Filament Run-out Sensor Add-on

Topics around mechanical design, controller and electronics. Mods & hacks welcome.
User avatar
Vicky@Raise3D
Posts: 3495
Joined: Fri Mar 25, 2016 3:54 am

Filament Run-out Sensor Add-on

Postby Vicky@Raise3D » Mon Jan 22, 2018 6:57 am

Dear All:

N Series owners can now pre-order the filament run-out sensor and get for free a thermal protector.

Learn more: https://www.raise3d.com/pages/filament-run-out-sensor-add-on
1.png

EldRick
Posts: 269
Joined: Thu Jun 22, 2017 5:55 am

Re: Filament Run-out Sensor Add-on

Postby EldRick » Mon Jan 22, 2018 5:09 pm

Many of us will need an adapter for Bondtech extruders. That's the only reason I haven't ordered the filament sensor yet.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Mon Jan 22, 2018 5:33 pm

My main complaint here is the thermal protection should be a free or near free separate add on. There should be instructions and open source for mass producing it, it's 2 resistors and the connection cables, basically the identical copy of the hardware mod just plug into the mainboard. Don't get me wrong, the design is nice with locking connectors, but tying this to the runout sensor and a total cost of $89, this isn't right. We need to get serious about promoting safety and this is not the right answer. I'm not saying customers cannot pay for it, I understand the cost implications here but somehow there needs to be a middle ground.

User avatar
Vicky@Raise3D
Posts: 3495
Joined: Fri Mar 25, 2016 3:54 am

Re: Filament Run-out Sensor Add-on

Postby Vicky@Raise3D » Tue Jan 23, 2018 3:16 am

Thank you for your feedback! As you know, we do value it.
Our team has analysed different options to make the Thermal-Protector available for those who may need it. We perfectly understand that some users may be heavy users but still do not need the Filament Run-Out Sensor Add-On.
The back-office management will be different and we are still analyzing the best way to address the needs of everyone. In our plan, this analysis will become a high priority in late March.
If you think that this is not acceptable in your case (as anyone else reading this post), please send me a PM and we will try to find an ad-hoc solution for you.
I am sorry that we cannot do better at this stage

User avatar
Vicky@Raise3D
Posts: 3495
Joined: Fri Mar 25, 2016 3:54 am

Re: Filament Run-out Sensor Add-on

Postby Vicky@Raise3D » Tue Jan 23, 2018 4:55 am

EldRick wrote:Many of us will need an adapter for Bondtech extruders. That's the only reason I haven't ordered the filament sensor yet.


May I ask what will be the adapter used for? The sensor will be located between the filament spool holders and the feeding tube. Should have no relationship to the extruder itself.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 5:22 am

DIY style $10 for a pair of detectors. https://www.amazon.com/MagiDeal-Consuma ... B074ZB1B2Z

Iamkar33m
Posts: 53
Joined: Fri Dec 01, 2017 2:53 pm

Re: Filament Run-out Sensor Add-on

Postby Iamkar33m » Tue Jan 23, 2018 6:39 am

Jetguy wrote:DIY style $10 for a pair of detectors. https://www.amazon.com/MagiDeal-Consuma ... B074ZB1B2Z


How would that connect to our printers?

sylus
Posts: 100
Joined: Fri Aug 05, 2016 9:47 pm

Re: Filament Run-out Sensor Add-on

Postby sylus » Tue Jan 23, 2018 11:48 am

Jetguy wrote:My main complaint here is the thermal protection should be a free or near free separate add on. There should be instructions and open source for mass producing it, it's 2 resistors and the connection cables, basically the identical copy of the hardware mod just plug into the mainboard. Don't get me wrong, the design is nice with locking connectors, but tying this to the runout sensor and a total cost of $89, this isn't right. We need to get serious about promoting safety and this is not the right answer. I'm not saying customers cannot pay for it, I understand the cost implications here but somehow there needs to be a middle ground.


Totally agree,if there is any risk regarding safety ,supplier have the obligation to propose to customer an software or hardware patch to avoid any sf
safety related issue.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 1:42 pm

Iamkar33m wrote:
Jetguy wrote:DIY style $10 for a pair of detectors. https://www.amazon.com/MagiDeal-Consuma ... B074ZB1B2Z


How would that connect to our printers?


