Mark 1 Hardware Graphs

Well, the real results are as shown at the top of the main roasting page, which is beans from the first and second runs. However, that's not really geeky enough, so here's some graphs. These are from the Mark 1 unit which didn't monitor temperatures - the graphs are generated by reading the thermometer and writing down the values! The unit does display elapsed time as it proceeds, so that helps a bit. [Note: these graphs are all for the first arrangement of rewired popper - in the current arrangement there's more heat input, so it would need lower heat value to obtain the same results.]

The second run. For the first minute the heat is set to 40% in order to get the beans warming and moving without singeing. Then it switches to 90% heat, runs for another 4 minutes then switches off all heat but leaves the fan running to cool the beans. This gave the lightish roast shown at the top of the main roaster page.

In case you're wondering, the first run did give an edible roast (though light for my taste), but I didn't record a useful temperature profile.

The third run. Once again there's a warm-up, then four minutes at 90%. The controller was programmed to give three minutes at 100%, but I cut the heat (with the switch on the side of the popper) at just under two minutes into this stage, because it was already a very dark oily roast by then. The temperature trace gives out because my LCD thermometer got too hot, and the display became unreadable. The last two points are probably wrong.

The fourth run. This is building on the third and aiming to have a steady ramp of temperature resulting in a dark roast without manual intervention on the heater element. This it did successfully, to uniform glossy beans, but once again my thermometer gave out. After 5 mins of cooling (fan full power, no heat input), the beans are just slightly warm to the touch, so I think the cooling profile is much like that shown on the second run graph (above).

This was an attempt to roast slightly lighter but slower. It ended too light - useable, but lighter than I like.

For this run I've relocated the thermometer where the display doesn't get so hot, meaning I get results right through the cooling too.

Just for fun, a similar program to the above roast, but without any beans in the chamber. Interesting that the temperatures don't plateau completely in the 2 or so minutes allowed, and also that it doesn't cool down to ambient within five minutes of fan with no heater input - it settles at about 10 degrees warmer than it started.

Last graph for a while, because this (run 7, sixth roast, since the one above is not really a roast) gave me beans pretty much how I wanted - a glossy dark roast.

I worked hard scribbling down times and temperatures for this one, and got lots more points. I'm glad I did, because it shows quite clearly the delay between changing the heater element drive, and the temperatures. It's most obvious when the heat goes off at the end - the element is switched off but the next two data points (5 and 10 seconds later respectively) show no drop in temperature, then a further five seconds later (now 15 since switch off) it's dropped 20 degrees, and carries on down at that sort of rate.


back to the main coffee-roasting page
back to Ian's contents page
back to www.astounding.org.uk top contents page

To comment on anything (please do) email ian@astounding.org.uk