Yeah you may be right. I need to really get it all in an excel doc and crunch the numbers. The money I'd save though, is more from the parts I already have. If I reduced the system to two outlets then made one for the lower element in the kettle then switched the hlt and upper bk element at the other outlet, I'm pretty much already wired past the outlets. I just need to add some code to the arduino. I'd need two more 2000w elements, nuts from bargain fittings, and the extension cords to them, one breaker, one gfi outlet, etc. Like I said though, I've still got numbers to crunch. I'll let you know what I come up with.
As far as pwm, that's all pids do already. Pulse width modulation is just turning the load on and off for a percentage of the latch time. Generally latch times are pretty fast, in the kHz, but even 2 seconds is still technically pwm. What makes it cheaper is that you could put together a pwm circuit for about 10 bucks using a 555 chip and a pot. Anything large is run this way. 240 volt 30 amp rheostats are fairly non existent.
As far as pwm, that's all pids do already. Pulse width modulation is just turning the load on and off for a percentage of the latch time. Generally latch times are pretty fast, in the kHz, but even 2 seconds is still technically pwm. What makes it cheaper is that you could put together a pwm circuit for about 10 bucks using a 555 chip and a pot. Anything large is run this way. 240 volt 30 amp rheostats are fairly non existent.