josh
/
Prusa-Firmware
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Prusa-Firmware/Firmware/mmu2.cpp

1104 lines
38 KiB
C++
Raw Blame History

#include "mmu2.h"
#include "mmu2_config.h"
#include "mmu2_error_converter.h"
#include "mmu2_fsensor.h"
#include "mmu2_log.h"
#include "mmu2_marlin.h"
#include "mmu2_marlin_macros.h"
#include "mmu2_power.h"
#include "mmu2_progress_converter.h"
#include "mmu2_reporting.h"
#include "strlen_cx.h"
#include "SpoolJoin.h"
#ifdef __AVR__
// As of FW 3.12 we only support building the FW with only one extruder, all the multi-extruder infrastructure will be removed.
// Saves at least 800B of code size
static_assert(EXTRUDERS == 1);
constexpr float MMM_TO_MMS(float MM_M) { return MM_M / 60.0f; }
#endif
namespace MMU2 {
template <typename F>
void waitForHotendTargetTemp(uint16_t delay, F f) {
while (((thermal_degTargetHotend() - thermal_degHotend()) > 5)) {
f();
safe_delay_keep_alive(delay);
}
}
void WaitForHotendTargetTempBeep() {
waitForHotendTargetTemp(3000, []{ });
MakeSound(Prompt);
}
MMU2 mmu2;
MMU2::MMU2()
: logic(&mmu2Serial, MMU2_TOOL_CHANGE_LOAD_LENGTH, MMU2_LOAD_TO_NOZZLE_FEED_RATE)
, extruder(MMU2_NO_TOOL)
, tool_change_extruder(MMU2_NO_TOOL)
, resume_position()
, resume_hotend_temp(0)
, logicStepLastStatus(StepStatus::Finished)
, state(xState::Stopped)
, mmu_print_saved(SavedState::None)
, loadFilamentStarted(false)
, unloadFilamentStarted(false)
, toolchange_counter(0)
, tmcFailures(0) {
}
void MMU2::Start() {
mmu2Serial.begin(MMU_BAUD);
PowerOn(); // I repurposed this to serve as our EEPROM disable toggle.
mmu2Serial.flush(); // make sure the UART buffer is clear before starting communication
extruder = MMU2_NO_TOOL;
state = xState::Connecting;
// start the communication
logic.Start();
logic.ResetRetryAttempts();
}
void MMU2::Stop() {
StopKeepPowered();
PowerOff(); // This also disables the MMU in the EEPROM.
}
void MMU2::StopKeepPowered() {
state = xState::Stopped;
logic.Stop();
mmu2Serial.close();
}
void MMU2::Tune() {
switch (lastErrorCode) {
case ErrorCode::HOMING_IDLER_FAILED:
{
// Prompt a menu for different values
tuneIdlerStallguardThreshold();
break;
}
default:
break;
}
}
void MMU2::Reset(ResetForm level) {
switch (level) {
case Software:
ResetX0();
break;
case ResetPin:
TriggerResetPin();
break;
case CutThePower:
PowerCycle();
break;
case EraseEEPROM:
ResetX42();
break;
default:
break;
}
}
void MMU2::ResetX0() {
logic.ResetMMU(); // Send soft reset
}
void MMU2::ResetX42(){
logic.ResetMMU(42);
}
void MMU2::TriggerResetPin() {
reset();
}
void MMU2::PowerCycle() {
// cut the power to the MMU and after a while restore it
// Sadly, MK3/S/+ cannot do this
// NOTE: the below will toggle the EEPROM var. Should we
// assert this function is never called in the MK3 FW? Do we even care?
