After I figured out what was going on and had some idea of how to set (or should I say, "Autoset") my PID I had a few days to use it before writing this. Mind you, there will be more to come with some experiments that are a bit more scientific, or at least to say, some subjective taste testing over a range of temperatures, but I thought I would share some early thoughts and observations based on nothing more that what I have seen in general use in that short time.

      One of the first things I noticed was that all my pucks were now quite dry almost immediately after pulling the shot. Before the PID was installed I would get pucks that were sometimes quite moist on the top and occasionally muddy or mushy on top. Now, all the pucks come out quite dry, even just a few seconds after the pull has completed. The wettest puck I have seen was just a bit moist on one edge. All the pucks have been quite solid as well- holding their shape even after being knocked. The driest puck was quite void of any moisture based on a visual examination, and this occurred at a PID setting of 232 f. Based on what I have seen here, if you are getting soupy pucks my first thought now has to be that the brew temperature is too low or the machine has not been allowed to sufficiently warm up.

      My very first pulls with the PID installed were done at a set point of 228f. if I remember correctly. These were very smooth in taste. The next day I tried 232f. and these were noticeably bitter, and presumably over-extracted as they also exhibited that nasty burnt taste behind them that reminds one that the BBQ needs emptying. The third and subsequent days I went back to 228f., and once again, these tasted very smooth with no bitterness or harshness. My wife commented that the 228f. coffee was very smooth and with no coffee-bitterness at all, with a hint chocolate and a natural sweetness. This could have been an effect of the lack of bitterness allowing the light sweetness of the soymilk to come through.

      The immediate and most obvious advantage of a PID is the consistency of the set temperature. Pick a temp, run the autoset one time, and from that point on you can be fairly well assured that the chosen temp is the one you will get each time, minus a bit of recovery time between shots. But there is also an underlying advantage of being able to adjust that temperature in controlled increments. It allows the user to select a slightly lower or higher temperature depending on various factors such as roast, blend, age of roast, and most importantly, personal preferences.

      I can certainly say that I do not at all miss temp surfing or time surfing. Beyond the fact that with the assistance of the PID I can now pull consecutive shots in less time, I do not feel the pressure of rushing through the dose and tamp to make it back to the brewhead in the requisite 45 or 50 seconds. I feel that time and temp surfing methods are superior to the hit-the-button-and-prey method, but they are neither accurate nor repeatable. Surfing lets the user begin the pull at some point during the temperature rise. It is sort of like trying to jump on a moving train. Sometimes I would hit the sweet spot, and at other times it would just run me over. The PID seems to produce far more consistent results. Even if I get back to the machine for a second double immediately after the first, and the temp is a degree off my setpoint, I still get better results than when surfing. Most of the time, the machine is idling at a set point and waiting for you to get on board- the machine has become a private train awaiting your arrival at the station.

      My PID has dual setpoints. It two different temperature settings can be selected through an external toggle switch. I have chosen to use the second setpoint to steam. Although the steam point comes up to temperature a bit slower than the stock T-stat allows (about the only negative point to the PID installation I have observed thus far), it holds the temperature better and since I can choose the setpoint, it can be higher than stock, and thus hold more total energy. The wide range of temperature allowed by the snap T-stat (between the on point and the off point, plus over-shoot) also makes it difficult to maintain consistency of steam volume. That is not a problem with the PID. At a setting of 285 f. I get tons of powerful steam, noticeably more powerful than the stock Silvia. In the future I may try some higher temperature for steaming, but an "experimental" setting of around 300f. produced so much steam that it was difficult to control. It stretched the milk much faster, but I was not accustomed to it and over-heated the milk.

      There has also been a psychological advantage of using the PID-equipped machine. I can now concentrate on all the other factors within my control without worrying about, or even thinking about the brew temp. In particular, it allows me to more carefully dose and tamp. Not that it takes all that long, but without the overriding feeling of being rushed to beat the time surf, it adds a more relaxed atmosphere to the process.

      added 4/30 - I have noticed that there seems to be no steam build-up in the boiler regardless of how long the machine is allowed to idle. This is an added benefit from the PID's ability to maintain a set temperature without overshoot. That is, if set to produce 205 degree water, the machine never goes above 205 or 206, and then, only ocassionally and for a very short period of time.