Kennebago wrote: snip ..
As someone mentioned earlier, to control a variable you want to measure it directly and the good news is in this day and age you can directly measure almost anything and with modern PLC controllers you can control it in a closed loop fashion. This does NOT have to mean a lot more cost - it just means taking an approach that is commensurate with the problem at hand. Many problems that were not solvable in the 40's (Compare that carburated DeSoto with a 2014 FI Chevy)are easily solved today. Seems to me there is an opportunity for someone (perhaps AHS) to re-engineer a boiler system, to make it efficient, automatic, cost effective and above all safe.
Things like combustion mixtures can be measured and controlled, the need for air can be anticipated and injected so its there when its needed. My guess is that one of the biggest problems is that the size of the market won't support the R&D necessary to put such a unit on the market -- Ohhh - if only I was a bit younger.
Kennebago ... Welcome to the forum. I'm particularly interested in your comment about measuring combustion mixtures. What types of sensors are you referring to? Brand names and/or part numbers would be helpful. I've considered making an instrument to do so but was thwarted in not being able to find a suitable combustion sensor. I found some, but they cost more that the boiler (-: As is see it there are two control loops required, one for burning height in the antratube and a second for combustion mixture control. Combustion height could be sensed with a linear array of thermocouples or perhaps as few as two and a software algorithm. The microcontroller specs would be really minimal since there is no need for high speed computations.