There is a coffee roasting machine called the Aillio Bullet. You have to read its specs hear the hype to see what it is capable of, and at £3,250 it appears to be worth it. However, it doesn't have a flu or chimney, just a vent, so all the steam and oil it releases just fill up the kitchen. Hopefully they'll see the short-sightedness of this. Not all of us have garages, basements or cellars.

It does though, have one intriguing piece of technology tucked into it's sleek features, that got me rather excited.

My Gene Café has a single thermometer, measuring the temperature of the exhaust air. This is good, you definitely want to measure that, but to have increased control you need to know the temperature of the beans. Most professional machines will have a thermal probe in amongst the beans.

You need to keep the temperature of the beans constantly rising, however you might want to slow things down a bit, so rather counter-intuitively you can drop the amount of energy you are pumping into the roasting chamber. On my Gene Café, this means lowering the air temperature. If the beans are a lot cooler than the air, they will still be 'on the rise'.

Therefore, knowing the temperature of the beans and the temperature of the air (usually measured in the exhaust) gives you an incredible amount of control over the process because you know what is going on. The Aillio Bullet does this differently than most - if not all - other roasting machines: instead of using a thermal probe, it uses a thermal camera. It measures the temperature of the beans by the amount of infra-red energy they emit.

You can buy infra-red thermometers. They come in two basic varieties. The first fires a red laser (possibly infra-red) at the target and measures the distortion as it bounces back. This looks pretty much the same as a hand-held radar gun. The problem with these kinds of sensors is they measure surface temperature. Although I have a glass roasting chamber, I could be measuring the temperature of the chamber itself.

The other kind is literally a thermal camera and uses the colour or brightness of the target to determine its temperature. The problem with these is that the infra-red radiation is being blocked by the glass of the roasting chamber - the beans are radiating in the infra-red, but that radiation is then being absorbed by the glass.

In both cases, I would be measuring the glass, not what's inside it. However, with the camera model I can see the temperature of the glass in the locations where the beans currently are, the temperature where they aren't, and how quickly that temperature changes as the beans tumble in the rotating chamber.

Can this actually be useful, or is it a complete red-herring? I'll let you know after my infra-red thermometers arrive arrive.

Horror of horrors, I bought pre-roasted beans from the supermarket this week. Why? I haven't figured out how to adjust the latest roasting profile.

I know that the yellowing stage is about the right length at 6 minutes, and the browning stage is also about right at approximately 4 minutes. The development stage though is a woeful 6 minutes and it should be ~2. I cant weak the final stage by increasing the temperature by 10°C but the machine literally can't go any higher than 250°C - and I wouldn't want to go any higher anyway because that'll burn the beans.

I could drop the pre-heat temperature and extend the yellowing phase by 2 mins, which is acceptable. That will also allow me to drop the temperature of the browning phase to extend that to 4 mins 50 seconds, and the development stage can take the full heat and hopefully allow it to complete in 3 mins 12, for a total time of 16 minutes - about the same as it was for Profile I. I'll just have to suck it and see, although the temperature may have to go up to 250 in the browning phase and stay there to complete on schedule.

After roasting, oxidation occurs and as part of that process oxygen combines with carbon to produce carbon-dioxide gas, which is then released - especially over the first three days. After these three days I have been putting my beans in a negative pressure tin - one where all the air has been removed - to preserve the flavour.

It turns out, that was a mistake. As there is no oxidation, carbon is locked in the beans, and there is a burnt taste. I thought at first I was burning my roasts, but this is not the case. I should have been putting my beans in a positive-pressure tin which should be opened daily to allow carbon-dioxide to be replaced with fresh air to allow oxidation to occur. Will this destroy other flavours? Yes, but the beans need 10 days to mature after roasting and it is after this period you need to make every effort to preserve flavour and aroma.