but also simple things like use of nylon pegs at all non-service times for most beers. Keep that condition in the beer!

So, basically, beer festivals give gravity dispense a bad name.[2]

An antidote to this is a sharper focus on cask ale “cellaring”/quality at festivals. Adequate cooling, adequate space and time to match bar capacity, and adequate knowledge and skills. With these three sorted a pint of cask ale whether from beer engine or gravity should be fantastic and sport a nice frothy head.

Gravity dispense in pubs is rare – but is a growing trend mainly thanks to the micropub “boom” and use of non-pub buildings as pubs. Where I have come across it it usually suffers the same problems as festivals. A lack of ability to give at-bar stillaged beer appropriate time and cooling are the enemies here – so too is the lack of knowledge of some of these new landlords who’re new to keeping cask ales.

Gravity Dispense - middle of 3-day festival

Gravity Dispense – middle of 3-day festival

Gravity Dispense - end of 3 day festival

Gravity Dispense – end of 3 day festival

If there is a natural superiority to the beer engine it is that it allows you to keep your beer in a properly cooled cellar away from the point of dispense. Gravity can be just as good… but due to various practical limitations the odds are often against it.

The story does not end there with respect to comparing cask dispense methods. I have tasted a few beers side-by-side sparkled-and-unsparkled and there appears to be a flavour difference. Albeit this has not been done in a blind tasting situation – just casually in a few northern pubs. The sparkled beers usually appear muted… blunted… you could say “smoother” perhaps. What is behind this? Tight head on a sparkled pint blocking some level of aroma perception perhaps? Is enough CO2 forced out of the beer to perceptively alter acidity? Is there something else there… does the sparkler add a rapid-oxidation effect, with an impact akin to putting wine through a blender to “breathe” it?

And if a change of flavour is evident in sparkled-v-unsparkled cask ale – is there similar between gravity-v-beer-engine? Can a beer engine actually change, and improve, the flavour of beer? I doubt it. But I keep an open mind…

Some further research is required… aka: drinking beer ;)

[1] I’m not saying sparklers make beer flat! If the cask is well kept and contains a good amount of CO2 in solution there will still be a good vol CO2 in the beer post-sparkler. In fact, as I see it, in order to serve non-flat beer sparklers demand the best in cellarmanship. (And perhaps this is why beer is generally kept better up north than it is in, say, London.) That said, I near-always find a sparkled pint is flatter than I would like. I do have a preference for a bit more carbonation in my beer however, dirty foreigner that I am…

[2] Though some are better than others – I pride myself in the quality of the beer at my own festival. But even I, with my pretty good practical knowledge and scientific understanding, cannot have 100% perfect beer at a cask ale festival. Cooling is a pain – even using the fancy CAMRA HQ cooling equipment (this year I had beer getting a bit too cold). And the “time” element is impossible to do anything about (luckily few beers come to a festival with insufficient condition in this day and age, but there are always some.)

Followup: Cask Ale Fallacies

Well, there has certainly been some interesting discussion out of the “Three Cask Ale Fallacies” post. On the back of some of this discussion some updates have been made to that post (all clearly labelled as such). It is time to post a follow-up I think, rather than muddy the waters by adding more updates to my own post.


There is some discussion worth reading on the Boak & Bailey post.


I just want to take a moment to state clearly what I am NOT claiming, and never claimed:

  1. Most importantly: I am NOT claiming that Ed is serving duff beer. I think his process is sound but I challenged some of the methods and justifications. I thought I was very clear in my post that a) I expect Ed serves fantastic beer, and b) I think Ed is the sort of cellar person we need in more pubs.
  2. I am NOT claiming that there is a one-true-way to keep cask ale. This would just be totally bonkers. I just seek to examine the reasons behind certain thinking and practices.
  3. I am NOT claiming that I am an expert on the subject. However, I do consider myself an informed commentator… who is seeking to extend and clarify his own knowledge on the subject.
  4. I am NOT claiming that the general public ought to be aware of any of the technicalities. This is a discussion for people who keep beer or are deeply interested in it. Albeit if I hear “the general public” peddling bullcrap as fact I will sometimes challenge them on it. This most often involves “I’m a CAMRA member” types who’re are proselytizing or trying to “educate”. It happens on the oxygen front predominantly. I suspect I find myself on the receiving end of such education more often than older folk purely because I spend too much time around beer and I’m under the age of 50 (I’m 34, for what that’s worth) and I’m a foreigner (Australian). So it is assumed that I need a proper education in “English beer”. This is usually hilarious… but anyway, I digress.

