External calandria retrofit

[sbradt]

Does anyone have experience with retrofitting a gas fired kettle to be heated with an external calandria? I'm looking at refitting a 35 Bbl kettle and I'm curious to any pros or cons to this kind of system.

[Moonlight]

The Merlin system looks really cool (hot), but it requires steam.

[arcangel]

Does anyone have experience with retrofitting a gas fired kettle to be heated with an external calandria? I'm looking at refitting a 35 Bbl kettle and I'm curious to any pros or cons to this kind of system.

Hi,

The advantages of external calandria:
1. Rousing boil, recirculation rates of roughly 10 per hour
2. Good protein breaks
3. Whole hops help to keep the calandria clean!
4. No need to enter copper to clean caramelised surfaces unless you like saunas!
5. Well designed calandrias can be cleaned in minutes.
6. Gas fired calandria with ceramic burners and modulating burner give excellent control.
7. Low hold up - typically a calandria with a 14 bbl (US) copper would contain 1 bbl (US) of wort.
8. Versatile - can be attached to any vessel and can be used to heat brewing water (I was going to say 'liquor', but I'm not sure that is the right expression!).

I would not advise steam calandrias. A well designed gas fired calandria can operate with a combustion efficiency of 85% and a flue gas temperature of 280 deg F. I make them.

I hope this helps.

Arcangel

[crassbrauer]

Steam or pressurized hot water calandrias are the only ones suitable for serious brewing. External calandrias are highly advantageous because their surface area is not limited to having to fit inside the wort kettle, therefore given the larger surface area, the steam can be slightly above the temperature of the boiling wort (100 C at sea level), greatly reducing thermal stress and therefore increasing the flavor stability of the finished beer.

[crassbrauer]

By the way, the Merlin system is a thin film evaporator, not a calandria.

[Moonlight]

Arcangel, I want to know more about what you make.

[arcangel]

Steam or pressurized hot water calandrias are the only ones suitable for serious brewing. External calandrias are highly advantageous because their surface area is not limited to having to fit inside the wort kettle, therefore given the larger surface area, the steam can be slightly above the temperature of the boiling wort (100 C at sea level), greatly reducing thermal stress and therefore increasing the flavor stability of the finished beer.

Hi Crassbrauer,

The success behind direct fired calandrias is to avoid flame impingement and manage the heat input over the surface area. A well designed gas direct fired calandria can operate with a flue gas temperature of 140 deg C (~280 deg F.

If you really want to avoid thermal stress, why not maintain a low evaporation rate and vacuum strip.

My market, at the moment, ranges up to 60 bbl (US) or 70 hl.

I have extensive experience of steam raising: it is an art/science and to carry out well is another job that demands another employee.

I look forward to exchanging experiences.

Kind regards
Arcangel

[arcangel]

Arcangel, I want to know more about what you make.

Hi Moonlight,

I design and manufacture:

1. Vapour condensers: put one of those on your copper and assuming (?) a 10% evaporation, i.e. 1bbl in 10bbl, you can recover 70% of the brew length as hot water at 200deg F.

2. Direct fired calandrias. So many people use electric elements, direct fired heated best used for melting tar, or direct fired spiral gas fired heaters. Clean up times are huge. My calandria work at ca 85% efficiency and flue gas temperature is ca 140 deg C (280 def F). Open the manways and hose down. Any deposits, use a brush. Very simple.

Thank you for your interest in my business.

Please let me know if I can help further.

Kind regards


Michael George
Arcangel

[crassbrauer]

It doesn't matter whether the boiling time is reduced or vacuum expansion is carried out after the whirlpool/hot trub removal. The thermal stress to the wort is caused by the difference between the medium used to heat the wort (usually steam or hot water) and the wort itself. Avoiding the excessive creation of Maillard products (produced through the application of heat in the presence of sugars and amino acids, like caramel or bread crust) is the primary goal of boiling wort not only for the benefit of the beer itself but also to reduce the frequency with which one must CIP the external calandria. This has been thoroughly investigated here in Germany, but I found an article in English for you to read:

Andrews. J., Axcell, B. Wort Boiling – Evaporating the Myths of the Past. MBAA TQ vol. 40, no. 4 (2003); pp. 249–254.

