Beta of updated firmware, improved Pause/Filament Chg, etc (Page 1 of 2)

It's a crime that no one has posted in appreciation, so I'd like to thank you for all of the hard work you've put into this.

Thank you Adrian.

Ok, Just synced the latest patches to Git.

Firstly, the broad changes:

• Starting to skirt the boundaries of  the 644 in size, so did some clean ups:

• I've Setup #ifdef's for Zwobble and Hysteresis patches - this is because the Zwobble adds 9k and the Hysteresis 1.9k, and both do the same thing but differently. Decide which you want, and enable it!

• All the Filament_Alarm code is setup the same, if you dont define a FILAMENT_ALARM_PIN value, it will not compile in any of the code.

• I've added some comments in the Configuration.h file of the various firmware size changes when you compile in the various options.

• Added the Filament Alarm Code

Big update is the addition of my Filament Alarm function. As described in http://www.soliforum.com/topic/2516/wip … when-idle/ this is run out of the main loop and every 2.5ms or so (to not impact timing functions otherwise) checks if the configured pin has been pulled low. If it has, it triggers the M600 command. Once triggered, if it remains active for the configured value of PAUSE_HEATER_DELAY, it will automatically shut off the heater. The loop also inspects if the pin returns to a normal state (open), and if more than 1.5mins has passed, it will reset the alarm status - this is to prevent multiple pause conditions when refeeding the filament.

This version also includes the pausing of Repetier Host when the alarm trips, or you send an M600 command manually.
The problem is there is a bug, and I'd love someone to help out troubleshooting it.. I've chased it for a day to no avail. Basically it seems a race condition of sorts occurs when I enqueue the M600 command - the M600 correctly parks the head to the configured location (based on SD2/SD3 definition) however it seems the last command from repetier is then still exectued. I've tried playing with adding delays after sending the repetier pause but it doesn't seem to help.

Anyway, M600 works otherwise as expected, and it shuts off the extruder motor allowing rapid filament changeover etc (as demonstrated in http://www.soliforum.com/topic/2514/dem … 0-seconds/ ).

M601 works as expected, and as stated, you simply need to enter M601 before clicking Continue in repetier and it all works as expected.

About 2 out of 5 times the code manages to cleanly enqueue the M600 and the head stays parked - at the moment I re-home the heads a second time manually...

If we can nut out the bug in my code when I first enqueue M600, then this will be an awesome function! At the moment, I manually home the head the second time.

To use this directly from repetier-host, just send an M600 from the console - it'll auto pause repetier for you as well.

Demo of M602 and the Filament Alarm:

Great job! Sorry for not having much time to play with it at the moment (I am struggling with MkIV and JHead and also I need to spend some time at my real job ). I'll give a hand as soon (and as much) as I can!

Cheers all! Keen still for feedback/assistance.

I'm going to spend most of tomorrow I think bashing the code again.. so any input would be awesome

Have you assigned pins already on the ramps board for the filament alarm? I just got a couple micro switches in that would be perfect.

Awesome, thanks!

(I know 53 was a typo) 63 works as would D1 and D57. I think 63 is a good choice.

I (think)am getting rid of my fans and going to small peltier cooling pads. Any chance those 5v pins next to D1 and D 57 could handle 3A max draw? If so I could just use these instead and forget about the mosfets.

http://www.ebay.com/itm/DC-6-2V-6-16W-3 … 0787605300

{{BUMP}}

Okay, My filament change is doing the same thing that you described. It's kind of disturbing sometimes because my extruder looks like it is possessed. I wish I had taken some video of it because the hobbed gear was going full speed and the carriage looked like it was just floating around my bed.

Have you tried sending a "buffer" command before m600. Maybe something like an endstop check M119 then M600. Would that help hook the command properly?

Forgive my ignorance (I am not a programmer )

I'm interested! was just waiting for the final release.

Nice setup on the filament switch

Moving the extruder should be a non issue - before it releases the head it saves the last print position. an M601 will return it to the correct location - ideally this will be a menu option off the LCD.. and as for Repetier-Host... not sure on how I'll tackle that bit besides the present advice of 'Issue M601 before clicking Continue'.

LCD Buzzer yes, it should work, but I had to tweak some stuff there which has meant its kind of flakey at the moment. In programming terms, the previous Buzzer Code was 'blocking' - not what you want when you are relying on the execution of further code.  It'll be back next upload to GitHub.

One of the tabs broke on my round end adapter. I need to use the other hole anyways which requires me to move some cables. I just haven't felt like rewiring everything.

the tab might have broken because of the weight. I am going to reprint with my black filament. It seems to he a bit more stout and better suited for the application.