Grapevine & Wine 02/01/2012

Two experience and some suggests for oenology refrigeration

The Australian Wine Research Institute studied the economic impact of the refrigeration on wine cellar operations and considered how to reduce costs without quality' problems

Trials related to improving refrigeration efficiency were performed at a small winery (~500 tonne crush) during and around the 2011 vintage. These trials considered the use of warmer brine temperatures, the operational strategy used to control the cooling system, cooling system maintenance, plant shut-down/infrequent operation, and the use of an external heat exchanger.
Running this winery’s cooling system with a brine temperature of +4°C year-round, instead of -5°C, translates to a 17% reduction in refrigeration related electricity usage (12% reduction in total winery electricity usage). In vintage trials, red wine ferments were adequately controlled using nominally +4°C brine circulated through the existing tank cooling jackets. However, the warmer brine temperature naturally meant a lower cooling rate. This was somewhat problematic for the trial winery whose specific fermentation paradigm involved rapidly cooling the ferment once it had reached a peak temperature. To facilitate this operational strategy, a brine cooled external heat exchanger was introduced to the red wine pump-over’s to provide additional cooling surface area. At many wineries the reduced cooling rates obtained during fermentation may not be a major limitation to using warmer brine, however operations like must chilling, cold settling and cold stabilisation could be unless alternative strategies are employed. As a general rule of thumb, warmer brine temperature should be used for as much of the time as practicable. This can be partly facilitated by scheduling operations that require very low brine temperatures to occur over the same period. Technologies that negate the need for very cold brine temperatures (e.g. flotation for white juice clarification and alternative cold stabilisation techniques) may enable some wineries to operate with warmer brine temperatures year-round.
The cooling system at the trial winery was functioning sub-optimally as refrigerant had leaked from one of the packaged chiller’s two refrigerant circuits rendering that circuit inoperable; reducing the cooling capacity of the packaged chiller and increasing the risk of total loss of winery cooling. This fault had gone undetected for a significant period of time. If winery staff had kept a log of operational parameters (e.g. number of compressor starts) the fault would have been quickly detected and a service technician could have been contacted. Wineries should establish a good working knowledge of their cooling systems, document correct operational procedures and keep a regular log of basic operational parameters. While improvements as a result of these practices may be difficult to quantify, they are likely to have much more significant economic impact than many other winery refrigeration related changes.
The winery didn’t require cooling for approximately four months each year and therefore shut down the packaged chiller altogether during this period. This strategy was very effective; reducing annual electricity consumption at the winery by approximately 24%. However, switching refrigeration systems off completely may mean that some of the brine freezing point suppressant could evaporate (depending on local weather conditions), partially negating the savings from reduced power consumption. For wineries that do not require cooling for a significant period of time a better strategy may involve changing the brine set-point and hysteresis settings on their cooling system so that the refrigeration plant only runs infrequently. A brine temperature set-point of around 10°C together with an appropriate hysteresis setting to limit starts/stops should achieve this.

Three side-by-side fermentation trials were performed during the 2011 vintage to investigate the influence of white juice/wine fermentation temperature on cooling requirements. The trials identified a reduction in over cooling requirements of between 0 and 8% for ferments controlled at elevated temperatures (e.g. 20°C as opposed to of 16°C). Reductions of this order were consistent with a review of the mechanisms of heat generation and transfer during fermentation. Increased evaporation of ethanol and water was likely a key source of the small reduction in cooling requirements at warmer fermentation temperatures.
For a winery fermenting 30,000 kL per annum it is estimated that the potential electricity savings associated with reduced cooling requirements at warmer fermentation temperatures was approximately $5,000 per annum. Given that annual electricity consumption for a winery of this size costs in the order of $1,000,000 and there remains some unquantified risk of sensory damage; the use of warmer fermentations of the order of 4°C is not likely to be justifiable on the basis of electricity savings alone. However, faster fermentations, resulting from warmer fermentation may assist wineries limited by fermentation tank capacity.
The 58 kL tanks used in these trials were fitted with only one cooling jacket; positioned towards the bottom of the tanks. The location of this jacket sometimes resulted in significant cooling-induced stratification when the brine was flowing through the jacket but the agitator was not on. Stratification did not generally occur when fermentation was actively proceeding as the generation of carbon dioxide induced mixing was sufficient to ensure liquid phase temperature homogeneity.
Poor tank cooling, agitation and temperature measurement configurations have the potential to negatively influence wine quality and consistency, increase energy use, and can result in misleading characterisation of tank temperature on winery monitoring systems, which could in turn result in sub-optimal decision making. Wineries should be aware of this and possibly consider auditing the temperatures at different points in tanks at their site to verify that temperatures reported on winery monitoring systems are accurate.

Source: AWRI - Copies of this study are available for download from or

di Simon Nordestgaard,