By the way, vacuum expansion has been shown not to be as effective at reducing DMS and other unwanted wort volatiles as atmospheric boiling, because the concentration of the volatiles in the gaseous state determine how much will leave the liquid state. With atmospheric boiling they're always exiting, and therefore the concentration is always relatively low above the wort; with vacuum expansion the concentration above the wort reaches an equilibrium with that in the wort and no more will leave.

[beertje46]

Who manufactures external steam calandria for retrofit?

[crassbrauer]

What size are you looking for?

[beertje46]

What size are you looking for?

30 US bbl.

[sbradt]

crossbrauer, I'm looking in the 35 bbl range

[crassbrauer]

Check with some of the small brewing equipment manufacturers. If you or they need technical data, I can be reached at texbrew@gmx.net for regular correspondence. I would actually like to be involved in a project to create these for smaller breweries, since they are quite practical for reasons already stated in this thread, primarily that low evaporation rates and minimal fouling are possible with them, which increases the flavor stability, and the overall flavor as well, of your beer. There is a size issue, however. I know that with a ten bbl system an external calandria is superfluous, because the surface area to volume ratio of the heated surfaces to the wort volume are sufficient to supply enough heat homogeneously to the wort. I'd have to look into whether it would be worthwhile to do it for a 30 - 35 bbl system, but since volume increases by one more dimension than surface area, it doesn't take much of an increase for it to be. I think the smallest I've seen is perhaps for a 50 hl wort kettle.

[arcangel]

It doesn't matter whether the boiling time is reduced or vacuum expansion is carried out after the whirlpool/hot trub removal. The thermal stress to the wort is caused by the difference between the medium used to heat the wort (usually steam or hot water) and the wort itself. Avoiding the excessive creation of Maillard products (produced through the application of heat in the presence of sugars and amino acids, like caramel or bread crust) is the primary goal of boiling wort not only for the benefit of the beer itself but also to reduce the frequency with which one must CIP the external calandria. This has been thoroughly investigated here in Germany, but I found an article in English for you to read:

Andrews. J., Axcell, B. Wort Boiling – Evaporating the Myths of the Past. MBAA TQ vol. 40, no. 4 (2003); pp. 249–254.

By the way, vacuum expansion has been shown not to be as effective at reducing DMS and other unwanted wort volatiles as atmospheric boiling, because the concentration of the volatiles in the gaseous state determine how much will leave the liquid state. With atmospheric boiling they're always exiting, and therefore the concentration is always relatively low above the wort; with vacuum expansion the concentration above the wort reaches an equilibrium with that in the wort and no more will leave.

Hi Crassbrauer

Funnily enough I have the article you kindly suggested I read in my library. This time I read it more carefully.

Do you remember the time you calibrated a thermometer in your physics class? I remember putting the thermometer in a beaker of boiling water and heating it with a bunsen burner. Even increasing the heat input did not change the temperature at which the water boiled. So it is with wort in a copper. In fact the author of the article you recommended accepted that if fouling occurred and the steam pressure, and accordingly the temperature, were raised to achieve the same evaporation rate, the wort would (provided other conditions had not changed) boil at the same temperature.

This is not a case of just heating an object using a heat source without a change of state, in which case the object would rise in temperature to roughly that of the heat source. In wort boiling any 'excess temperature/heat' produces a change of state, from liquid to vapour, that absorbs that heat (latent heat of vaporisation).

Where problems arise is where heat input is so great that nucleate boiling (bubbles of vapour in liquid) goes to the transitional state where a blanket of vapour between wort and the calandria tubes is formed that impairs heat transfer. As a consequence tube temperatures rise. When this happens undesirable high temperature reactions occur, such as you mention, the Maillard reaction.

In direct fired calandrias, the secret is to avoid flame impingement and design for maximum recirculation rates.