The conversations coming out of my original post ranged far and wide with all kinds of wild tangents of discussion and heated argument taking place. Even “CAMRA bashing” became the topic at one point. I’m sticking purely to my original points in this post.

1. Oxygen (O2) – needed in secondary/conditioning?

I see my job as the cellarman is to use what the brewer has given me (namely yeast + sugars), add some oxygen, and let rip.

There’s all many of biochemical reasons why oxygen is important – one of the most is allowed aerobic respiration to occur which re-metabolises compounds like diacetyl (produced during anaerobic).

This point hasn’t really been discussed much. I stand by my statement that any requirement for O2 to help/start “secondary” & “conditioning” is a myth.

I don’t know where the diacetyl item comes from… the material linked to in the comments appears to me to claim the opposite. (Find the “OXIDATION” section, it does explain that plenty of oxygen is needed in the wort to ensure a healthy fermentation with little diacetyl production.)

I take it by the lack of discussion on the topic that there isn’t any real technical challenge or counterexample.

With respect to Ed’s cellaring process I don’t think O2 plays any role in the quality of the beer coming out at the tap. When a cask is vented it’ll be ejecting CO2 which will mostly keep air/O2 away from the beer. If the cask is sealed with a nylon peg when it is not active then this ensures O2 continues to stay out of the picture. It won’t be until the beer is served that air/O2 is drawn into the cask as per usual with cask ale. (Where we hit a whole different kettle of fish with respect to the desirability of oxidation to “soften” beer.)

3. “Secondary” (cask conditioning) – does not occur at the brewery?

you say “do any brewers send out beer that isn’t in condition these days though?” Every single one does.

Ever drunk cask beer straight after it’s been racked and tapped? There is no condition in it whatsoever.

There has been a little discussion on this topic but not a lot. I’m standing by my position that _most_ cask beer leaves the brewery with suitable “condition” (sufficient vols CO2) to serve. (Meaning only that there is sufficient CO2, not that the beer is ready to serve… this will depend the beer and maturation required to achieve peak/desired flavour.)

There has been some lighthearted challenge on the use of the term “secondary”. A bit nitpicky perhaps… we all know that it is just a continuation of fermentation, right? And it is common practice to use the term to refer to that phase of fermentation used to achieve sufficient CO2 is in the beer. This “secondary” fermentation might take place in bottle, cask, or tank – and it doesn’t magically cease when the beer leaves the brewery. (Ignoring sterile filtered and/or pasteurised beer of course.)

As for the original point – I fully accept that sometimes beer does leave the brewery without sufficient carbonation. I said as much in my post. There is a variety of reasons for this – the yeast might just be a bit lethargic, possibly due to cold weather, or the brewery might not have given it enough time… and in this latter case I believe it is normal for a brewery to give a warning. It is also not unusual for a pub to reject a beer that does not arrive with a reasonable CO2 level. This has been my personal experience on this subject.

I have asked a few brewers about this in the last week and the uniform answer has been that they don’t let the beer out of the brewery until sufficient time has been given for the beer to reach a desirable carbonation level. (It is an amusing co-incidence that one brewer who I don’t believe was aware of this discussion warned me: you can have this beer but we’ve only just racked it so it will need a week.) The most useful bit of information on the topic I’ve found is a blog post from Jon of Stringers Brewery: Gas, and Hot Air – this discusses tank versus cask conditioning at the brewery, the relevant part of his post is the bit where cask leaves the brewery at 1.4 vols CO2 to be ready for venting at the pub at 1.5 vols CO2 (and will be finally be served below this after being vented).

Why is Ed’s experience of this at odds to the reality? It may be due to venting immediately after shaking the cask… this will cause a violent exit of CO2 from solution. So if you immediately sample the beer from this cask it probably is seemingly flat. However as Ed explains he then gives his beer plenty of time to generate some more CO2 to recover its condition.