PowerOff();
safe_delay_keep_alive(1000);
PowerOn();
}
void MMU2::PowerOff() {
power_off();
}
void MMU2::PowerOn() {
power_on();
}
bool MMU2::ReadRegister(uint8_t address) {
if (!WaitForMMUReady())
return false;
do {
logic.ReadRegister(address); // we may signal the accepted/rejected status of the response as return value of this function
} while (!manage_response(false, false));
return true;
}
bool MMU2::WriteRegister(uint8_t address, uint16_t data) {
if (!WaitForMMUReady())
return false;
// special cases - intercept requests of registers which influence the printer's behaviour too + perform the change even on the printer's side
switch (address) {
case 0x0b:
logic.PlanExtraLoadDistance(data);
break;
case 0x14:
logic.PlanPulleySlowFeedRate(data);
break;
default:
break; // do not intercept any other register writes
}
do {
logic.WriteRegister(address, data); // we may signal the accepted/rejected status of the response as return value of this function
} while (!manage_response(false, false));
return true;
}
void MMU2::mmu_loop() {
// We only leave this method if the current command was successfully completed - that's the Marlin's way of blocking operation
// Atomic compare_exchange would have been the most appropriate solution here, but this gets called only in Marlin's task,
// so thread safety should be kept
static bool avoidRecursion = false;
if (avoidRecursion)
return;
avoidRecursion = true;
mmu_loop_inner(true);
avoidRecursion = false;
}
void __attribute__((noinline)) MMU2::mmu_loop_inner(bool reportErrors) {
logicStepLastStatus = LogicStep(reportErrors); // it looks like the mmu_loop doesn't need to be a blocking call
if (isErrorScreenRunning()) {
// Call this every iteration to keep the knob rotation responsive
// This includes when mmu_loop is called within manage_response
ReportErrorHook((CommandInProgress)logic.CommandInProgress(), (uint16_t)lastErrorCode, uint8_t(lastErrorSource));
}
}
void MMU2::CheckFINDARunout() {
// Check for FINDA filament runout
if (!FindaDetectsFilament() && check_fsensor()) {
SERIAL_ECHOLNPGM("FINDA filament runout!");
stop_and_save_print_to_ram(0, 0);
restore_print_from_ram_and_continue(0);
if (SpoolJoin::spooljoin.isSpoolJoinEnabled() && get_current_tool() != (uint8_t)FILAMENT_UNKNOWN){ // Can't auto if F=?
enquecommand_front_P(PSTR("M600 AUTO")); // save print and run M600 command
} else {
enquecommand_front_P(MSG_M600); // save print and run M600 command
}
}
}
struct ReportingRAII {
CommandInProgress cip;
explicit inline __attribute__((always_inline)) ReportingRAII(CommandInProgress cip)
: cip(cip) {
BeginReport(cip, (uint16_t)ProgressCode::EngagingIdler);
}
inline __attribute__((always_inline)) ~ReportingRAII() {
EndReport(cip, (uint16_t)ProgressCode::OK);
}
};
bool MMU2::WaitForMMUReady() {
switch (State()) {
case xState::Stopped:
return false;
case xState::Connecting:
// shall we wait until the MMU reconnects?
// fire-up a fsm_dlg and show "MMU not responding"?
default:
return true;
}
}
bool MMU2::RetryIfPossible(uint16_t ec) {
if (logic.RetryAttempts()) {
SetButtonResponse(ButtonOperations::Retry);
// check, that Retry is actually allowed on that operation
if (ButtonAvailable(ec) != NoButton) {
logic.SetInAutoRetry(true);
SERIAL_ECHOLNPGM("RetryButtonPressed");
// We don't decrement until the button is acknowledged by the MMU.
//--retryAttempts; // "used" one retry attempt
return true;
}
}
logic.SetInAutoRetry(false);
return false;
}
bool MMU2::VerifyFilamentEnteredPTFE() {
planner_synchronize();
if (WhereIsFilament() == FilamentState::NOT_PRESENT)
return false;
uint8_t fsensorState = 0;
uint8_t fsensorStateLCD = 0;
uint8_t lcd_cursor_col = 0;
// MMU has finished its load, push the filament further by some defined constant length
// If the filament sensor reads 0 at any moment, then report FAILURE
const float delta_mm = MMU2_CHECK_FILAMENT_PRESENCE_EXTRUSION_LENGTH - logic.ExtraLoadDistance();
// The total length is twice delta_mm. Divide that length by number of pixels
// available to get length per pixel.
// Note: Below is the reciprocal of (2 * delta_mm) / LCD_WIDTH [mm/pixel]
const float pixel_per_mm = 0.5f * float(LCD_WIDTH) / (delta_mm);
TryLoadUnloadProgressbarInit();
/* The position is a triangle wave
// current position is not zero, it is an offset
//
// Keep in mind that the relationship between machine position
// and pixel index is not linear. The area around the amplitude
// needs to be taken care of carefully. The current implementation
// handles each move separately so there is no need to watch for the change
// in the slope's sign or check the last machine position.
// y(x)
// ▲
// │ ^◄────────── delta_mm + current_position
// machine │ / \
// position │ / \◄────────── stepper_position_mm + current_position
// (mm) │ / \
// │ / \
// │/ \◄───────current_position
// └──────────────► x
// 0 19
// pixel #
*/
// Pixel index will go from 0 to 10, then back from 10 to 0
// The change in this number is used to indicate a new pixel
// should be drawn on the display
uint8_t dpixel1 = 0;
uint8_t dpixel0 = 0;
for (uint8_t move = 0; move < 2; move++) {
MoveE(move == 0 ? delta_mm : -delta_mm, MMU2_VERIFY_LOAD_TO_NOZZLE_FEED_RATE);
while (planner_any_moves()) {
// Wait for move to finish and monitor the fsensor the entire time
// A single 0 reading will set the bit.
