Filament runout sensor kills printing -- fixed

Topics around mechanical design, controller and electronics. Mods & hacks welcome.
User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Filament runout sensor kills printing -- fixed

Postby JohnSays » Sat May 19, 2018 3:05 am

My install of the sensor and wiring went smoothly. Before installing the sensor, I pushed some filament into it. I was surprised by the amount of resistance and wondered how well my Bondtec BMG was going to handle all that resistance. Well, it can't. No matter how tight I make the thumb screw I get lots of skips. Basically, at very low speeds like 35 mm/sec, I can print with ABS. At 70mm/sec I get about 2/3 of my print and the reats is air.

So not good. This brings up another question: How on earth are we now supposed to print flexibles? No way could any flexible filament be pulled through that sensor. I'm taking it off tonight as I simply cannot print with it on there.

One thing of note is that the filament is 1.79mm. I pushed some filament that is 1.72mm through and it did slide easier -- although still with too much resistance IMO.

Has anyone else had this problem? Is mine broken?
Last edited by JohnSays on Sat May 19, 2018 5:02 am, edited 1 time in total.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

Jetguy
Posts: 2700
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Filament runout sensor kills printing

Postby Jetguy » Sat May 19, 2018 3:16 am

I think you missed this post about printing an alternate cover for the sensor?
viewtopic.php?f=4&t=8185#p30776
For the friction of loading filament, we just developed a new cover based on users' suggestions. If possible, would you like to print it out and give a try?
Attachments
Filament Run-out Sensor Back Cover Upgraded Ver.STL
(408.38 KiB) Downloaded 62 times

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing

Postby JohnSays » Sat May 19, 2018 3:53 am

More data: So I've been playing with the sensor off of the machine. I noticed that there was white filament powder nearly filling the input hole. I cleaned that out and pushed some filament in and it too left white powder. Because I am near the end of a spool, my filament has a curve to it. What I see happening is that the edge of the inside hole of the sensor is scraping the filament. That hole edge is sharp enough to dig into the filament and tug on it. That edge needs to have a chamfer.

How did this get into our hands without this being caught? That's a pretty elementary feature.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing

Postby JohnSays » Sat May 19, 2018 3:54 am

Thanks JetGuy. Honestly though, that seems a bit extreme. Couldn't the input hole be tapered a bit?
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Sat May 19, 2018 5:02 am

Well I went ahead and modified my sensor block by chamfering the input holes. First I disassembled it. Then I took a 1/4" carbide bit for boring a hole in glass (Amazon: https://www.amazon.com/gp/product/B002FY8ZQA) and used my fingers to twist it in the hole. Because of the nice long taper of the bit, it only chamfered the filament hole. I twisted it several times until I could plainly see a chamfer form. Then I took a Dremel bit -- the smallest tapering aluminum oxide stone I could find and which is about 3/16" diameter tapering to about 1/32" -- and twisted that in the hole using my drill press. I went around the hole several times putting a polish and finish round on any potentially sharp edge. Lastly I blew it out with compressed air and reassembled it.

It is now back in my printer and I am printing at 60mm/sec with no problem. Before I reinstalled it, I ran filament in and out of it and it did not drag any more than the internal friction that is expected. The edge of the input hole no longer grabs the filament and no filament dust is forming there.

So I am calling this a fix.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Sat May 19, 2018 5:24 pm

Next subject (RANT WARNING): Sensor does not work. Software downloaded and installed:frs version of 1.1.6 and latest LCD software. Check. I double and triple checked my wiring against the docs. Then I went online to see if I could get a better look at the pictures by expanding them. I then look at the N1 pictures by mistake (I have two N2's) and see that the green and white wires are in exactly the opposite connectors. So I try that on the N2 and it works... Honestly, would it be too much effort to email us about this mistake or to revise the docs online?! Every single person who downloads and uses these docs is going to have the same problem I had. This is again so sloppy. Sloppy assembly of the machines and sloppy customer support that causes hours to days of work for us. I do not see how this behavior is raising the standard. One more thing, the 14 pin connector appears to be wired upside down. The notch should be installed to the left per the picture hint on the board. However, IF I am seeing the picture correctly the wires go to the top of the connector; which places the notch to the right. Not sure because nobody made sure the photo was (a) big enough or (b) bright enough. More sloppiness.

Yes I am PO'd because this sort of slop (substitute preferred invective) totally throws my day off -- AGAIN. So instead of getting a print going and handling my list of things to do today, I am still trying to get R3D (redacted) to work. I was up until 1:30am making the the sensor fit for use by chamfering the hole and testing it. I paid $89 for this sensor kit and now I find out I could have gotten better from Amazon and JetGuy's instructions. For $89 I think we should get a working kit that includes workable -- read correct -- instructions. In my software company I always always always tested the documentation against the software. I had a person in my Help Dept just for the purpose of testing the documentation.

And just as a little teeny weeny aside and speaking of testing: I cannot believe that anodized aluminum sensor block was fully tested. Surely any experienced printer operator would have caught that the filament was being scraped and jamming going into the sensor block. Now the "solution" is to 3D print another workaround. No. Fix the problem. Email or message the buyers that they have a faulty product and here's what to do. Bing bang. Don't let us flounder and waste our time each one of us finding out that the product is faulty or the docs are faulty, That is not Raising the Standard. If R3D does not think that improving communication should be part of their efforts, then they will not get past middling in this market.

R3D you need to test your documentation against the hardware. PLEASE!