Regards
Arcangel

[crassbrauer]

Sorry, I've been away for a while...

Yes, the wort boils at the same temperature regardless of how much heat is transferred to it, provided the pressure in the calandria does not change, because a phase change is involved, and the energy directed into the liquid causes the wort components which possess a boiling point at 100 °C or below to enter the gaseous state. (Actually, due to the wort being pumped into the calandria, the pressure does increase slightly inside of it and therefore the boiling point does, as well.) Since water is the primary component of wort, the boiling point and enthalpy required to achieve boiling is similar to that of water. And it is true what you say about nucleate boiling and two-phase flow and that it is important that a blanket of vapor should not appear between the liquid and the heat transfer surface; otherwise, as you stated, this impairs heat transfer, and the result is a substantial increase in Maillard products. However, external calandria often exhibit single phase flow in the heating tubes due to the increased pressure in the tubes; a portion of the wort then turns to vapor upon exiting the tubes again.

This is all necessary for understanding the process. Nevertheless, an important point not to be glossed over is that wort is made up of over 3000 compounds, which interact with one another during the brewing process. The distilled water you used to calibrate your thermometer is pretty much 100% H20. Therefore, the two cannot really be compared, at least from the standpoint of fouling, thermal stress to the wort, the formation of Maillard products, keeping coagulable nitrogen within the desired range and evaporative efficiency. Another factor to consider is how the heat transfer occurs across the tube wall of the calandria.

What is relevant on the wort side is how much the components that make up the wort are stressed and altered during the heating process. Pretty much any sort of directed-fired boiling system is a thing of the past for modern breweries, because increased flavor stability through reduction of thermal stress has become a key issue over the last few decades. This is one reason why decoction is rarely practiced anymore (its high energy requirement, well-modified malt and hot-side aeration are others). There are, on the other hand, breweries for which flavor stability plays a secondary role. I can name a few who still have direct-fired kettles, but usually this is for one of two reasons: lack of funds or tradition. Believe it or not, according to a book about brewing from 1898 written by J. Thausing, brewers on the Continent were some of the last to give up their direct-fired kettles, after British and American brewers had already adopted steam usage. That’s why you’ll still see a few with old direct-fired kettles here. Continental brewers are very conservative and take a while to accept new technology, but they finally did take up steam heating because they recognized its benefits.

Evaporative efficiency is described as the amount a certain indicator in the wort is reduced through boiling over the total evaporation. Total evaporation alone is not a good indicator of an efficient boiling system. Usually for comparing boiling systems a substance that is not produced during boiling is used as an indicator, like hexanal. But for a brewery’s laboratory to be able to keep a handle on what its own boiling systems is doing, three values are often used to judge whether the evaporative efficiency is adequate:
1. increase in the thermal stress
2. DMS
3. coagulable nitrogen

The goal of a good boil in a modern brewery is to have a low rate of evaporation (3 – 6%) but to rid the wort of unwanted volatiles while isomerizing hops, etc., etc. Direct-fired kettles are good at driving the free DMS out of the wort, provided the boil kicks really well. However, the primary benefit of steam heating is that thermal stress on the wort is reduced and the coagulable nitrogen is treated in a gentler manner. Thermal stress is easily testable using thiobarbituric acid. If this number is high (depends on beer style) this is an indication that a large amount of Maillard products are present in the wort. These are all associated with distinct off-flavors in the finished beer. If the beer is drunk within a week in a brewpub this will hopefully not be a huge problem unless it’s extreme, and for brewers who make dark beers, Maillard products (melanoidins) are desired but they most often come from using dark malt (i.e. only a small amount should be created in the wort kettle, unless you’re brewing something like “stone beer”).