fsensorStateLCD |= (WhereIsFilament() == FilamentState::NOT_PRESENT);
fsensorState |= fsensorStateLCD; // No need to do the above comparison twice, just bitwise OR
// Always round up, you can only have 'whole' pixels. (floor is also an option)
dpixel1 = ceil((stepper_get_machine_position_E_mm() - planner_get_current_position_E()) * pixel_per_mm);
if (dpixel1 - dpixel0) {
dpixel0 = dpixel1;
if (lcd_cursor_col > (LCD_WIDTH - 1)) lcd_cursor_col = LCD_WIDTH - 1;
TryLoadUnloadProgressbar(lcd_cursor_col++, fsensorStateLCD);
fsensorStateLCD = 0; // Clear temporary bit
}
safe_delay_keep_alive(0);
}
}
Disable_E0();
TryLoadUnloadProgressbarEcho();
TryLoadUnloadProgressbarDeinit();
if (fsensorState) {
IncrementLoadFails();
return false;
} else {
// else, happy printing! :)
return true;
}
}
bool MMU2::ToolChangeCommonOnce(uint8_t slot) {
static_assert(MAX_RETRIES > 1); // need >1 retries to do the cut in the last attempt
for (uint8_t retries = MAX_RETRIES; retries; --retries) {
for (;;) {
Disable_E0(); // it may seem counterintuitive to disable the E-motor, but it gets enabled in the planner whenever the E-motor is to move
tool_change_extruder = slot;
logic.ToolChange(slot); // let the MMU pull the filament out and push a new one in
if (manage_response(true, true))
break;
// otherwise: failed to perform the command - unload first and then let it run again
IncrementMMUFails();
// just in case we stood in an error screen for too long and the hotend got cold
ResumeHotendTemp();
// if the extruder has been parked, it will get unparked once the ToolChange command finishes OK
// - so no ResumeUnpark() at this spot
UnloadInner();
// if we run out of retries, we must do something ... may be raise an error screen and allow the user to do something
// but honestly - if the MMU restarts during every toolchange,
// something else is seriously broken and stopping a print is probably our best option.
}
// reset current position to whatever the planner thinks it is
planner_set_current_position_E(planner_get_current_position_E());
if (VerifyFilamentEnteredPTFE()) {
return true; // success
} else { // Prepare a retry attempt
UnloadInner();
if (retries == 2 && cutter_enabled()) {
CutFilamentInner(slot); // try cutting filament tip at the last attempt
}
}
}
return false; // couldn't accomplish the task
}
void MMU2::ToolChangeCommon(uint8_t slot) {
while (!ToolChangeCommonOnce(slot)) { // while not successfully fed into extruder's PTFE tube
// failed autoretry, report an error by forcing a "printer" error into the MMU infrastructure - it is a hack to leverage existing code
// @@TODO theoretically logic layer may not need to be spoiled with the printer error - may be just the manage_response needs it...
logic.SetPrinterError(ErrorCode::LOAD_TO_EXTRUDER_FAILED);
// We only have to wait for the user to fix the issue and press "Retry".
// Please see CheckUserInput() for details how we "leave" manage_response.
// If manage_response returns false at this spot (MMU operation interrupted aka MMU reset)
// we can safely continue because the MMU is not doing an operation now.
static_cast<void>(manage_response(true, true)); // yes, I'd like to silence [[nodiscard]] warning at this spot by casting to void
}
extruder = slot; //filament change is finished
SpoolJoin::spooljoin.setSlot(slot);
++toolchange_counter;
}
bool MMU2::tool_change(uint8_t slot) {
if (!WaitForMMUReady())
return false;
if (slot != extruder) {
if (/*FindaDetectsFilament()*/
/*!IS_SD_PRINTING && !usb_timer.running()*/
!marlin_printingIsActive()) {
// If Tcodes are used manually through the serial
// we need to unload manually as well -- but only if FINDA detects filament
unload();
}
ReportingRAII rep(CommandInProgress::ToolChange);
FSensorBlockRunout blockRunout;
planner_synchronize();
ToolChangeCommon(slot);
}
return true;
}
/// Handle special T?/Tx/Tc commands
///
///- T? Gcode to extrude shouldn't have to follow, load to extruder wheels is done automatically
///- Tx Same as T?, except nozzle doesn't have to be preheated. Tc must be placed after extruder nozzle is preheated to finish filament load.
