Today I'm playing around with controlling the Highlighter with a PID. It simply turns it on and off to maintain the temperature... not the best for bulb life, but it does seem to work. My aim is to have it crank a bit harder when in use to eliminate the refractory period between back-to-back baskets. I'll let y'all know how it goes...
I'm curious if there's a thermocouple/thermistor that can be placed within the beads for a more responsive reading. I'd found
this amphenol glass-encased thermistor that looked somewhat promising, but I'm not 100% certain about the safety of putting dumet wire (copper coated nickel IIRC) in the vapor path, and the maximum operating temp is 300C, pretty close to the highest temps you'd see in the barrel already. One idea I had was using a dremel to grind two shallow grooves in the ground glass joint and put the thermistor leads there, mounting the thermistor right above the proforma screen so it would only be measuring airflow, although you'd have to make your PID controller figure out when you're taking a draw and when you aren't, since there would be drastic differences in readings between when there's air flowing over it and when there isn't.
Also, regarding the method of temp controlling: I did some research awhile back, and I think a better solution for bulb health would be to build a VVPS that could be controlled by an arduino or other PID capable microcontroller. Integrated circuits like the LM338K use a potentiometer (variable resistor) for their voltage control, and digital potentiometers are available for arduinos that could allow for .1V stepping (or likely even finer steps). This way you could keep the bulb always on but have fine-tuned control over exactly how much heat the core would be generating.
One nice thing is these two problems can be solved independently, so you don't have to figure out a solution for the temp-monitoring issue before proceeding with working out the arduino VVPS (far easier IMO). Once I get one of these myself I'll probably go right into making the arduino controlled VVPS for it sans PID functionality, which will still provide some nice features like being able to implement voltage boost profiles. Then later down the line, I can work on installing a thermistor and PID code for actual temp control.
Materials safety is a pain, though, and I'm thinking it's gonna be pretty hard to find a responsive thermometer that'll be fully safe in the vapor path. One possible idea I had was a "passive TC coil", where you'd put a coil of SS wire into the core outside the beads, and measure with an ohmmeter to calculate the temp (similar to how TC ecig mods work). The main problems here are I'm not sure how responsive this method would be (if it takes 20 seconds after a change to yield accurate temps, kinda not worth it) and you'd only be getting an average of temps across the whole heating tube, not measuring the air temp at the end which would be more accurate/responsive to what your bowl temps should be. Regarding thermistors, it also looks like the majority top out their operating temp at 300 degF, which makes measuring right beside the bulb way more iffy. Maybe there'll be some miracle thermocouple someone finds one day with a 400 deg operating temp, glass encased, with leads made from a vapor safe material.