Wednesday, 10 August 2011

Cool Snake

If you follow me on twitter you might occasionally see me arguing with Tandleman over various issues. In many ways this is a bit bizarre because I have on more than one occasion had a very enjoyable sociable experience drinking with someone with the name Peter Alexander. Recent discussions have been over naming and shaming individual establishments over their beer serve temperature. Although I remain unconvinced about the value of public naming and shaming on twitter, I agree that there are some shocking examples of poor beer dispense.

Beer should be served at the correct temperature. For cask beer this is generally considered as being around 12 degrees centigrade1. Too cold and flavours can be masked as well as inevitable problems with chill haze. Much worse is the crime of serving beer too warm; the drinking experience is not enjoyable and an otherwise well brewed beer can be ruined as a result.

There is unfortunately a pattern; tied pub estates often have significant care given to the quality of installed equipment. OK, the beer range may not be particularly varied or imaginative, but you can be more certain that the beer is better looked after. There is very good reason for this.

With a tied house the beer sold is entirely supplied through one route to market. It might be a single regional brewer or it may be a PubCo but there is at least a central purchasing route and maximising sales is crucial to the profitability of the owning business. Cellar support is inevitably very good with great care given to maximising the quality of the beer.

By contrast many free houses have poorer investment in cellar equipment and dispense technology. A very noticeable but in my view completely unacceptable omission is, as I think Tandleman put it, "python2 cooling to the point of dispense". This should include jacketed handpulls and carefully regulated circulating cooling water from a dedicated cooler, i.e. NOT from the keg cooling circuit.

OK, the investment might well be significant for a free-house without benefit of a large brewery to supply the technical support and equipment investment, but it will reduce wastage in "pulled through" beer and also inevitably increase the quality of the beer therefore improving the customer experience. I suspect the payback time will be very much shorter than most establishments expect.

Moreover, some of my very favourite beer outlets do suffer from beer serve temperature problems. This results in public naming and shaming of the very places I love. So, perhaps these places could help to prevent Tandleman and I from falling out by thinking about investing a little in dispense equipment?

On Beer, Birra, Bier there is an interesting reflective post on my most recent twitter discussion with Tandleman.

Expensive but very good handpulls can be bought here. Pythons and other such wonderful things can be bought here. Personally I find the technicalities of putting these things together very straightforward, but if your practical skills aren't up to it a good cellar technician shouldn't cost too much.

If any cellar technician tries to tell you that it's OK, it's "trad beer" and doesn't need python cooling, look for someone else.


1personally I think there is an argument for some very light and hoppy beers being served a little cooler and things like strong stouts and barley wines a little warmer. 12 degrees is a good compromise however and I totally reject the excuse some dinosaur cellar-men use to say cask beer should be warm. Cask Marque is one organisation that has set some parameters and this cannot be a bad thing.

2A python is a thermally lagged bundle of pipes that includes a flow and return cooling circuit. It's really good at keeping beer at the right temperature from the cellar right to your glass. Generally, if beer is too warm, it is highly likely that this technology is not in use, or it's broken.


Eddie86 said...

I personally don't like them from a barman's point of view. Because of the narrow lines, the 1 I had installed as a trial needed a flojet.

The end result was the beer would come out of the pump at the same speed, regardless of the speed you pulled the pump.

On traditional lines, you have more of a 'feel' for the rate of flow, making topping up easier and resulting in less ullage.

HardKnott Dave said...

Ed, sounds like you didn't have a check valve installed. Trust me, a properly installed system is much better than the traditional system.

Owen said...

There's absolutely no need for a flojet, and assuming the cask is below the pump, no need for a check valve either.

It's possible that the one given to you as a trial used a micro bore line that's only suitable for gas dispense. There's no reason why you couldn't have the same line in the python as you use without, with no difference other than the temperature of the beer coming out.

Eddie86 said...

Sounds like I'll go back to Marstons and talk to them again - they offered me this equipment. The one I trialled before was a couple of years ago from Brains.

I know the handpumps have got the water jackets around them with the separate cooler in the cellar, just need it all plumbing in. Looks like my day off on Friday could be fun...

HardKnott Dave said...


If the python is a large bore version then I'd totally agree that a FloJet is not required and neither is a check valve.

However, FloJet pumps are regularly used in conjunction with micro-bore pythons in many up-to-date installations. Micro-bore is not just for gas dispense and it saves significantly on beer wastage and installation space.

Moreover, micro-bore pythons are more readily available off-the-peg. Large bore pythons are generally, in my experience, made by the cellar technician as an afterthought when bits are left over from keg install. These wide bore installation fail to have the same performance, in my view, as the micro-bore systems.

HardKnott Dave said...

Eddie, if only you were closer I'd come and give you a hand. In fact, perhaps we can come to an arrangement, I think we still need to get casks back to you at some point.

Eddie86 said...