///- Tc Load to nozzle after filament was prepared by Tx and extruder nozzle is already heated.
bool MMU2::tool_change(char code, uint8_t slot) {
if (!WaitForMMUReady())
return false;
FSensorBlockRunout blockRunout;
switch (code) {
case '?': {
waitForHotendTargetTemp(100, [] {});
load_filament_to_nozzle(slot);
} break;
case 'x': {
thermal_setExtrudeMintemp(0); // Allow cold extrusion since Tx only loads to the gears not nozzle
tool_change(slot);
thermal_setExtrudeMintemp(EXTRUDE_MINTEMP);
} break;
case 'c': {
waitForHotendTargetTemp(100, [] {});
execute_load_to_nozzle_sequence();
} break;
}
return true;
}
void MMU2::get_statistics() {
logic.Statistics();
}
uint8_t __attribute__((noinline)) MMU2::get_current_tool() const {
return extruder == MMU2_NO_TOOL ? (uint8_t)FILAMENT_UNKNOWN : extruder;
}
uint8_t MMU2::get_tool_change_tool() const {
return tool_change_extruder == MMU2_NO_TOOL ? (uint8_t)FILAMENT_UNKNOWN : tool_change_extruder;
}
bool MMU2::set_filament_type(uint8_t /*slot*/, uint8_t /*type*/) {
if (!WaitForMMUReady())
return false;
// @@TODO - this is not supported in the new MMU yet
// slot = slot; // @@TODO
// type = type; // @@TODO
// cmd_arg = filamentType;
// command(MMU_CMD_F0 + index);
if (!manage_response(false, false)) {
// @@TODO failed to perform the command - retry
;
} // true, true); -- Comment: how is it possible for a filament type set to fail?
return true;
}
void MMU2::UnloadInner() {
FSensorBlockRunout blockRunout;
filament_ramming();
// we assume the printer managed to relieve filament tip from the gears,
// so repeating that part in case of an MMU restart is not necessary
for (;;) {
Disable_E0();
logic.UnloadFilament();
if (manage_response(false, true))
break;
IncrementMMUFails();
}
MakeSound(Confirm);
// no active tool
extruder = MMU2_NO_TOOL;
tool_change_extruder = MMU2_NO_TOOL;
}
bool MMU2::unload() {
if (!WaitForMMUReady())
return false;
WaitForHotendTargetTempBeep();
ReportingRAII rep(CommandInProgress::UnloadFilament);
UnloadInner();
return true;
}
void MMU2::CutFilamentInner(uint8_t slot) {
for (;;) {
Disable_E0();
logic.CutFilament(slot);
if (manage_response(false, true))
break;
IncrementMMUFails();
}
}
bool MMU2::cut_filament(uint8_t slot, bool enableFullScreenMsg /*= true*/) {
if (!WaitForMMUReady())
return false;
if (enableFullScreenMsg) {
FullScreenMsgCut(slot);
}
{
if (FindaDetectsFilament()) {
unload();
}
ReportingRAII rep(CommandInProgress::CutFilament);
CutFilamentInner(slot);
}
extruder = MMU2_NO_TOOL;
tool_change_extruder = MMU2_NO_TOOL;
MakeSound(SoundType::Confirm);
ScreenUpdateEnable();
return true;
}
bool MMU2::loading_test(uint8_t slot) {
FullScreenMsgTest(slot);
tool_change(slot);
planner_synchronize();
unload();
ScreenUpdateEnable();
return true;
}
bool MMU2::load_filament(uint8_t slot) {
if (!WaitForMMUReady())
return false;
FullScreenMsgLoad(slot);
ReportingRAII rep(CommandInProgress::LoadFilament);
for (;;) {
Disable_E0();
logic.LoadFilament(slot);
if (manage_response(false, false))
break;
IncrementMMUFails();
}
MakeSound(SoundType::Confirm);
ScreenUpdateEnable();
return true;
}
bool MMU2::load_filament_to_nozzle(uint8_t slot) {
if (!WaitForMMUReady())
return false;
WaitForHotendTargetTempBeep();
FullScreenMsgLoad(slot);
{
// used for MMU-menu operation "Load to Nozzle"
ReportingRAII rep(CommandInProgress::ToolChange);
FSensorBlockRunout blockRunout;
if (extruder != MMU2_NO_TOOL) { // we already have some filament loaded - free it + shape its tip properly
filament_ramming();
}
ToolChangeCommon(slot);
// Finish loading to the nozzle with finely tuned steps.