End of rant -- for now.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

User avatar
3DWP
Posts: 69
Joined: Fri Dec 15, 2017 5:48 pm

Re: Filament runout sensor kills printing -- fixed

Postby 3DWP » Wed Aug 01, 2018 11:20 am

I just installed a filament sensor, same issues! The filament is stripped by the metal - there's a ball of plastic at the entrance.

Also, my sensor seems not to work even though I think I followed the manual very closely - probably left and right mixed up..

I'm also not amused that this is still not adressed properly, I ordered from an official reseller and did not gat any info with it. My version did have the cable/adapter next to the first cable that now apparently isn't used.

Why is the manual still not changed? And why should I print and change my just received filament sensor? I should have gotten the latest proper working version :x

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Thu Aug 02, 2018 9:28 pm

Yes, well read the above and you too can fix up your own sensor. You do not have to print anything. Just chamfer the hole.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

AJM
Posts: 37
Joined: Tue May 02, 2017 11:20 pm

Re: Filament runout sensor -- worse than the cure

Postby AJM » Sat Sep 08, 2018 3:27 pm

I made the modification to enlarge the opening and the path, yet the sensor is causing more trouble than it is worth.

1. It still adds significant resistance in pulling the filament and increased the chance of breakage at the extruder motor.

2. The filament scraping, however slight, accumulates near the sensor window and eventually blocks the light which makes the sensor blind to the breakage. The indicator lights are not visible after the installation, so you cannot tell that the sensor is no longer functioning and results in failed print. I had to cut out the side of the printer to be able to see the indicators.

3. Some filaments are bent at the end of the reel to facilitate winding. The sensor path is still too long and narrow to allow the remaining filament to go through without getting stuck, thus the filament is breaking *after* the sensor. So the printer never sees the breakage.

Overall, this sensor implementation is worse than not having one. I would not recommend purchasing it as it is now.
A better approach would have been to use a sensor like a laser mouse which checks for filament movement. It only needs to check to see if movement matches approximate extrusion (G0/G1) within a certain time period. The movement check only needs to be approximate because there will be times when there is no movement at the sensor because of slack in the filament. This setup will detect filament running out as well as breakage anywhere.

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Sat Sep 08, 2018 3:39 pm

AJM, Did you chamfer the entrance to the filament hole? Enlarging it will not help.

Any 1.75 filament I put through mine slides easily without more than a very small -- insignificant -- friction. Secondly, I would not put bent filament through my machine. That is just asking for trouble at various spots. I just cut off the bent stuff at the end of the spool where it sticks out.
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

AJM
Posts: 37
Joined: Tue May 02, 2017 11:20 pm

Re: Filament runout sensor kills printing -- fixed

Postby AJM » Sun Sep 09, 2018 10:56 am

John,
The path was enlarged as well as chamfered using the tool you had mentioned.
Some filaments go though fine, but the stiff ones come off the reel with curvature and do not go through the long restrictive path of the sensor. Enlarging, not chamfer, is what helped the most for me, but still not enough.

And I have no control over the bend at the end of the reel because that portion cannot be seen until it runs off. Yes, you can cut the end off on many reels, but it depends on the reel. The bends do not cause a problem and go through the printer fine without the sensor.

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Tue Sep 18, 2018 4:51 pm

AJM wrote:John,
The path was enlarged as well as chamfered using the tool you had mentioned.
Some filaments go though fine, but the stiff ones come off the reel with curvature and do not go through the long restrictive path of the sensor. Enlarging, not chamfer, is what helped the most for me, but still not enough.

And I have no control over the bend at the end of the reel because that portion cannot be seen until it runs off. Yes, you can cut the end off on many reels, but it depends on the reel. The bends do not cause a problem and go through the printer fine without the sensor.


Okay, I get that. It just occurred to me that we may be mounting the sensors in different positions. Here's where I mount mine:

IMG_8900.JPG
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+

AJM
Posts: 37
Joined: Tue May 02, 2017 11:20 pm

Re: Filament runout sensor kills printing -- fixed

Postby AJM » Tue Sep 18, 2018 7:44 pm

John, you are right. Mine was installed a few millimeters higher and that created a sharper bend in the guide tube which created more resistance.
It was mounted higher for a couple of reasons:
1. I had made holes in the chassis to be able to see the LED indicators. Kind of silly to hide indicators!
2. That allowed mounting via the upper screw instead of using a double-sided tape. Being able to remove the sensor easily is important in cleaning the path of debris from filament scraping. which has preventing its actuation.

Anyway, the sensor is still not working properly because of false triggering with clear filaments. It is unpredictable when it thinks the filament is out, and there is also an apparent software bug because it needs multiple button actuation to resume printing. Yet another bug is that a sub-second filament-out condition makes it think the filament is out when it is just a temporary bad sensor read. All in all, this is not worth the trouble or the cost with more failed print than not using it. Very disappointing. I am going to use a Palette 2 instead.

User avatar
JohnSays
Posts: 637
Joined: Tue Apr 26, 2016 2:42 pm
Location: Saint Petersburg, FL

Re: Filament runout sensor kills printing -- fixed

Postby JohnSays » Tue Sep 18, 2018 7:51 pm

Wow! You have had a bad time with this. I haven't tried clear filament, so thanks for the heads up. I have a Palette+ for sale that I never used except to set it up. I was going to use it to print support and then I realized I can use the same filament I am printing my object with to do support. Dope slap. Duh!
- John
2 Raise3D N2 Duals, V2 nozzles, Bondtech BMGs, adjustable table, Panucatt SD2224 drivers, run-out sensor, thermal overload protection resistors on motion controller board, Firmware 1.1.8ABH - with Lin_Advance, Palette+


Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 8 guests