Ultimaker S5 - one material appears to be empty

[Shadowman 53]

3 hours ago, PFolland said:

Ok an update on my situation.

 

It turns out the feeder for the PVA filament was shedding some material as it passed through feeder mechanism. This in turn somehow clogged/blocked the flow sensor from detecting the material.

 

I simply opened up feeder 2 and cleaned it out.

 

All works fine now, been printing for 3 days non stop.

 

Thank you for your update

[Zel_man 0]

I just started to receive this error message after upgrading to Firmware 5.2.11.2010503.

Printing with Ultimaker Nylon on extruder 1, seems to happen about 3 - 3.5 hours into the print.  

 

Any solutions beyond disabling the filament sensor?

[robobartz 3]

We have the same error. I agree it should display which extruder it thinks is out of material. (even though neither of them are)

We are getting frequent error: “One material appears to be empty”.  The error is always wrong because there is plenty of material. This is extremely frustrating and making 3d prints take 5X longer because the error happens after we go home and the unit sits all night idle losing 10-12 hours of print time. 

 

 

Is this as simple as just swapping out the flow sensor? Looks like 1 screw & 1 plug.

 

Here is what we have tried/observed:

1.       https://ultimaker.com/en/resources/52577-flow-sensor-error

2.       Upgrading firmware

3.       Took apart and cleaned out the material feeder blew with light air (breath)

4.       Blew out the Bowden tube to remove any residual dust.

5.       Ensured the material is not binding

6.       Measured the material: It is perfectly in spec at 2.85MM

7.       Happens on multiple jobs

8.       Message occurs at random time intervals ½ hour to many hours.

9.       When the error happens it is using black PLA on extruder 1. Ultimaker brand PVA is in the other.

10.   We are printing over the network. I wonder if  printing using USB make a difference?

[gr5 2,167]

I assume someone mentioned it above but you can disable the sensors if you want.  It's somewhere in the settings right on the printer screen.

 

Also if you type in the ip address of your printer into a browser there are a few things you can click on including the "temperature graph" which also shows sensor position.  If you leave that graph visible while it prints it is supposed to cylce from 0 to 4095 and instantly back to 0 (a sawtooth pattern).  This may help you see when the sensor is slipping.  It may be easier to diagnose if you can see partial slipping (that don't cause the "empty" message).

[aag 6]

I found out what happens to the PVA extruder. When you insert a new PVA, sometimes the extruder has a huge resistance. Then the gripper wheels slide over the filament, and everything comes to a halt. My solution is to release the gripper, and push the PVA manually until it starts extruding. At that point, I engage the gripper, and then everything works fine.

[cboath 0]

Having the same problem here as well. New S5 set up in May. 

Our firmware is on the latest as well.

 

Putting the PVA in a dryer overnight to see if it helps at all. I'm also wondering if it could be a bad spool of PVA as i've cut sections of it at recently as 30 minutes ago and there are areas with excess material on them. Short of unspooling the whole thing, not sure how to adequately address that as a potential issue.

[aag 6]

try my "manual push" trick. It will cost you 10 seconds, and it is likely to solve your issue,

[cboath 0]

Would that still apply when the error doesn't kick in for 3ish hours after the print begins?

[gr5 2,167]

If the filament has been ground down, then yes that fixes the problem.  However if you hit "continue" and it starts printing just fine then pushing the filament through the feeder (with the lever up) will not help.

 

Did you try looking at the "temperature graph"?  It can help you realize which feeder is the problem for one thing.  You can also see the problem before it stops the printer.

[cboath 0]

I put the PVA in a 'PrintDry' and am letting it sit in there knocking the humidity down to 40.

 

Looking at the graph, temperature0 is a long solid orange line, with a bit of fluctuation at the end. Temperature1 is green and is jagged all over the place. There's a log of jaggyness on the BedTemperature line as well. The blue seems to get worse when the green is calm and vice versa.

Edited July 1, 2019 by cboath
Added image

[robobartz 3]

Ok we have solved this issue for our company and are 95% sure of what the two causes were:

It may or may not apply to your situation.

We were getting the error every 1/2 hour to 3 hours. Since we did this we have not seen the error in 2 days of printing.

 

CAUSE 1 (80%)

Our Ultimaker 5S is on a counter that is too narrow to allow the dual spool spindle in the back to be used. Therefore, we have separate spools on each side. At first we positioned, each spool on the side of the bowden tube that it used. In other words, if the PLA material is on the right, we put the PLA spool on the right. The PVA material was on the left, so we put the PVA spool on the left. The dual factory spindle(which we were not using) is in the center of back. Our side spools both feed from the outside in. This changed the angle/curve at which the material enters the material feed motors and possibly skewed the material away from the sensors.

 

Therefore, if you have external spools, put them on the opposite side of the Bowden tube which the feed into, so the 2 filaments criss-cross in back of the printer. This will mimic the intended setup.

 

CAUSE 2 (20%)

We have a 5 Killogram spool left over from the previous employees. Our toolmaker made a nice plastic holder for it, however the axles he made turned too hard. So the feed motor (designed for a 1k spool)  was unwinding a 5k spool with more resistance. We remade the axle to use a slippery undersized UHMW material and it spins freely enough to overcome the weight of the spool. Also the spool from a 3rd party was wound very poorly with criss-crossing strands and binding at the sides of the spool. 

Edited July 1, 2019 by robobartz

[cboath 0]

I see this is tagged solved, but it doesn't seem as if there's a de facto solution anywhere in here.

