It’s that time of year again when winemakers lose sleep worrying about getting their beautifully crafted wines into bottle without ‘stuffing up’ the quality – and praying that there won’t be any re-fermentation or other issues. Vineyard finds out the steps winemakers take to ‘get it right’ before the wine goes into bottle, to ensure that wines reach the customer in the best possible condition.

Preparing for still wine bottling is a significant task in the winery, aside from blending there are critical checks on stability, dissolved gases, filterability and sulphite levels – as well ensuring the correct dry goods are ordered. Some tests or analyses are done in house, others are sent away to the laboratory, but all are part of the overall quality control plan. “Why do we do these tests – well it’s much easier and cheaper to check and rectify any issues when the wine is not in bottle. 

“Tartrate crystals or hazes in the wine are not acceptable and can lead to expensive product recalls – and loss of reputation,” explains Josh Donaghay-Spire, Head Winemaker at Chapel Down Wines, speaking with other winemakers at the recent WineGB webinar on preparing for still wine bottling. “Getting a pre-bottling analysis is, in my view, well worth the cost. This is the last chance the winemaker has to influence the quality of the wine – as once in bottle it’s too late,” added David Cowderoy, Consultant Winemaker and Director of BevTech Ltd. 


Protein, colloidal stability and tartrate stability are all necessary pre-bottling checks. Most winemakers check protein stability using a difference in turbidity before and after a heat test. Typically, 70°C for 30 minutes, followed by refrigeration, but other parameters are used.

“Subtracting the second reading from the first gives the delta NTU (Nephelometer Turbidity Units) and indicates the protein stability – the winemaker’s interpretation and decision will depend on where the wine is likely to be stored and sold. For example, a wine travelling overseas and experiencing temperature changes, will behave differently to a wine being sold locally,” commented Josh. “It’s important to use the same bentonite mix in the trials as the final additions. And if you plan to use CMC (carboxymethyl cellulose) the wine will need to be protein stable,” added David. “The rules are changing for CMC and it can now only be used for whites,” added Sam Lindo, Winemaker at Camel Valley. 

When it comes to filtration at bottling, the wine has to be able to pass through the sterile filter and not cause blockages. “Filtration at bottling is not to remove anything at this stage but as a safeguard, to stop microbes getting into the bottle,” explained David.

Testing for tartrate stability is required to ensure the wine is stable and avoid crystals being deposited when in bottle. “This can be done by traditional cold stabilisation, CMC, metatartaric, KPA (Potassium polyaspartate), electrodialysis – by whatever method, it must be done,” commented Josh. “The cold test involves taking a sample, putting it in a freezer or water bath for a set period of time, and then looking for crystals. Or you can use the conductivity method for a quicker answer – we use an automated set up called CheckStab, which also indicates likely stability over time,” added Josh. “A quick and easy test I do, is the freeze-thaw test – it’s a severe test, but if it passes, it’s a pretty good indicator that your wine is stable,” commented David.

Dissolved gas levels

Dissolved oxygen (DO) is picked up from ullage space, transfers, operations and bottling. “It is important to consider DO pick up and I am surprised by how few wineries have a DO meter,” commented David. “DO pick up during bottling can vary hugely, a good line will limit pick up to around 0.1 mg/l, a mediocre line around 1.0 mg/l, but I have seen poor quality bottling lines result in a DO pick up of 5.0 mg/l. A key step in limiting this is the ability to fill the bottles with CO2.” 

CO2 produced during fermentation will leave the wine near saturation point. “Depending on the style of wine, this may be unacceptably high – no one likes fizzy red,” said Josh. “We test using a Carbodoseur. If levels are too high when bottling at 20°C, the CO2 will start to come out and cause foaming and problems in the filler. If needed, we reduce the dissolved CO2 levels by sparging with nitrogen – which has the added advantage of reducing dissolved oxygen also,” commented Josh. David adds: “I’ve seen a lot of red wines where the dissolved CO2 is too high and it has a really negative impact on the sensory profile – the tannins taste harsher, the acidity tastes higher, less fruit flavour, and simply de-gassing it makes a massive difference.” From Sam’s point of view: “Dissolved CO2 at around 1,000 mg/l makes for an easier life when bottling.”