The same basic way the one Raise 3D is supplying there may be minor differences, but not anything so drastic. The sensor is mounted at the entrance to the guide tubes near the spools, the sensor is nothing more than an end stop switch inside the housing to detect filament is present in the path through the device. If you read the firmware thread it gave the details of how the switch is connected. viewtopic.php?f=2&t=593&p=16495&hilit=filament+monitor#p16495

Yes, these DIY versions for sale I linked are currently switches with all 3 wires so on the mainboard plug you need to make it a 2 wire plug. The switch has 3 wires, Common, Normally Open, Normally Closed. "Please use sensors normally closed for this. We define Close as no filament, On as has filament there." So again, per their statement, you determine the Common and Normally Closed wires, put them in a 2 pin connector and plug into the end stop ports on the mainboard per the diagram. An enterprising person could purchase these off the shelf, buy a JST-XH 2 pin connector shell, and spend all of 5 minutes of time to put the 2 wires into 2 wire version VS the 3 wire version- then sell to you folks for markup if you want true plug and play. Again, just the reason for linking these is to give you folks an understanding of what is possible, what the technical details are so that you can properly evaluate and understand what Raise3D is selling. Part of the logic here is that you are buying a "solution" that is tested, proven, approved, and is worth the $$$$. I have no doubt the Raise3D system works, I have no doubts there was testing done. The Raise 3D version is using optical non contact sensing VS a mechanical switch lever obviously rubbing against the filament. The question is, in a long term test with dozens if not hundreds of spools of filament run through one of these what wear, what failures, what false detection or other issues happens? For $89, it should look and feel like $89. By comparison, the $10 version at least has "the right stuff" that you the user even buying switches and 3D printing them, to assemble and your time is worth $10.


All this said, one thing that is more feedback for the Raise3D version is concern that the filament path entrance and exit need wear guides and cannot just depend on a 3D printed or molded hole. This is because filament passing through from a spool has a bend or curve in it and will rapidly wear the hole larger and potentially wear against the optical sensor given enough filament passing through. This is why the $10 version I linked has brass inserts for this very reason of concerns of wear over time.
Attachments
filament_runout_sensor_02_1024x1024.png
filament_runout_sensor_02-1_1024x1024.png
51HwlECw4dL._SL1024_.jpg

EldRick
Posts: 269
Joined: Thu Jun 22, 2017 5:55 am

Re: Filament Run-out Sensor Add-on

Postby EldRick » Tue Jan 23, 2018 4:57 pm

"May I ask what will be the adapter used for?"
As I understand it, your filament sensor appears to be designed to mount on top of the N2 printhead.
With the Bondtech extruders, you will need an adapter of some sort to mount it on their extruder.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 6:06 pm

EldRick wrote:"May I ask what will be the adapter used for?"
As I understand it, your filament sensor appears to be designed to mount on top of the N2 printhead.
With the Bondtech extruders, you will need an adapter of some sort to mount it on their extruder.


The detector is NOT mounted to the extruder head (and wiring to it would be a problem anyway) it mounts in the location near the spools and entrance from the spools to the vertical guide tubes thus placing it nearer the electronics bay and easier to wire.

Again, the first assumption was that it mounted to the head and that is false. Second problem is, if you did mount it to the head then you have to ALSO install wiring to it. Third problem is, if the distance between the sensor and the intake of the feeder is too short, by the time a lack of filament is detected and pause is invoked you may have well run out of filament. Pause must run the last segment of gcode to completion. Depending on how long that segment is, you easily could run out of filament if the detector is only mm above the drive gears of the feeder.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Filament Run-out Sensor Add-on

Postby Squenz » Tue Jan 23, 2018 6:15 pm

My very first question would be, what exactly happens the moment when the filament triggers detects a runout ?

Printer goes in "Pause" like hitting the button on the touch screen or does it STOP printing right away ?
If it stops, does the head drive to X0, Y0 and waits for a user to respond ?

And then, how's the proceeding, pull out the rest of the filament from the old spool, feed in from a new spool and then hit a button to resume ?

I'd be interested, but not enough information to decide if i need that gimmick.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 6:34 pm

"what exactly happens the moment when the filament triggers detects a runout ?"
Answer: the Marlin motion control firmware sends a gcode response across the USB cable to the front panel LCD that is streaming the gcode print file. The logic then in the latest update of the front panel Raisepack LCD touch firmware makes a decision to enter a triggered pause function.

ABH
Posts: 168
Joined: Sun Aug 14, 2016 7:31 pm

Re: Filament Run-out Sensor Add-on

Postby ABH » Tue Jan 23, 2018 8:20 pm

Hmm.., the Marlin "FILAMENT_RUNOUT_SENSOR" feature isn't enabled in the official Raise3D motion controller firmware, so I don't see how this will work? Isn't the sensor connected to the Touchscreen computer?

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 8:34 pm

Well the GPIO pins face outward so plugging anything in is kind of a hacky official solution right? It's not a USB sensor- that's pretty obvious from looking at the circuit board. Pause was already implemented via gcode response, so an about face and suddenly then connecting the sensor to the pcduino is a complete reversal of all previous work to this point.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Filament Run-out Sensor Add-on

Postby Squenz » Tue Jan 23, 2018 9:48 pm

Jetguy wrote:"what exactly happens the moment when the filament triggers detects a runout ?"
Answer: the Marlin motion control firmware sends a gcode response across the USB cable to the front panel LCD that is streaming the gcode print file. The logic then in the latest update of the front panel Raisepack LCD touch firmware makes a decision to enter a triggered pause function.


