My coffee buddy, Richard, bought my Silvia/Rocky setup. For the last few weeks he's been complaining about sub-standard shots, describing them as bitter / harsh. My answer was always the same: "Your temperature is off." He tried temperature surfing
but still complained of good shots one day, terrible the next, and no great shots at all. Yesterday we got together for the Rocky/Mazzer shootout rematch
for the first match results by CraigA). Afterward, we turned to diagnosing Silvia. Another CoffeeGeek, David Morris, mentioned the idea of temperature surfing after reaching max temperature. I wanted to investigate this idea of "reverse temperature surfing."
We did a lot
of temperature trials to reverse temperature surf Silvia accurately. Normally when you temperature surf, you introduce water into the boiler, wait for the heating element to come on, then wait n
seconds for the water to heat up to your target temperature. Problem is, my measurements showed variance, probably because split-second response can make a big difference (e.g., two seconds too much water and you're off by several degrees), not to mention "warm versus very hot" Silvia variations. I also noted that his machine put out a lot of water, almost 10oz in 30 seconds. Again, that makes it hard to temperature surf accurately by introducing water.
So instead, we tried the other direction, hence why we call it "reverse temperature surfing." We introduced water until the boiler came on. Then we waited for the heating element to turn off. Then we plotted the drop off in temperature (1 minute, 2 minutes, 3 minutes, 4 minutes, 5 minutes... really boring stuff). By trial and error, we found that 2 minutes after the boiler clicks off with a well heated Silvia, the temperature reached 201F. At first we would get variances of 196, 197, 199, 199... and then 201, 201, ..., 201. It was that dead-on, time after time after time. His wife is a statistician and says we'd have to take at least 25 measurements to have any true statistical certainty; we stopped after six consecutive results of 201F. Obviously this type of temperature surfing requires lots of patience, since you're waiting about three minutes to get to the right temperature by the time it heats up to boiler shutoff and cools down to your target temperature.
To make an already long story short(er), this technique seemed to produce pinpoint accurate temperatures. Since the boiler is cooling (not heating as would be the case in 'normal' temperature surfing), we suspect the temperature stability during the shot is enhanced. The Intelligentsia Black Cat shots we pulled yesterday afternoon using this technique were some of the best Silvia has ever produced. Two shots from the Silvia/Mazzer Mini trial were "rock your world
" calibre and we rated the third "very good to excellent".