execute_load_to_nozzle_sequence();
MakeSound(Confirm);
}
ScreenUpdateEnable();
return true;
}
bool MMU2::eject_filament(uint8_t slot, bool enableFullScreenMsg /* = true */) {
if (!WaitForMMUReady())
return false;
if (enableFullScreenMsg) {
FullScreenMsgEject(slot);
}
{
if (FindaDetectsFilament()) {
unload();
}
ReportingRAII rep(CommandInProgress::EjectFilament);
for (;;) {
Disable_E0();
logic.EjectFilament(slot);
if (manage_response(false, true))
break;
IncrementMMUFails();
}
}
extruder = MMU2_NO_TOOL;
tool_change_extruder = MMU2_NO_TOOL;
MakeSound(Confirm);
return true;
}
void MMU2::Button(uint8_t index) {
LogEchoEvent_P(PSTR("Button"));
logic.Button(index);
}
void MMU2::Home(uint8_t mode) {
logic.Home(mode);
}
void MMU2::SaveHotendTemp(bool turn_off_nozzle) {
if (mmu_print_saved & SavedState::Cooldown)
return;
if (turn_off_nozzle && !(mmu_print_saved & SavedState::CooldownPending)) {
Disable_E0();
resume_hotend_temp = thermal_degTargetHotend();
mmu_print_saved |= SavedState::CooldownPending;
LogEchoEvent_P(PSTR("Heater cooldown pending"));
}
}
void MMU2::SaveAndPark(bool move_axes) {
if (mmu_print_saved == SavedState::None) { // First occurrence. Save current position, park print head, disable nozzle heater.
LogEchoEvent_P(PSTR("Saving and parking"));
Disable_E0();
planner_synchronize();
// In case a power panic happens while waiting for the user
// take a partial back up of print state into RAM (current position, etc.)
refresh_print_state_in_ram();
if (move_axes) {
mmu_print_saved |= SavedState::ParkExtruder;
resume_position = planner_current_position(); // save current pos
// lift Z
MoveRaiseZ(MMU_ERR_Z_PAUSE_LIFT);
// move XY aside
if (all_axes_homed()) {
nozzle_park();
}
}
}
// keep the motors powered forever (until some other strategy is chosen)
// @@TODO do we need that in 8bit?
gcode_reset_stepper_timeout();
}
void MMU2::ResumeHotendTemp() {
if ((mmu_print_saved & SavedState::CooldownPending)) {
// Clear the "pending" flag if we haven't cooled yet.
mmu_print_saved &= ~(SavedState::CooldownPending);
LogEchoEvent_P(PSTR("Cooldown flag cleared"));
}
if ((mmu_print_saved & SavedState::Cooldown) && resume_hotend_temp) {
LogEchoEvent_P(PSTR("Resuming Temp"));
// @@TODO MMU2_ECHO_MSGRPGM(PSTR("Restoring hotend temperature "));
SERIAL_ECHOLN(resume_hotend_temp);
mmu_print_saved &= ~(SavedState::Cooldown);
thermal_setTargetHotend(resume_hotend_temp);
FullScreenMsgRestoringTemperature();
//@todo better report the event and let the GUI do its work somewhere else
ReportErrorHookSensorLineRender();
waitForHotendTargetTemp(100, [] {
marlin_manage_inactivity(true);
mmu2.mmu_loop_inner(false);
ReportErrorHookDynamicRender();
});
ScreenUpdateEnable(); // temporary hack to stop this locking the printer...
LogEchoEvent_P(PSTR("Hotend temperature reached"));
ScreenClear();
}
}
void MMU2::ResumeUnpark() {
if (mmu_print_saved & SavedState::ParkExtruder) {
LogEchoEvent_P(PSTR("Resuming XYZ"));
// Move XY to starting position, then Z
motion_do_blocking_move_to_xy(resume_position.xyz[0], resume_position.xyz[1], feedRate_t(NOZZLE_PARK_XY_FEEDRATE));
// Move Z_AXIS to saved position
motion_do_blocking_move_to_z(resume_position.xyz[2], feedRate_t(NOZZLE_PARK_Z_FEEDRATE));
// From this point forward, power panic should not use
// the partial backup in RAM since the extruder is no
// longer in parking position
clear_print_state_in_ram();
mmu_print_saved &= ~(SavedState::ParkExtruder);
}
}
void MMU2::CheckUserInput() {
auto btn = ButtonPressed((uint16_t)lastErrorCode);
// Was a button pressed on the MMU itself instead of the LCD?
if (btn == Buttons::NoButton && lastButton != Buttons::NoButton) {
btn = lastButton;
lastButton = Buttons::NoButton; // Clear it.
}
switch (btn) {
case Left:
case Middle:
case Right:
SERIAL_ECHOPGM("CheckUserInput-btnLMR ");
SERIAL_ECHOLN(btn);
// clear the explicit printer error as soon as possible so that the MMU error screens + reporting doesn't get too confused
if (lastErrorCode == ErrorCode::LOAD_TO_EXTRUDER_FAILED) {
// A horrible hack - clear the explicit printer error allowing manage_response to recover on MMU's Finished state
// Moreover - if the MMU is currently doing something (like the LoadFilament - see comment above)
// we'll actually wait for it automagically in manage_response and after it finishes correctly,
// we'll issue another command (like toolchange)
logic.ClearPrinterError();
lastErrorCode = ErrorCode::OK;
lastErrorSource = ErrorSourceNone; // this seems to help clearing the error screen
}
ResumeHotendTemp(); // Recover the hotend temp before we attempt to do anything else...
if (mmu2.MMULastErrorSource() == MMU2::ErrorSourceMMU) {
// Do not send a button to the MMU unless the MMU is in error state
Button(btn);
}
// A quick hack: for specific error codes move the E-motor every time.