I'm sorry, i'm not too famliar of Firmware/USB connection to the Mainboard and such on these machines.

If i get this info right, there's a look ahead buffer on the main board and this initiated gcode pause will be infused into the GCode stream the moment the Runout signal is triggered.

The machine will continue to print until the buffer is empty and the pause-code is the next executed gcode-line. Then it will pause and drive the print head to a designated position where i can change the filament and prep the continuance. When the fresh filament is ready to go, the pause/resume button on the LCD will start heating and then start to fill the buffer again.

Does that sound like the correct description ?

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 10:41 pm

Correct, however key point- there is no pause gcode. By that I mean from a motion control firmware perspective we are not using a firmware based pause command that would happen when you print on traditional 3D printer architecture from an SD card that is direct local storage to the processor. This is a print server style system where the motion board is blindly following whatever gcode you send it. A pause is simply lack of further new motion commands being sent. Since you didn't know this, it's imperative that you then are also told of a known issue in the motion firmware version 1.1.1 where there is a stepper timeout- and this can happen if you enter a pause and failed to update the motion control firmware- boom- you have an unexpected ruined print. You have 3 choices. Never pause ever, update to firmware at least firmware 1.1.6rev1, or compile custom firmware from source code and ensure you disable that setting in the source. https://www.raise3d.com/pages/download

The expectation is that Raise 3D when they provide the filament sensor will provide an appropriate Marlin firmware file to also flash, and maybe yet another LCD touch update as a matched pair. Until then, details again in this thread document DIY versions and community supported functionality viewtopic.php?f=2&t=593&start=60#p21416 based on the comments, I have to think user Sylus is testing or running with a filament detector now.

Squenz
Posts: 140
Joined: Fri Dec 01, 2017 11:18 pm
Location: Stuttgart/Germany

Re: Filament Run-out Sensor Add-on

Postby Squenz » Tue Jan 23, 2018 11:11 pm

Jetguy, as always, very good information, thank you for pointing things out.


Do you happen to know what size of buffer we're dealing with ?

From what I'd observed when pausing a print over the display, i'd venture the guess of it beeing around 32 or 64 lines of code.




On a sidenote, mid december, when i unpacked the machine, first thing after confirming basic operation, i updated everything to the current stable versions. So, i'm almost recent, because i didn't do the latest update to the Display 0.9.7, yet. Might happen over the coming weekend.

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 11:30 pm

Good read http://3digitalcooks.com/2014/12/marlin-planner-101/
and also as to the specific question of depth https://github.com/MarlinFirmware/Marlin/issues/4828
// The number of linear motions that can be in the plan at any give time.
// THE BLOCK_BUFFER_SIZE NEEDS TO BE A POWER OF 2, i.g. 8,16,32 because shifts and ors are used to do the ring-buffering.
#if ENABLED(SDSUPPORT)
#define BLOCK_BUFFER_SIZE 16 // SD,LCD,Buttons take more memory, block buffer needs to be smaller
#else
#define BLOCK_BUFFER_SIZE 16 // maximize block buffer
#endif


Note, I did not grab that from the official Raise 3D Marlin source, just pointing out a quick Google search on the topic. Probably search for that parameter in the source code, and then we know for sure. https://github.com/Raise3D/Marlin-Raise ... tion_adv.h

That said, it's 16

Jetguy
Posts: 2352
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament Run-out Sensor Add-on

Postby Jetguy » Tue Jan 23, 2018 11:53 pm

Sorry I know I've destroyed this thread. For those who don't know, my stepfather passed away the end of November. My entire life was kind of turned upside down and I traveled home to help my mother out. As such, I was completely away from 3D printing and Raise 3D until just recently when I finally returned home.

My intent here is to see you fellow users get the best features and support and safety we can from this hardware. As such, the firmware, the motion control board, filament monitoring, and thermal safety have been ongoing and active deep technical discussions for some time. During the last couple of months, we have a lot of newer members.

What I want to say is, I'm happy for seeing Raise 3D finally taking real orders on a filament sensor and the safety board. While I'm giving them heck for it, the design of both parts is perfectly fine, it's the how they are selling it and the cost to perceived value. At the same time, this was and has been a community of open source contribution and when we get a situation where Raise 3D doesn't produce something to our liking, we have the power of open source to go the DIY route.

I'm hoping this does 3 major things:
#1 puts some pressure on Raise 3D to consider pricing and how this is bundled
#2 Improve the actual sensor (details like wear guides and other details)
#3 educate the users on known related issues now that exist in certain versions of firmware related to a pause function (firmware 1.1.1 stepper timeout), educate users on the functional underlying system(dual boards talking to each other with a limited protocol) so they can better use, operate, and even troubleshoot the printer, and last, help shape and educate on the limits that exist based on the architecture, but also how to take that knowledge and make BETTER suggestions and improvements.


Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 4 guests