And I need to get some back to you - off out now but I'll catch you on twitter shortly

StringersBeer said...

It makes you wonder how people managed to get a decent pint in the olden days, without all this technology, eh?

Barm said...

In the olden days nearly everyone was drinking the cask beer and there was only one kind of it, or two if you were lucky.

Brewers Union Local 180 said...

I'm trying to remember ... were you using a separate and independent chiller for the python cooler lines and cask jackets back at the ol' Woolypack? And was it glycol or water?

Tandleman said...

Stringers: Short lines, bar and cellar people who knew what they were doing, heat exchange via wetting, cellar runs if need be. More turnover, less beers on sale, etc etc.

There were many ways and of course we just got a lot of warm beer. No excuse now and with short runs and good cellar practice plus turnover, cellar cooling is just fine, but it costs of course. Maybe you wouldn't be surprised at those that turn it off, but many might be.

Best idea is to factor the cost of cooling into the pint and then you are paying for what you get.

StringersBeer said...

Increased choice not unalloyed good thing shock!

Dave's right of course, handling a low(er) turnover, temperature sensitive food product needs special care and equipment. From the cask (keg), to the pump (tap), if it's not flowing - it's effectively stored (even in the line) - and should really be at the recommended storage temp. Ask your EHO about it.

HardKnott Dave said...

BUL180, all water; there is no need for glycol when all you need to do is maintain around 12 degrees centigrade. The same remote cooler in the cellar can handle the python and jackets if it is all done right. However, sometimes it is necessary for reasons of practicability to have more than one cooler, normally where a system is being upgraded.

Tandy and Stringers, choice may well have it's problems, but without a desire for choice most of the smaller breweries would not be able to exist.

Tyson said...

Couldn't agree more, Dave. I'm lucky in that my local-not a freehouse, but given wide discretion by the pubco in all matters, has micro-bore pythons. They work great and are a vast improvement on previous systems.

Tandleman said...

I was giving reasons, not agreeing with the premise. I agree with Dave too.

HardKnott Dave said...

Tandy, I don't think you needed to defend yourself there, it's just a fact. I also note Barm makes the same point, sorry I didn't acknowledge.

StringersBeer said...

Sure Dave, clearly, choice is a good thing - overall. But as has been pointed out, scaling up the offering on the bar won't be a simple matter of bolting on a few more pumps and threading some tubing around. You're right about the cooling, of course. For that matter, 4 9s take up much more room in the cellar (particularly on stillage) than a barrel would have.

I'm pretty sure that when a pub is designed from scratch nowadays, the traditional cellar, and long runs of manky tube, aren't in the picture.

Saga Of Nails said...

Twelve degrees is at the highest end of what temperature a cellar should be. Certainly in the warmer months I tend to put the cooling closer to nine degrees than twelve and I think that a lot of other landlords do as well. Remember that ale keeps fresh for longer at ten degrees than it does at fifteen degrees, and of course lager will start to fob loads beyond that temperatures, so I cannot imagine any intelligent landlord actually turning his cellar cooling off. But we do have some pretty stupid landlords in this country.

Of course an awful lot of pubs have vintage beer engines built into the bar and don't really want to replace them with less attractive modern ones. The last pub that I ran had beer engines which were installed in 1947, and I would never have changed them for more modern ones, as long as they were still working. I am lucky that I have only ever ran pubs with cellars directly below the bar, as there is only one metre of line that allows the ale to warm up, in these pubs.

I do find it strange that you find the omission of 'python cooling to the point of dispense. This should include jacketed handpulls and carefully regulated circulating cooling water from a dedicated cooler.' unacceptable in a freehouse. In a pub that has four handpulls and you are serving say twenty pints an hour, why would you need this? It seems an unnecessary expense.
Morning/evening pullthrough is more about temperature, it is about oxygen getting to the ale as well, and no amount of cooling is ever going to fully eliminate this.
I do agree that some pubs with long lines NEED this kind of system. My friend has just started working in a pub near Bath, and following an extensive million pound refit the ale lines are now over fifty foot long. (He may have even said fifty metres, which is mind boggling.)

StringersBeer said...

Of course cellars are themselves expensive to construct, which is why in non-traditional pubs we often find "cellars" at ground level, often backing on to the bar. We also come across uncooled cellars (with cask/keg cooling setups), in these places the line length actually in the cellar is also subject to heat pick-up. It's perfectly possible (as S.O.N says) to use over-cooled beer to offset the temperature rise on the way to the glass, but Dave's suggesting that there's away that's better, & probably cheaper to run.

If it works - don't fix it I guess, but there are a few places I've seen where the first step would be to shut the sodding cellar door, rather than trying to cool the entire universe down to 12°C

Owen said...

Chilling the beer at the pump isn't in any way ideal as condition will have been lost and can't be regained by cooling; the only thing you'll achieve is cold flat beer rather than warm flat beer.

