High temperature thermistor (PT1000) for high temp materials?
I'm wanting to print PPS-CF on my Mk4S, and it looks like the only modification needed (other than having an enclosure) is a high temperature hotend thermistor. Has anyone done this, yet? Is there a source for a "plug and play" thermistor for the nextruder system? I know I'd need to modify the firmware and break the little fuse thing. I'm not worried about that. I just don't want to put a bunch of effort into something I can just... buy and install.
RE: High temperature thermistor (PT1000) for high temp materials?
Thinking about the same thing for my Core One. Did you try it?
RE: High temperature thermistor (PT1000) for high temp materials?
Not yet. Apparently, with the stock resistors, the thermal resolution is pretty bad at the desired temperatures. Something like ~1°C per bit, compared to the stock 0.1°C per bit. Ideally, you'd change the resistors on the board, to get the resolution back. Not sure I'm ready to commit to that, just yet.
RE:
One could use an amplifier board instead of switching the resistor. Something like https://www.adafruit.com/product/3648 or https://e3d-online.com/products/pt100-amplifier-board
Is <1°C resolution even necessary though? I'd think that the temperature fluctuations due to melting the filament would greatly exceed 1°C.
RE:
One could, but that would require even more wiring and modification than replacing a single resistor. Especially considering the loveboard, and limited available space.
Necessary? I think yes. It's not about holding the hotend to less than 1 degree of change. It's about letting the processor actually know what's going on, so it can ramp the PWM to the heater, accordingly, and not resort to the old "bang-bang" control method, when it can't see the *velocity* of temperature change. The PID loop used to control the hotend temperature depends on "rate of change" as one of the variables. If the computer can't see but 0.5℃ changes, it can't really see the rate of change, and the PID loses its effectiveness. This is particularly critical when you're printing tiny features.
Amended math, based on this post. The factory thermistor (100kΩ) has roughly 0.08℃ steps (linearly interpolating. It'll be wider at higher temps, and narrower at lower) in the 0-300℃ range. The PT1000, without a driver or resistor change, would have approximately 0.5℃ steps (again, wider steps at higher temperatures). While this is "usable", it is insufficient for reliable use at high detail, with high speed. Just because a printer "can work" doesn't mean it's good enough for me to trust as a tool. For me, the printer is a tool that I need to be able to trust, to hold tight tolerances. I'm not making low-poly-fox models. I'm making engineering prototypes. Microns matter, and therefore tiny fractions of a degree do, too.
Also: watch your hotend temperature, while you're printing something using high temperature filament. The hotend rarely moves more than a single degree off of the setpoint, which means that the PID loop is actively working, constantly varying the power to the hotend, to prevent that deviation. It is far more tightly controlled than you think.