Free and total SO2 levels 

The reason for adding SO2 to a wine is to achieve microbial stability and protect against oxidation. “A level of 0.6 mg/l molecular is the minimum and 0.8mg/l is the optimum. There are charts available to be able to check this, as levels will depend on pH,” explained David. 

“DO impacts SO2 levels, as 1 mg/l of dissolved oxygen can consume 4mg/l of free SO2. So, knowing the likely pick up at bottling influences the amount of SO2 added when preparing for bottling. Oxygen is also picked up after bottling, depending on the closure used,” he added. 

Making adjustment to SO2 levels should be done in good time explains Josh: “As it will reach a new equilibrium – don’t leave it to the morning of bottling and expect it to be there as free. A few days beforehand check the levels, adjust, check again – then check on the day as well. It’s not a difficult test to do – either by aspiration or with ripper. If you send samples off to a lab, package them carefully with some inert gas, otherwise the readings will be meaningless. The wine has to be legal for sale as there are maximum limits, and the level has to be below this and be aware of how the wine will evolve over time.” 

Hydrogen sulphide

“If you think you have a problem, it is only going to get worse in bottle, so it is best to check beforehand. With good control of fermentation, the use of appropriate yeast strains and yeast nutrition, problems should be avoided – but I still always screen,” advises Josh. If using copper Sam warns: “That it is a blunt instrument and can cause reduction issues down the line and I prefer to use some aeration at the end of ferment.”

Residual sugar

“It’s important to know the level of residual sugar (RS) post-fermentation, and again when doing the blend and making sweetening addition decisions – then to double check again before bottling,” explains Josh. “The RS levels might affect decisions around stability, for example, as it changes the risk matrix, especially with a lower sulphite regime, and with lower alcohol levels,” he added. David sees problems occurring with the origin of the sugar which is often must concentrate or süss reserve. “If süss reserve is added just before bottling it can lead to instability problems. Sometimes RCGM (rectified concentrated grape must) is not filtered well and can lead to filterability problems,” he said. 


Wines that are not properly filtered can cause blockages of the sterile membrane, resulting in costly downtime, and possibly replacements of expensive membranes. Filters can block for various reasons, so, Josh says: “The easiest thing to do is check the filterability index of the wine before bottling – it’s not hard to do.” David advises: “A 0.4 μm nominal depth filter ahead of the 0.45 μm sterile membrane filter, to avoid problems.”

Sterility at bottling

At the end of previous bottling, it is essential that all wine is flushed out, with plenty of water, from the bottling line and filters, or there will be microbial contamination. Then steam before bottling – it is the only sure way of sanitising equipment. If using cleaning chemicals, they have to be in contact with all the intricate parts and complex areas such as nozzles. “We use ATP swabs to check for any residues. It’s also important to do an integrity test on the filter membrane, as it needs to pass this to ensure sterility – it amazes me how many people do not do this,” explained David. “We also send the bottled wine for post-bottling analysis as sterility needs to be achieved for the PDO scheme,” added Josh.

The day of bottling

“Even after much preparation it can still go horribly wrong,” exclaimed Sam. The temperature of the wine, and also the winery or bottling hall should be around 20°C, to ensure smooth running of the line, avoid condensation on the bottles – particularly if labelling when bottling. The bottling line needs to be set up correctly for the size and shape of bottle, and the closures that are to be used. There needs to be enough staff in place, the dry goods checked – to avoid the frustration of running out of closures or boxes.

Pallets of empty bottles can have the odd broken bottle, and breakages can occur during handling, and on the bottling line. The procedures for dealing with broken glass is high on the list for inspectors from supermarkets or any customer. “The first thing to do is have a robust glass breakage procedure. The staff need to be trained so that they know how to handle broken glass. Broken glass can also be found amongst the empty bottles on the pallet, and for this reason we prefer the bottles to be rinsed inside – in fact, it’s all part of a HACCP plan,” commented David.

These winemakers were contributing to a recent WineGB webinar on preparation for still wine bottling. The WineGB webinar recording is available to its members on the WineGB website.