The beer needs to be at serving temperature (or only just below) from cask to glass, and the only way to achieve that (bar very short runs) is to have the cellar at the right temperature and to have cooling all the way along the line, ie with a python.

StringersBeer said...

The Professor suggests that, if you've got a floject on a cask line and the pressure is affecting the normal operation of a hand pull you might want a flow control valve rather than a check valve (although you might need one of those if you dribble). But it's all theory with the Prof...

HardKnott Dave said...

Stringers, I would wonder where the Prof gets his information from. The flojet will create a fixed(ish) head of pressure to the input of the handpull, effectively making the situation similar to having the cask above the level of the bar. Without a check valve the beer will flow all the time even if the handpull isn't being operated.

I think the head of pressure is proportional to the gas supply to the flojet. The check valve works by sensing the slight depressed pressure relative to atmosphere created by the handpull on operation, whereby the check valve opens allowing beer to flow.

A flow control valve is only a restriction and will not shut to stop the flow. The flow control valve is not required if the gas pressure(s) on the flojet(s) are set correctly.

Flow controllers are absolutely invaluable on keg beer installations however, and can be used to slow the delivery rate when keg beer fobs too much.

StringersBeer said...

floject = flojet of course.
You've still got all that "lone voice" link spam Dave.

HardKnott Dave said...

Hmmmm, have removed the "stuff that links here" widget. Not sure I'm happy that this appears to be the only way to get rid of such spam.

StringersBeer said...

I asked the Professor about this, and while he didn't exactly look sheepish, I could tell he was thinking on his feet. Then he brightened,
"See here, no dribbling & no checkvalve means no pressure, so this system must be exactly, or somewhat under balance, which is exactly what you'd hope for"

"Except", he went on, "this probably means the handpump resistance is dependent on the flow-rate achieved by the flojet. Which with these settings is likely to be low. Or possibly too high. Either way, probably the cause of the user-interface issue reported!"

"Clearly", his attitude all confidence now, "It should be should be possible to set a reasonable value for handpump resistance without going down to the cellar! A flow restrictor would be just the ticket!"

I wasn't 100% convinced, "But Professor, surely this would only work if the system was producing overpressure at the pump, in which case you'd also need a checkvalve just like Dave said"

He said something about needing some gas, and went back to his workroom. Personally, I suspect he wasn't sure about the problem Eddie86 was having. I'm sure we'll hear from the Prof again. After some experiments.

P.S. Yes it's rubbish that there doesn't seem to be a way to "moderate" backlinks with blogger.

Anonymous said...


I went through all this 6 years ago. Built a dedicated ale cellar underneath the hand pumps, and installed aircon there to keep it at 12C. Was unable to obtain a large bore python anywhere so constructed our own with 12mm food grade hose.

Ran 6 lines round a central cooling core and insulated it all with a rubber pipe sock and duct tape. Obtained a u-shaped piece of stainless pipe to return the cooling water at the bottom of the python.

Purchased 6 water jacked beer engines and fitted non-retrurn valves under the intakes. Connected the cooling jackets in series to one end of python's core.

Hardest part was to source a python cooler with a good strong pump. There was only one on the market at that time. These days can still only find the same one on the market by now they have a minimum order of five. This means I'm maintaining a cooler that should really have been replaced this year, not very cost effective :(

The hardest part to crack was between the python and the hand pumps. It necessary to break the lines out of the python about 2 feet before the pumps in order to create sufficient space to connect the lines. With a 12mm hose this meant half a pint sitting in the line, and even with good insulation that's going to warm up in no time. What I did was so sleeve this section down to 6mm line, by slipping a 6mm food grade hose inside the line (just fitted) and melding them at each end of the section with jubilee clips. This reduced the volume of that section to 1/4 pint. The narrow hose sits about 2" deep into the line in order for the line to connect to the pump as per usual. This means, worst case scenario, 3/4 of a pint at 12C and 1/4 of a pint at 22C, equating to a temp range of 12-14C depending on when the pint is poured, which is bang within the recommended range.

The only other thing I suggest is that the first pull out of each line every day will have been sitting oxidising in the pump over night. We fit an elastic band round the top of each pump hand. If the band is at the top of the handle then we throw away the first 2 pulls of the first pint ordered, then move the band down to the bottom of the handle, signifying that the stale beer has been evacuated from that pump, and its now safe to serve for the rest of the day.

Walk in the Light


camelia said...


We bumped into your blog and we really liked it.
We would like to add it to the

We would be delighted if you could add your blog to Petitchef so that our users can, as us,
enjoy it.

Petitchef is a french based Cooking recipes Portal. Several hundred Blogs are already members
and benefit from their exposure on

To add your site to the Petitchef family you can use or just go to and click on "Add your site"

Best regards,

Confetteria Filarmonica said...

I always wonder the best temperature serve of my cask ale as i have found that every pubs has a different one. Hope pub owner are picky on that in order to get customers like me happy!