// Not sure if we can rely on the fsensor.
// Just plan the move, let the MMU take over when it is ready
switch (lastErrorCode) {
case ErrorCode::FSENSOR_DIDNT_SWITCH_OFF:
case ErrorCode::FSENSOR_TOO_EARLY:
HelpUnloadToFinda();
break;
default:
break;
}
break;
case TuneMMU:
Tune();
break;
case ResetMMU:
Reset(ResetPin); // we cannot do power cycle on the MK3
// ... but mmu2_power.cpp knows this and triggers a soft-reset instead.
break;
case DisableMMU:
Stop(); // Poweroff handles updating the EEPROM shutoff.
break;
case StopPrint:
// @@TODO not sure if we shall handle this high level operation at this spot
break;
default:
break;
}
}
/// Originally, this was used to wait for response and deal with timeout if necessary.
/// The new protocol implementation enables much nicer and intense reporting, so this method will boil down
/// just to verify the result of an issued command (which was basically the original idea)
///
/// It is closely related to mmu_loop() (which corresponds to our ProtocolLogic::Step()), which does NOT perform any blocking wait for a command to finish.
/// But - in case of an error, the command is not yet finished, but we must react accordingly - move the printhead elsewhere, stop heating, eat a cat or so.
/// That's what's being done here...
bool MMU2::manage_response(const bool move_axes, const bool turn_off_nozzle) {
mmu_print_saved = SavedState::None;
MARLIN_KEEPALIVE_STATE_IN_PROCESS;
LongTimer nozzleTimeout;
for (;;) {
// in our new implementation, we know the exact state of the MMU at any moment, we do not have to wait for a timeout
// So in this case we shall decide if the operation is:
// - still running -> wait normally in idle()
// - failed -> then do the safety moves on the printer like before
// - finished ok -> proceed with reading other commands
safe_delay_keep_alive(0); // calls LogicStep() and remembers its return status
if (mmu_print_saved & SavedState::CooldownPending) {
if (!nozzleTimeout.running()) {
nozzleTimeout.start();
LogEchoEvent_P(PSTR("Cooling Timeout started"));
} else if (nozzleTimeout.expired(DEFAULT_SAFETYTIMER_TIME_MINS * 60 * 1000ul)) { // mins->msec.
mmu_print_saved &= ~(SavedState::CooldownPending);
mmu_print_saved |= SavedState::Cooldown;
thermal_setTargetHotend(0);
LogEchoEvent_P(PSTR("Heater cooldown"));
}
} else if (nozzleTimeout.running()) {
nozzleTimeout.stop();
LogEchoEvent_P(PSTR("Cooling timer stopped"));
}
switch (logicStepLastStatus) {
case Finished:
// command/operation completed, let Marlin continue its work
// the E may have some more moves to finish - wait for them
ResumeHotendTemp();
ResumeUnpark(); // We can now travel back to the tower or wherever we were when we saved.
logic.ResetRetryAttempts(); // Reset the retry counter.
planner_synchronize();
return true;
case Interrupted:
// now what :D ... big bad ... ramming, unload, retry the whole command originally issued
return false;
case VersionMismatch: // this basically means the MMU will be disabled until reconnected
CheckUserInput();
return true;
case PrinterError:
SaveAndPark(move_axes);
SaveHotendTemp(turn_off_nozzle);
CheckUserInput();
// if button pressed "Done", return true, otherwise stay within manage_response
// Please see CheckUserInput() for details how we "leave" manage_response
break;
case CommandError:
case CommunicationTimeout:
case ProtocolError:
case ButtonPushed:
if (!logic.InAutoRetry()) {
// Don't proceed to the park/save if we are doing an autoretry.