The precipitation of proteins through excessive heat needs to be monitored, because too much results in poor foam stability and an empty-tasting beer. Excessive protein coagulation also results in a loss of trace minerals, vitamins, etc. necessary for a healthy fermentation. Protein coagulation is influenced by the temperature of the heating medium and the intensity of wort boiling. The article I recommended used the EBC Manual of Good Practice as a source, and it states: “The more steam bubbles that are formed, the more intensive the boil, the more proteins are enriched at the gas/liquid interface. This results in a higher coagulation rate. Pumping and re-circulating wort increases turbulence.” You mentioned that because your heating medium is quite hot that you have to “design for maximum recirculation rates”. This creates a problem, because a slowing of the wort recirculation in order to protect important wort compounds from sheering forces is not possible. External calandria are already known for their damaging sheering forces and maximizing circulation makes it even worse – aside from coagulable nitrogen, another example is hot break formation. The lipids bound to proteins in hot break material desperately need to be taken out of the wort. Destroying these flocs through violent circulation can result in off-flavors in the finished beer. This is often a problem with forced circulation unless it is done gently. This is why, for example, Steinecker when designing their “Stromboli” internal calandria, use a gentle pumping system with no sharp angles and “low steam temperatures [to] prevent any thermal load on the wort”.

Another reason to reduce the flow rate through the calandria is that the higher back pressure caused by a rapid flow rate results in the pressure rising in the tubes, which raises the boiling point, which in turn leads to more thermal stress to the wort. Obviously, thermal load on the wort should be avoided. One way is to reduce the time the wort is exposed to the heat; another way is to reduce the temperature difference between the heating medium and the wort. You must remember that not only is the wort undergoing a phase change, but normally the steam on the other side is, as well. It is close to the same temperature of the wort and is giving up its latent heat (heat of vaporization), which is similar to what wort needs to boil. This heat transfer takes place over the surface of the tubes in the calandria (the steam undergoes “film” or “drip” condensation depending upon the conditions in the calandria). The overall heat transfer coefficient and the surface area of the transfer surface (known together as “heat flux”) are factors which determine the rate at which fouling of this surface on the wort side occurs (i.e. Maillard production). There are several factors influencing the overall heat transfer coefficient; one of them is the differential temperature across the tube wall between the heating medium and the wort; another is the enthalpy of the heating medium itself. The article I recommended quotes the EBC Manual of Good Practice, which states that “a high temperature at the boiling surface leads to a loss of foam-positive proteins and also to a darker wort due to Maillard reaction products. These also change the flavor and impair the flavor stability of the finished beer.” The article also states: “Very-large-surface-area external wort heaters operated as thermosyphons and, using very-low-temperature steam, provided an effective method of wort boiling” and that the “beers [brewed on this system] did differ [from the control]; this was seen to be positive. The test beer had better flavor stability and a lower DMS level.”

Therefore, equipment manufacturers supplying breweries with brewhouse equipment should consider these points. As already mentioned, the EBC Manual of Good Practice is one good place to look for information on this subject in English.

Cheers,
Crass

[sbradt]

“Very-large-surface-area external wort heaters operated as thermosyphons and, using very-low-temperature steam, provided an effective method of wort boiling”

Crass,

Very interesting observations. Would you care to offer an opinion on what constitutes "very large surface area"? I've been told that I should be looking for at least 75 sq feet for a 35 Bbl kettle (about .17 sq m/hl). I'm also curious as to what constitutes "very low temperature steam". I assume that, in the interest of reducing sheer forces, one wants the tubes large enough to avoid excessive velocity of flow. What kind of velocity of flow is appropriate?

Thanks, I'd like to get this right the first time and I'm afraid that I'm not an engineer.

[crassbrauer]

There are a lot of factors to consider here. For a standard external calandria with an evaporation rate of 3 - 5% in which the total wort volume circulates around 8 times per hour and a steam temperature of 110 C, I'd say you'd need (very ballpark-ish) around 5 - 6 tubes in your calandria with a total area of around 15 to 20 m2 (you'd have a few 180 degree turns). I know that's a lot, but they did say they were large. I'm eyeballing some formulas here without doing a whole lot of calculating, so I hope I considered everything. You could even try circulating less. A steam temperature of 105 C would make the necessary area double, i.e 45 m2 or so. Hope that's helpful...