Heater option
-
I would really like to see an optional parameter in M307 to allow a startup delay before the system checks for the correct heating rate. This could also be in the form of a relaxed parameter (dead time maybe) that is only applied at the very early stages of heating.
I have a bed heater that I have tuned a number of times that will, intermittently (50%) throw a heating rate error but ONLY when heating up from ambient temperature. Once the bed has gone past whatever hump in the beginning (there seems to be a noticeable delay from heater activation to when I see the temperature rising) of the heat up phase, everything works just peachy.
I have seen that on different printers and also on hot end heaters on occasion.
I believe I can just manually set a different dead time parameter but of course that affects the entire PID system when things are working just fine under normal circumstances.
It is annoying to send a print job to the printer and then, when you check on the printer maybe 15 minutes later, you find that the system barfed with a heater error at the very beginning of the print job.Update: Of course as soon as I sent this firmware wish on it's way, I find out that the error isn't necessarily just happening in the beginning. It took 3 restarts just now to get the bed up to temperature. Maybe some sort of slowed reaction time in the heat up phase? If that can be implemented without compromising safety ....
Update 2: I think I will replace the thermistor just in case
-
I would like to leave the firmware wish stand because this is a frequently encountered problem with other users and not just myself. A lot of times the sensor hickups during the heat-up phase and the safety system in RRF kicks in very fast. Introducing a delay or some sort during the heatup phase would probably fix many of these errors.
-
@KenW, this printer IS line powered (750W 300 x 300)
I am not entirely sure what actually is happening. Before I turn on the heater, I get a nice steady temperature reading. As soon as the heater goes on, temperature jumps around and, for a few seconds, that's all it does. Then the temperature starts to go up and the temperature reading appears to be generally stable (and increasing).I just had an interesting thought .... with the heater being line powered and the thermistor being glued to the heater, maybe I am seeing interference. This would be more of a problem when cold as the resistance is higher. I do not use a shielded cable for the thermistor and it runs together with the power lines from the bed right to the electronics compartment. It's not even a twisted line.
Funny thing is that all three of my printers are configured that way and they see to be mostly ok (although I had intermittent problems like this in the past)
Interference would explain a lot though ..... I wonder if it would make sense to stick a substantial capacitor across the sensor input pins on the Duet to absorb quick spikes? Any comments on a bandaid such as that?I am curious though why a spike would trigger the heater error because the error has to persist for a while before a fault is triggered. I seem to recall that the time for error detection can be extended - I will give this a try first since it's just a line in config.g. I am reluctant to do this because it applies to the entire print job and not just the startup phase.
Now that I am thinking about it, maybe since the heater is hard on during startup it causes more interference than a slowly cycling heater?