SaveAndPark(move_axes);
SaveHotendTemp(turn_off_nozzle);
CheckUserInput();
}
break;
case CommunicationRecovered: // @@TODO communication recovered and may be an error recovered as well
// may be the logic layer can detect the change of state a respond with one "Recovered" to be handled here
ResumeHotendTemp();
ResumeUnpark();
break;
case Processing: // wait for the MMU to respond
default:
break;
}
}
}
StepStatus MMU2::LogicStep(bool reportErrors) {
CheckUserInput(); // Process any buttons before proceeding with another MMU Query
StepStatus ss = logic.Step();
switch (ss) {
case Finished:
// At this point it is safe to trigger a runout and not interrupt the MMU protocol
CheckFINDARunout();
break;
case Processing:
OnMMUProgressMsg(logic.Progress());
break;
case ButtonPushed:
lastButton = logic.Button();
LogEchoEvent_P(PSTR("MMU Button pushed"));
CheckUserInput(); // Process the button immediately
break;
case Interrupted:
// can be silently handed over to a higher layer, no processing necessary at this spot
break;
default:
if (reportErrors) {
switch (ss) {
case CommandError:
ReportError(logic.Error(), ErrorSourceMMU);
break;
case CommunicationTimeout:
state = xState::Connecting;
ReportError(ErrorCode::MMU_NOT_RESPONDING, ErrorSourcePrinter);
break;
case ProtocolError:
state = xState::Connecting;
ReportError(ErrorCode::PROTOCOL_ERROR, ErrorSourcePrinter);
break;
case VersionMismatch:
StopKeepPowered();
ReportError(ErrorCode::VERSION_MISMATCH, ErrorSourcePrinter);
break;
case PrinterError:
ReportError(logic.PrinterError(), ErrorSourcePrinter);
break;
default:
break;
}
}
}
if (logic.Running()) {
state = xState::Active;
}
return ss;
}
void MMU2::filament_ramming() {
execute_extruder_sequence(ramming_sequence, sizeof(ramming_sequence) / sizeof(E_Step));
}
void MMU2::execute_extruder_sequence(const E_Step *sequence, uint8_t steps) {
planner_synchronize();
const E_Step *step = sequence;
for (uint8_t i = steps; i ; --i) {
MoveE(pgm_read_float(&(step->extrude)), pgm_read_float(&(step->feedRate)));
step++;
}
planner_synchronize(); // it looks like it's better to sync the moves at the end - smoother move (if the sequence is not too long).
Disable_E0();
}
void MMU2::execute_load_to_nozzle_sequence() {
planner_synchronize();
// Compensate for configurable Extra Loading Distance
planner_set_current_position_E(planner_get_current_position_E() - (logic.ExtraLoadDistance() - MMU2_FILAMENT_SENSOR_POSITION));
execute_extruder_sequence(load_to_nozzle_sequence, sizeof(load_to_nozzle_sequence) / sizeof(load_to_nozzle_sequence[0]));
}
void MMU2::ReportError(ErrorCode ec, ErrorSource res) {
// Due to a potential lossy error reporting layers linked to this hook
// we'd better report everything to make sure especially the error states
// do not get lost.
// - The good news here is the fact, that the MMU reports the errors repeatedly until resolved.
// - The bad news is, that MMU not responding may repeatedly occur on printers not having the MMU at all.
//
// Not sure how to properly handle this situation, options:
// - skip reporting "MMU not responding" (at least for now)
// - report only changes of states (we can miss an error message)
// - may be some combination of MMUAvailable + UseMMU flags and decide based on their state
// Right now the filtering of MMU_NOT_RESPONDING is done in ReportErrorHook() as it is not a problem if mmu2.cpp
// Depending on the Progress code, we may want to do some action when an error occurs
switch (logic.Progress()) {
case ProgressCode::UnloadingToFinda:
unloadFilamentStarted = false;
planner_abort_queued_moves(); // Abort excess E-moves to be safe
break;
case ProgressCode::FeedingToFSensor:
// FSENSOR error during load. Make sure E-motor stops moving.
