this was discussed in discord by me, atom and legitsu but I thought it prudent to document here even though it doesn't particularly alter usability of the printer. It would be nice to fix if it was not resource intensive as we could issue a warning to the user if fan speed dropped by 15% from spec value which is the value sunon says signals replacement IIRC.
Firmware: COSMOS 0.0.2
Symptom: Fan tach readings are lower than expected in COSMOS for at least the hotend and part cooling fan (12k rpm stated by elegoo for the hotend fan ). And fan speed readings are lower at 100% than 90%.
Reproduction steps: Self explanatory
Diagnosis and mitigation attempts: Legitsu and atom attempted to alter tach ppr, cycle time, and poll interval but were not able to eliminate anomalously low readings at 100% PWM relative to 90% but I am unsure of the parameters tested. I measured tach output for both hotend and part cooling fans and and found that tach output is a set 50% duty cycle with variable frequency with 2 PPR. Hotend and part cooling fans read at ~400 hz at 100% duty cycle and thus 12k rpm. Fan speed is higher at 100% for the part cooling fan as expcted. COSMOS reads both fans as being similar speed as expected so the issue appears systematic and to affect both equally.
Fan speeds in COSMOS v0.0.2
Identification of speed discrepancy at 90 and 100%:
hotend fan tach signal measured by oscilloscope at 100% duty cycle
Plot of part cooling fan speed vs pwm duty cycle as measured by oscilloscope
I do not know if this affects the mainboard fan tach signal as I misplaced this fan after replacing it, if I find it I can check. If anyone else with the OEM mainboard fan still installed can comment on if it displays the same discrepancy at 100% duty cycle that would be helpful
this was discussed in discord by me, atom and legitsu but I thought it prudent to document here even though it doesn't particularly alter usability of the printer. It would be nice to fix if it was not resource intensive as we could issue a warning to the user if fan speed dropped by 15% from spec value which is the value sunon says signals replacement IIRC.
Firmware: COSMOS 0.0.2
Symptom: Fan tach readings are lower than expected in COSMOS for at least the hotend and part cooling fan (12k rpm stated by elegoo for the hotend fan ). And fan speed readings are lower at 100% than 90%.
Reproduction steps: Self explanatory
Diagnosis and mitigation attempts: Legitsu and atom attempted to alter tach ppr, cycle time, and poll interval but were not able to eliminate anomalously low readings at 100% PWM relative to 90% but I am unsure of the parameters tested. I measured tach output for both hotend and part cooling fans and and found that tach output is a set 50% duty cycle with variable frequency with 2 PPR. Hotend and part cooling fans read at ~400 hz at 100% duty cycle and thus 12k rpm. Fan speed is higher at 100% for the part cooling fan as expcted. COSMOS reads both fans as being similar speed as expected so the issue appears systematic and to affect both equally.
Fan speeds in COSMOS v0.0.2
Identification of speed discrepancy at 90 and 100%:
hotend fan tach signal measured by oscilloscope at 100% duty cycle
Plot of part cooling fan speed vs pwm duty cycle as measured by oscilloscope
I do not know if this affects the mainboard fan tach signal as I misplaced this fan after replacing it, if I find it I can check. If anyone else with the OEM mainboard fan still installed can comment on if it displays the same discrepancy at 100% duty cycle that would be helpful