My experience of cask ale is that is usually arrives from the brewery well carbonated. It is left on stillage for at least half a day and then vented carefully to try and retain CO2 in solution. Retaining CO2 is ideal for beer festivals, the context in which my own experience lies. I don’t have the experience of a pro cellarman, but I have vented at least 1000 casks in this way across a very wide variety of breweries and beers.

I cannot say whether one or the other method results in better beer. One could ponder that Ed’s violent “whoosh” of CO2 release carries away undesirables – sulphur compounds perhaps that give “green” beer its “Burton Snatch”/struck-match aroma? On the flip-side this could perhaps cause the beer to give up volatile hop aromas that brewers try so hard to keep inside the beer. One for some experimentation perhaps.

2. Yeast rousing (shaking/rocking casks) – is essential for good beer?

This item has been rearranged to be last as it is the “myth” I’m least convinced about. When doing my training I was instructed by a good brewer to *not* agitate casks unnecessarily. I took it as being a strict rule, coming from an experienced chap with a Heriot Watt masters degree. However I’m now thinking this was a “lies for children” sort of instruction… a rule of thumb that hides a hell of a lot more “under the bonnet”.

Where I now stand on this is: know your beer.

The factor that is most important here seems to be finings performance. A complicated subject in which there are several variables to consider (brewery fining practice, chain of supply, yeast properties, beer properties, temperature… at the least). This is why “know your beer” is the rule here.

The main item of interest seems to be performance of finings across “drops”. The feedback seems to go something like: On the first “drop” after the finings have been added to the cask the trub layer will be relatively loose. On the second drop a little tighter. On the third and maybe fourth drops the tightest. After which the finings will start to get “tired” and be less effective. This is the collected anecdotal evidence from several brewers and does not represent a set of hard rules, but seems to be a generally accepted pattern.

If you buy your beer from a brewery that fines their beer when they rack it it would have dropped once after racking. Then a second time after transit to your cellar. Then the third time after you agitate it before putting on stillage. (Unless it has been shaken about additionally along the way!)

If you buy your beer from a distributor then there is probably at least one extra step along the way, likely more! As a new distributor myself I am now ultra-aware of this and will be careful not to move casks unnecessarily in the future.

Obviously there is a vast array of possibilities here. I hear some breweries add finings only when the cask leaves the brewery – so if receiving direct from the brewery it is only on its 1st drop when it hits your cellar. Based on the ideas discussed so far there would seem to be a definite benefit to thorough agitation before putting this cask on stillage.

So, what’s the worst that can happen if you agitate? Due to chain of supply it’ll have been through 6 or more drops and simply won’t clear. (Ever? Or will it just be very slow?)

So, what’s the worse that can happen if you don’t agitate? It’ll be on its 1st or 2nd drop and may not form as tight & small a trub layer as it could – thus you have a higher ullage and you get a lower return on your cask of beer.

There are people who swear by doing it either way… I think perhaps a few more have stated that they do rouse than those who have stated that they don’t. However I’ve not had time to try and find and count all the responses.

Another question is: if a cask has been sat on its end for 24+ hours and you move that cask from that position to being sat on a traditional stillage, what happens to the trub layer and how does it settle without the additional agitation? It was specifically this situation in which I was originally instructed that there was no need to go shaking the cask unnecessary. However that is in opposition to this statement:

I’ve done some searching on the topic but haven’t found anything particularly enlightening. I did find a treasure-trove of interesting beer information along the way however: – in this I found papers on some experimentation with finings performance: Improving the Effectiveness of Isinglass Finings for Beer Clarification by Optimisation of the Mixing Process. Part 1: Laboratory Scale Experiments (Also: Part 2: Pilot Scale, Part 3: Full Size). But whilst these papers are interesting the context is not right in the context of this discussion.

Ed Wray did a little digging to see if he could come up with any data on the topic and came up with: So no definitive answer yet. I suspect it’s one of those questions to which the answer is “it depends”.

Ed Razzall has used his connections to do a little experiment for us all. He’s going to get two glass fronted casks of Ghost Ship from Adnams and treat them both exactly the same way aside from the vigorous agitation pre-stillaging. This is exactly the sort of thing I love… experimental evidence! I’m looking forward to finding out the results. It will not give us a definitive answer to the “do” or “don’t” agitate question… but it will give us a useful and interesting datapoint.


