Hi Anthony
Thanks for the kind words!!! I'm excited to hear you're enjoying it, and glad to see you've made use of "events". Personally, I've converted all of my profiles from reading/control curve combinations over to just reading curves with events. I'm finding I like that setup much better.
About RoR curves... When I designed them, I had originally intended to employ curve "smoothing", since they are extreme amplifications of curve data, and irregularities show up as the ugly spikes. I decided against it, though, since I realized that not every user employs data logging in their roasting, and in those cases, I couldn't make any guesses about how often a user was creating nodes.
I will probably rethink that, and use smoothing for data logging only, or create a preference setting. But, in the mean time, unfortunately, there's only two things that would help.
- Increasing the sample frequency. I use 15 seconds, but you can go much lower than that. That will make them "appear" a little smoother.
- Using insulated (ungrounded) thermocouples. This has the net effect of curve smoothing at the hardware level. Since insulated probes take long to register temps, they are much less prone to random spikes, and will show a much smoother transition during the roast. Now, the tradeoff, of course, is that they're not as responsive as grounded probes. I, personally, don't find that to be a problem - I've always used insulated probes. But your mileage may vary.
Danny