This year I’ve decided to take a more experimental approach to cheese making, and during my last experiment into salt quantity, I noticed that the milk coagulation time was about double what I was expecting, taking 3 hours rather than the anticipated 1.5 hours.
At the time, I put it down to a combination of low ambient temperature, and possibly using slightly too little rennet, and resolved to increase the quantity substantially in my next make.
I thought it might be interesting to perform a direct comparison between cheeses which have been coagulated using a variety of different quantities of rennet – so keeping the milk, salt and maturation procedure the same, and just altering one variable.
I was interested in seeing whether the perceived wisdom of long slow sets, advocated by everyone from Dora Saker to Stichelton Dairy, would have much of an impact on a relatively short-aged cheese, and how it might effect yield and texture.
My understanding from various cheese making books is that using too much rennet, results in a dry textured cheese, caused by the casein bonds which trap the milk constituents forming too quickly, leaving the whey cloudy, and the cheese lacking moisture.
Too long a set can result in weak curd due to a lack of available casein bonds, and the potential for undesirable reactions taking over, which without careful acidity monitoring (for example, when making lactic cheese) can result in bad flavours.
As with the salt experiments, I wanted to be able to compare the resulting cheeses side by side to see whether there was a noticeable flavour and texture difference caused by different quantities of rennet.
Gastronorm to the rescue!
The ability to compare multiple cheeses side by side presented bit of a problem, as it meant I had to find a way of maintaining a number of identical isolated vats at the same temperature at the same time, some of which would set earlier than others due to the nature of the experiment.
Browsing around Nisbets, which is a constant source of inspiration and temptation, I found a bain-marie, which allowed six 1/6 size Gastronorm pans to be suspended in a large vat of water.
Measuring my existing vat, the fit looked like it was going to be tight, but I took the plunge anyway, buying six pans, each with a volume of 2.2l, and two dividing bars for them to rest on.
Luckily the pans fit perfectly, and after filling the vat (now a double bain-marie) up with water, I was ready to begin!
I found this raw milk at Blackheath Farmers Market, which is from a herd of Ayrshire and Guernsey cows.
Milk from mixed herds have never struck me as such a good idea, since I would assume the different milk properties would cause an inconsistent curd.
Ayrshire and Guernsey is an interesting mix though – Ayrshire is generally considered a perfect cheese milk due to high casein concentration and relatively small fat molecule size, whereas Guernsey is less appropriate due to it’s large fat molecule size.
Mixed together, I thought the high casein concentration might be able to trap enough of the Guernsey fat to add some intricate flavour to the cheese.
I think now I’ve figured out how to do multiple makes at the same time with my divided vat, milk type is a natural future experiment!
I considered starting the vats which would receive higher quantities of rennet earlier so all subsequent actions would be in sync, but felt a few hours difference in salting and maturation probably wouldn’t impact the results too much.
The vats were numbered from top left to bottom right, moving left to right, and renneted 30 minutes after adding a tiny amount of DVI starter, and geotrichum candidum.
Each individual vat contained 2,080 – 2,090 ml of milk – I’m not sure where the 2.2l quoted on the item description came from, but this was definitely the most each could handle without overflowing!
I decided to vary the rennet quantity from 0.02% to 0.045%, as most recipes recommend the lower end of this scale.
Rather than trying to make up each solution individually, I diluted the total required quantity of rennet (4.062ml) in 3 times the quantity of water (12.18ml), although to account for wastage during measurement, I actually used slightly larger quantities (although in the same proportion) of 4.2ml rennet and 12.6ml water.
Rennet solution, from left to right: 1.668ml, 2.084ml, 2.500ml, 2.916ml (two syringes), 3.332ml (two syringes), 3.744ml (two syringes).
It was only at renneting time I realised that the temperature had increased quite substantially from starting.
Adding another layer of water seems to introduce a substantial lag which resulted in the milk temperature raising to a 36C peak in one of the vats.When using the vat normally (i.e. not as a double bain-marie), heat is transferred incredibly quickly from the vat to the milk, allowing easy control of the milk temperature.
Interestingly, it seems the two vats closest to the front of the container vat seemed to be a fair bit hotter than the other four vats – I’m not sure exactly why this is, although at a guess it may down to the positioning of the heating element within the container vat.
Over the next 210 minutes, I checked the temperature and pH of each vat at regular intervals, as well as checking for a clean break (shown in column “Cut?”).
The first observation is that vat 1 & 2, which had the lowest quantity of rennet (0.02% and 0.025% respectively) took a substantially longer time to show a clean break than vat 3 (0.03%).
Secondly, increased quantities of rennet seem to cause a sharper drop in pH – vat 6 (0.045%) dropped from 6.4 to 6.2, whereas vat 1 (0.02%) didn’t really decrease from pH 6.3.
This makes sense, since rennet is an acid, and pH measures the acidity of a solution.
One thing this did highlight was that I may need to find a pH meter which has a higher resolution than one decimal place, and perhaps see whether there’s some way I can increase the sample rate to get a better idea of how it changes over time.
I ladelled out the curd as a clean break emerged on each, keeping the moulds in the same formation (separated into two sets of six) so I could keep track of each rennet quantity once out of the vat.
After draining overnight I noticed something strange which I didn’t expect – the curd coagulated with a lower quantity of rennet had drained far more than the higher ones.
In the picture above, the right hand vat is number 2 (0.025%), the middle is 4 (0.035%) and the left hand vat number 6 (0.045%).
At a guess, I would say this result is due to the weakness of the casein bonds, allowing liquid to escape more quickly from the curd.
After draining for a couple of days at 17C @ 60% humidity, the cheeses were salted at 3% concentration, using Cornish sea salt, as per the success of the previous salt experiments.
Another couple of days drying, and they were ready to go in the fridge at 12C @ 85% humidity.
After 12 days maturing, they were nice and fluffy with white mould, but weren’t as soft as expected, so I decided to give them another few days to ripen before having a tasting session!