Print breaking

netzmark

Hi Everyone,

Everything was ok for long time and rapidly I'm faced on some strange situation.

When I start printing

- the first line is printed on the edge of the bed in accordance to the start g-code,

- on the displays appears a screen with pause/stop switches instead of normal progress bar

- first 5-6cm of the brim is printed then head lifts up and moves to the left edge of the bed

- and the buzzer starts screaming

- display looks like not working and the only I can do is turn off the power

- when turn the power back on, get message that missing system files. But after ignore or cancel I can get access to every usb files.

This is not "sometimes", it's at every trying to print anything.

Filament sensor works properly, it was the first that I thought it can be.

Tried with few different files to print thinking that in the code there is something realy strange, but it's not this.

Is it something typical?? What can I check to find an issue?

(this is Marlin 2.0 but really it was working properly with not any one accident through last whole week)

Samuel Pinches

How do you know the filament sensor is working ok? I think the LCD module is misbehaving.

netzmark

Hi Sam,
This is ordinary microswitch giving cirquit closed when the filament is in and being opened when filament is missing. It's trivial to check with any multimeter. II see the cirquit is closed so telling to LCD that filament is ok.

Samuel Pinches

Yes, the function of the switch is almost never the problem.

The problem is the LCD board does not respond to the signal correctly. 

I suggest to rotate the connector 180deg, if not just unplug it.

DaHai has a great video on youtube about it. Search the forum on this topic too.

netzmark

Sam I always try to find the answers before writing anywhere.

I have seen DaHail video but is useless in my cases. I have this connector broken and the cables soldered.

Regarding to Aetek investigation, I don't believe that some noises have increased so rapidly that need extra pullup, if it worked throgh two years.

Disconneting the connector tells the filament is constantly off, so I don't understand how it may work then, does the input react on the signal edge not the level?

I will investigate it yet so. Can we disable this sensor in the LCD firmware?

netzmark

Uff, I have found the reason and it's already working.

I don't know "how" but it happens that my sensor got disconnected. The cables of the sensor are leaded along the Z-motor cables inside the plastic shielding. Inside this shielding, some 5-8cm from the motor there is a conector of this sensor that I have found disattached.

Regarding to @AETEK advice with pullup resistor (https://jgmakerforum.com/discussion/214/how-the-filament-out-sensor-works#latest) I don't understand the sense. When the filament is "in" the sensor is closing the sensor output to the ground, the same the pullup resistor gets connected just between +3.3 and GND (because OUT=GND in this state). So it's doing nothing except the current charging (throught the resistor to GND). The resisor may help only in case of faster reation for the filament ending and the sensor output opening then.

Regarding to the sensor programmable blocking I have found in the TFT35 config code following option in "Advanced Function Options":

#the function of power detecting and filament detecting(1:disable;0:enable)
>cfg_mask_PB0_PB1_function:0

It's here https://github.com/makerbase-mks/MKS-TFT/issues/324

Of course I don't know is it served by TFT28 firmware code and I didn't tested it since have the sensor repaired.

Sam, If you have a good will, you could refresh download links to the LCD firmwares, they don't work here:

https://jgaurorawiki.com/a5/lcd-firmware

However, thank you for your being sure it is something wrong whith the sensor anyway. Hard to understand why this shity firmware can't just display "filament finished" instead of what it's doing rather pointing on USB communication problem.

Samuel Pinches

Hi @netzmark - I thought you would have seen it. Personally, I have approximately 0% confidence in the reliability of the MKS LCD module. I don't fully understand the trigger mechanism either.

If you go download the LCD custom firmware, one of the files is mks_config.txt

In there you can change the line:

>cfg_disable_MT_det:0

Change the 0 to 1, save the file, and then flash the LCD firmware. It should then ignore the filament sensor.

Best,

Sam

netzmark

I haven't seen this declaration but really see this now when you told me, thx.

However it's already fixed by hardware luckily, so no need software disabling on the moment.

About the mentioned links to download, for everyday I use Firefox and links don't work for me. But works normally when using Chrome. Strange a bit but works!

brgds

Marek

Samuel Pinches

Apologies @netzmark - Dropbox has mucked up my account, and I have contacted them. Unfortunately dropbox links are broken till further notice. Hope to have this restored in the next few days. Please post if you need a file urgently.