Summary – recipe for set custard
Ingredients (for ~400g = enough for 4 people)
150g semi-skimmed milk
200g double cream
100g egg yolks (6 medium eggs)
1 tsp vanilla essence
30g caster sugar
3 g (1.5 leaves) leaf gelatin
Put all the ingredients apart from the gelatin in the Thermomix, at 80 degrees Centigrade, speed 2, for 7 minutes (or use a saucepan and care not to heat it too hot).
5 minutes from the end, put the Gelatine sheets in cold water
Once finished, stir the gelatine sheets into the custard, put into container you want it to set in, and cool over ice before refrigerating.
A couple of weeks ago we made a custard with our Thermomix (our new highly indulgent kitchen toy). Pretty much everything else we have done has been fab, and this wasn’t – the flavour wasn’t great, and the set wasn’t good either. And, many recipes for custard have loads more sugar in than tastes good at the end (to us), so I decided to suss out a decent custard recipe, both for pouring and set. This post is for me, to make notes as I go
Based on Thermomix cookbook recipe, p 189. 50% recipe, asked for 3 large egg yolks; I used 4 smaller ones. Note that the temperature is 80 degrees rather than the 90 degrees used for the original recipe that I didn’t like – less likely to over-cook the egg.
Tasted lovely when just done. Trial on setting for 200 ml using 1.5 gelatine sheets.
Based on Custard part of trifle recipe in Heston Bulmethal’s ‘Further adventures in search of perfection’, p 294. Target temperature is 75 degrees. Used 80 degrees (in choice between 70 and 80) as it worked pretty well in trial 1. All ingredients added in one go (like trial 1). Used 3 yolks (45g vs. 50g recipe quantity) vs. the 4 in trial 1. Used half fat milk rather than whole. Used vanilla essence rather than saffron (same qty as trial 1). Run at speed 2 to get slightly less air in. Set was with 1g of gelatine sheet (half a sheet) – volume was also lower that trial one, and in fact this is still twice what was in Heston’s recipe (oops!). But, good to try, as approx 50% of what was used in trial 1.
Hot it tasted not only creamier, but also sweeter. Same amount of sugar was included, but liquid volume was only 150ml which probably explains it. So, might be able to drop the sugar by 40% to match trial one. There was less air in both the hot and set versions. Much yellower colour than for trial 1.
Outcomes from trial 1 and 2
Trial 1 was better for non-set, whether hot or cold, and even guessing for the correction for sugar in trial 2. But, needed less air, so speed 2 more appropriate.
Trial 2 was better for set, with a much creamier mouth feel and deeper yellow colour. Only issue was that it hadn’t set enough after an hour – worth trying with slightly more gelatine, along with less sugar. That was true even allowing for the extent of setting,which was much too much for trial 1.
Same basis as trial 2, but only 60% of the sugar, and using the seeds from a vanilla pod rather than vanilla essence. Used 75g of milk rather than 50g. And, 150% (trial 3) and 200% (trial 4) of the gelatine.
Came out runnier (as you’d expect for slightly more milk). But, only intended to be used as a set custard, so need to wait to assess. As they went into the fridge, looking at trial 3, the vanilla seeds appear to have settled out somewhat – for trial 4 they are still well blended.
Outcomes from trial 3 and 4
In trial 3 the gelatine sheet hadn’t been properly melted – the result was lovely, but not set. Trial 4 was firmly set, so splitting the difference, trial 3 should be the right amount of gelatine (if actually melted!). Use of Vanilla seeds rather than vanilla essence didn’t really feel like it added much. Sweetness was about right.