As is often the case more questions are raised than answered. I still stand by my original statements with respect to “myths” 1 (oxygen) and 3 (condition). In the case of myth 2 (rousing) we seem to have hit a clash of two differing schools of thought neither of which has any hard evidence to back itself up or shoot down the other side. Whilst the true answer may be “it depends” – I’d like to know why? Because that’s just the sort of person I am.

But I do accept there may be no clear (haha) answer to some of these questions…

Three Cask Ale (semi)Fallacies

[There is a followup to this post based on the extensive discussion that occurred around it on Twitter.]

This is in response to Ed Razzall’s cask ale cellaring tips. I write it because in the post and comments items arise which are persistent cask ale fallacies. (With some minor updates since original writing, including insertion of “semi” above!)

  1. Myth: “Oxygen is needed for conditioning”: It is often heard that oxygen (O2 in air) is needed for cask ale to “start working” i.e. begin secondary fermentation. I encourage anyone who believes this to go ask a proper brewer, or learn some of the science/(bio)chemistry of brewing! (The latter may be safer… if you do ask a brewer if he likes O2 anywhere near his beer do it from a safe distance.) O2 contact with beer post pitching of yeast is BAD. The initial aerobic phase of fermentation that needs plenty of O2 in solution is for breeding yeast cells and after this phase O2 is not desirable. This healthy cell population then start munching yeast until the beer is ready for secondary/conditioning. Yeast does not need O2 in the conditioning phase, it will munch sugar and fart CO2 quite happily without O2, it will mop up diacetyl without O2. Look up the Embden-Meyerhof-Parnes pathway [Ref: I recommend reading Principles of Brewing Science]. The only thing O2 will do to beer once it has come out of a brewery is RUIN IT and make brewers cry. (With an exception for those drinkers who like cardboardy oxidised flavours to “soften” their beer… that’s a matter of taste… *mumblepervertsmumble*)

  1. (Partial) Myth: “rouse the yeast”: (See update below.) Many seem to believe that before stillaging & venting you need to “rouse the yeast” to get it into suspension. I’ve heard this often and I don’t understand it at all. The cask has been manhandled onto a lorry/van, driven over bumpy English roads, and then tossed down a hole (cellar). So even if it needed “rousing” I’d say the job is well done already. But… does it need rousing? Why rouse it? To get yeast into suspension so it can “work” possibly? Even if the beer needs to “work” more (see next myth), shaking it up isn’t going to improve things much – there will be plenty of live yeast in suspension already. There is another factor here to consider: finings. Finings aren’t forever… in time their usefulness degrades (note that usually unfined beers come out of the brewery with longer best-before periods). The effectiveness of finings also degrades between them having to work. I.e. if you keep “rousing” the yeast the finings get “tired” and the beer takes longer to drop bright and the trub layer will not be as “tight”. [Ref: Cellarmanship 5th ed. page 99.] “Rousing”: at best achieves nothing, at worst means you’ve got to wait longer for clear beer. Now, Ed likes to stillage his beer for a good long maturation phase. So his rousing isn’t likely to matter one way or the other. He’ll end up with clear beer because he gives it plenty of time. Even good unfined beer should drop bright eventually.

    UPDATE 2014-05-13: There is additional complexity here regarding the isinglass finings used to help nearly all cask ale drop bright. As in James’s quoted Tweet above finings are more effective past the first drop. If the brewer fines as casks leave the brewery and delivers straight to your pub you’ll possibly be on the 1st drop, if the brewer fines at racking you’ll be on at least 2nd drop, if you’ve gone via a distributor you’re probably past the 2nd drop. The danger is that if you’re not at least at 2nd drop you could have a higher ullage. My suggestion here is that you get to know your brewer very well and ask their advice – if they’re fining as the beer leaves the brewery then there is some evidence to suggest that you could do with a good old rousing! But: If in doubt I would not rouse… because you just don’t know where you’re at and if you go over that critical 5th drop then there is a chance the cask will simply not drop bright at all. I’ve not found any good references on this feature of isinglass finings but will try and do so.