loadFilamentStarted = false;
planner_abort_queued_moves(); // Abort excess E-moves to be safe
break;
default:
break;
}
if (ec != lastErrorCode) { // deduplicate: only report changes in error codes into the log
lastErrorCode = ec;
lastErrorSource = res;
LogErrorEvent_P(_O(PrusaErrorTitle(PrusaErrorCodeIndex((uint16_t)ec))));
if (ec != ErrorCode::OK && ec != ErrorCode::FILAMENT_EJECTED) {
IncrementMMUFails();
// check if it is a "power" failure - we consider TMC-related errors as power failures
static constexpr uint16_t tmcMask =
( (uint16_t)ErrorCode::TMC_IOIN_MISMATCH
| (uint16_t)ErrorCode::TMC_RESET
| (uint16_t)ErrorCode::TMC_UNDERVOLTAGE_ON_CHARGE_PUMP
| (uint16_t)ErrorCode::TMC_SHORT_TO_GROUND
| (uint16_t)ErrorCode::TMC_OVER_TEMPERATURE_WARN
| (uint16_t)ErrorCode::TMC_OVER_TEMPERATURE_ERROR
| (uint16_t)ErrorCode::MMU_SOLDERING_NEEDS_ATTENTION ) & 0x7fffU; // skip the top bit
static_assert(tmcMask == 0x7e00); // just make sure we fail compilation if any of the TMC error codes change
if ((uint16_t)ec & tmcMask) { // @@TODO can be optimized to uint8_t operation
// TMC-related errors are from 0x8200 higher
IncrementTMCFailures();
}
}
}
if (!mmu2.RetryIfPossible((uint16_t)ec)) {
// If retry attempts are all used up
// or if 'Retry' operation is not available
// raise the MMU error sceen and wait for user input
ReportErrorHook((CommandInProgress)logic.CommandInProgress(), (uint16_t)ec, uint8_t(lastErrorSource));
}
static_assert(mmu2Magic[0] == 'M'
&& mmu2Magic[1] == 'M'
&& mmu2Magic[2] == 'U'
&& mmu2Magic[3] == '2'
&& mmu2Magic[4] == ':'
&& strlen_constexpr(mmu2Magic) == 5,
"MMU2 logging prefix mismatch, must be updated at various spots");
}
void MMU2::ReportProgress(ProgressCode pc) {
ReportProgressHook((CommandInProgress)logic.CommandInProgress(), (uint16_t)pc);
LogEchoEvent_P(_O(ProgressCodeToText((uint16_t)pc)));
}
void MMU2::OnMMUProgressMsg(ProgressCode pc) {
if (pc != lastProgressCode) {
OnMMUProgressMsgChanged(pc);
} else {
OnMMUProgressMsgSame(pc);
}
}
void MMU2::OnMMUProgressMsgChanged(ProgressCode pc) {
ReportProgress(pc);
lastProgressCode = pc;
switch (pc) {
case ProgressCode::UnloadingToFinda:
if ((CommandInProgress)logic.CommandInProgress() == CommandInProgress::UnloadFilament
|| ((CommandInProgress)logic.CommandInProgress() == CommandInProgress::ToolChange)) {
// If MK3S sent U0 command, ramming sequence takes care of releasing the filament.
// If Toolchange is done while printing, PrusaSlicer takes care of releasing the filament
// If printing is not in progress, ToolChange will issue a U0 command.
break;
} else {
// We're likely recovering from an MMU error
planner_synchronize();
unloadFilamentStarted = true;
HelpUnloadToFinda();
}
break;
case ProgressCode::FeedingToFSensor:
// prepare for the movement of the E-motor
planner_synchronize();
loadFilamentStarted = true;
break;
default:
// do nothing yet
break;
}
}
void __attribute__((noinline)) MMU2::HelpUnloadToFinda() {
MoveE(-MMU2_RETRY_UNLOAD_TO_FINDA_LENGTH, MMU2_RETRY_UNLOAD_TO_FINDA_FEED_RATE);
}
void MMU2::OnMMUProgressMsgSame(ProgressCode pc) {
switch (pc) {
case ProgressCode::UnloadingToFinda:
if (unloadFilamentStarted && !planner_any_moves()) { // Only plan a move if there is no move ongoing
switch (WhereIsFilament()) {
case FilamentState::AT_FSENSOR:
case FilamentState::IN_NOZZLE:
case FilamentState::UNAVAILABLE: // actually Unavailable makes sense as well to start the E-move to release the filament from the gears
HelpUnloadToFinda();
break;
default:
unloadFilamentStarted = false;
}
}
break;
case ProgressCode::FeedingToFSensor:
if (loadFilamentStarted) {
switch (WhereIsFilament()) {
case FilamentState::AT_FSENSOR:
// fsensor triggered, finish FeedingToExtruder state
loadFilamentStarted = false;
// Abort any excess E-move from the planner queue
planner_abort_queued_moves();
// After the MMU knows the FSENSOR is triggered it will:
// 1. Push the filament by additional 30mm (see fsensorToNozzle)
// 2. Disengage the idler and push another 2mm.
MoveE(logic.ExtraLoadDistance() + 2, logic.PulleySlowFeedRate());
break;
case FilamentState::NOT_PRESENT:
// fsensor not triggered, continue moving extruder
if (!planner_any_moves()) { // Only plan a move if there is no move ongoing
// Plan a very long move, where 'very long' is hundreds
// of millimeters. Keep in mind though the move can't be much longer
// than 450mm because the firmware will ignore too long extrusions
// for safety reasons. See PREVENT_LENGTHY_EXTRUDE.
// Use 350mm to be safely away from the prevention threshold
MoveE(350.0f, logic.PulleySlowFeedRate());
}
break;
default:
// Abort here?
break;
}
}
break;
default:
// do nothing yet
break;
}
}
} // namespace MMU2
Reference in New Issue View Git Blame Copy Permalink