  1. Myth (mostly): “secondary occurs at the pub”: It is perhaps a legacy of historic practices that people believe breweries ship beer to pubs before substantial secondary fermentation has occurred. I hear this still happens sometimes… but in almost all cases: no. Most breweries do their best to ensure secondary has progressed sufficiently before the beer leaves the brewery. This is by the demand of the modern pub market which, for better or worse, often wants beer ready to serve within days of delivery. Historically this was not the case… which is why I think this myth persists in the collective “cask ale” memory. I mark this myth as “mostly” as every now and then I’ve come across beers that are “working hard” (enthusiastically fermenting) in the field. Plus secondary fermentation abates gradually, whilst the important fermentation phase will be complete by the time the beer reaches the pub yeast activity will not have ceased entirely. I also know that if you’re on a tight schedule to receive beer a brewer will sometimes let it out early – making it clear it won’t be ready to go to pubs/vent for a few days. I’m guessing there may be some pubs that have this as part of a regular relationship with some brewers they know (but that is only supposition).

None of this means Ed’s beer isn’t good. I suspect it is actually very good despite the myths. Because Ed likes to give his beers a good long time on stillage it’ll be bright regardless of rousing. Because beer ejects CO2 when vented very little O2 will actually manage to get into the cask to contact the beer – and note that Ed seals the cask airtight once it has stopped noticeably ejecting CO2.

In fact the impact of the fallacies in Ed’s cellaring is insignificant compared to the GOOD he does. Sanitisation! The puzzled looks I’ve had from people responsible for caring for cask ale when I ask about sanitising shives/keystones/taps/spears/etc scare me. Hard pegs! When not being served seal up you casks! Two things happen if you don’t, 1) more CO2 escapes: flat beer; 2) more O2 gets in: oxidised beer. TIME! Ed only serves his beer when he thinks it is ready, it doesn’t go on as soon as it looks right, it goes on when it smells and tastes right too. He gets to know his beers and learn what the right timing is for them. This is the skill of a good cellarperson. I bet his beer is the best for miles.

So the myths do not mean that no “cellarmanship” is required. Far from it! “Condition” in beer comes from secondary fermentation but the process of “conditioning” includes proper venting and maturation at the pub – you need all of these done well to end up with a pint “in good condition”. We need more cellarfolk with Ed Razzall’s interest and attention to detail.

Of course – you only have me saying all this. I am not a a pub cellarperson, so from what point of authority can I speak? I’m an analytical nerd, a researcher and a learner, a computer “scientist” with a foundation of engineering which came with a solid dose of physics and chemistry. I homebrew a little, but I’m not a great homebrewer. I talk to real brewers and bother them for knowledge when I can. I read… reading is useful. I’d recommend two books out of those I’ve read: the imperfect Cellarmanship and the deeply technical Principles of Brewing Science. I’m a CAMRA “cellarperson” and care for cask ale at festivals, I’ve done the week-long CAMRA Bar Management training at GBBF under the guidance of many key CAMRA technical folk and the direct tutelage of Buster Grant from Brecon Brewing (a Heriot Watt educated brewer). I’ve recently done the 2-day Beer Academy “advanced” course with the excellent master brewer Derek Prentice. I deal increasingly with brewers and breweries – buying/managing beer for CAMRA and now as part of my own small beer distribution business.

The best from-the-brewer’s-mouth advice on the topic I know of on the web comes from Justin of Moor Beer. Justin sells his beer unfined… and prefers a slightly different method to the usual (vent first, tap later), but much of his advice applies just as well to fined beers and either venting/tapping order: Cellar Management Tips – if anyone can point me to similar quality advice published by other brewers I would be grateful. There is – but it pushes this “redistribute the finings” idea which I find dubious (see update above on when rousing my be a good/bad idea), and I have trouble trusting any source of information presented in such an eye-watering colour scheme. UPDATE 2014-05-13: This is a thorough guide to cellarmanship by the respected Mark Dorber, Ed’s teacher and boss: Cellarmanship & Real Ale – which is good, but still not entirely in agreement with brewing professionals I know (conflict around yeast rousing and where/when secondary occurs remains).

I can take as good as I give… I’m a scientist not a believer, so if you have some evidence to say any of my myths are not myths I will consider it. It is a complicated subject and my own education is far from complete.

Update 14-05-2014: Much debate ensued, but it veered far and wide off-topic. Most of this is lost in Twitter (for the better), but some debate is recorded on Boak & Bailey’s blog: TRADITION AND SCIENCE IN THE PUB CELLAR