 

I'm seeing promising results using the PrintDry, however, that has it's own issues. I'd hardly consider it the catch all solution though. Provided it actually works.

[robobartz 3]

I wouldn't call what we(robobartz) posted the ultimate solution- its just what worked for us.

[CarloK 203]

The flow sensor and feeder are sensitive tools. As Ultimaker we work hard to make them reliable and we are monitoring this thread to see where we can improve upon.

 

Just a few reminders:

- Check the pressure setting of the filament feeder. On the side of the feeder is a gauge which should sit in the middle of the scale. It might be a production error when this feeder was not tigthened. When the feeder wheel is slipping you can increase pressure here.

- Ultimaker tested with a lot of materials but some materials are problematic since their coating is too slippery or too hard. Flexible materials like TPU are also problematic; TPU 95 is supported, but TPU 85 is not.

- The flow sensor is sensitive for the curvature in the filament since it uses a spring activated sensor wheel to push against the filament. So, moving the filament onto your own spool holders will go wrong when the filament enters the feeder from a different direction.

- A 5kg spool is outside the specifications. If the flow sensor triggered, then this was correct since the feeder is slipping. Fix by using a spool holder with less friction (ball bearings, etc.).

- Never use abrasive materials in your UM3 feeder! From tests we know that printing only 1 spool of abrasive filament destroys grip of the filament wheel forever. The feeder uses a knurled wheel where the teeth wear of fast when using abrasive materials. The S5 feeder uses wheels that received an extra (expensive) treatment to harden them.

[CarloK 203]

@cboath You posted a screen dump of the temperature graph. The most interesting aspect of this graph are the feeder-info lines which are missing from your picture...  😞  

Can you post again?

 

As you can see in the legend a few graphs are not drawn (strikedthrough). Enable these graphs by clicking the text in the legend. For clarity disable the lines that are not interesting, like the bed temperature.

[PKlemm 0]

21 hours ago, CarloK said:

The flow sensor and feeder are sensitive tools. As Ultimaker we work hard to make them reliable and we are monitoring this thread to see where we can improve upon.

 

Just a few reminders:

- Check the pressure setting of the filament feeder. On the side of the feeder is a gauge which should sit in the middle of the scale. It might be a production error when this feeder was not tigthened. When the feeder wheel is slipping you can increase pressure here.

- Ultimaker tested with a lot of materials but some materials are problematic since their coating is too slippery or too hard. Flexible materials like TPU are also problematic; TPU 95 is supported, but TPU 85 is not.

- The flow sensor is sensitive for the curvature in the filament since it uses a spring activated sensor wheel to push against the filament. So, moving the filament onto your own spool holders will go wrong when the filament enters the feeder from a different direction.

- A 5kg spool is outside the specifications. If the flow sensor triggered, then this was correct since the feeder is slipping. Fix by using a spool holder with less friction (ball bearings, etc.).

- Never use abrasive materials in your UM3 feeder! From tests we know that printing only 1 spool of abrasive filament destroys grip of the filament wheel forever. The feeder uses a knurled wheel where the teeth wear of fast when using abrasive materials. The S5 feeder uses wheels that received an extra (expensive) treatment to harden them.

 

Thank you! This helps to understand issues.

[cboath 0]

I'm fairly certain that I found the issue in our case.

 

As noted, I ran the last print through the PrintDry we have. I got a 30+ hour print complete with one unrelated issue. I couldn't get more than 3 or so before. The setup resulted in very little 'webbing' or whatever people call it. I have less than an 1/8" of that running up the side pointing directly at the material tower printed in the back right corner. That seems fairly normal to me.

 

The issue i did have is the design of the print dry. When you set the material on the spool holder, it has line a 1/2" or more gap underneath. As the PVA is pulled into the printer, it unspools a bit. That unspooled are can and will go below your spool and get caught in the assembly under the spindle which will lock up your PVA and case the same error - which is this one single instance was because the material could not spool from being jammed.  I fixed the jam and put the spool flat on the bottom of the dryer and it completed the last 20 hours or so without issue.

 

To be clear, prior to using the printdry is when we got this error of a material being out. And it did occur multiple times inside the 3 hour mark, and it would happen repeatedly. I believe the PrintDry fixed that issue. Whether that is a solution for everyone, i can't say. But it does seem to be the answer in our case. And, FWIW, we have these throughout our building in the general office area and not in a controlled room/area.

[robobartz 3]

In regards to CarloK post:

In regards to CarloK's post:

I found this post very helpful.

In looking at the pressure settings, i found that one indicator was all the way up.

 

Reason:  the head of adjusting screw was on top of the feeder instead of inside. In other words, if you can see the head of the tension adjusting screw on top of the material feeder, then it needs to be pushed back inside body of the feeder. To do so turn the screw into the thread all the way so your not fighting the spring. Then put the cover on and back it out to the setting CarloK recommended in middle.

 

The only thing you should see from the top of the material feeder is a 2-3MM hole through the white plastic. Then you insert the Ultimaker supplied tool into it fits into the screw.

 

The head of the screw is internal pushes up against the inside of the body. While removing the cover from the material feeder, the screw must have popped out through the hole, therefore, it was not supplying any tension to the spring. So if you take the cover off the back of the material holder (with the 4 screws) be careful the screw does not "pop out" of the body and to thru the hole.

 

I couldn't agree more... don't use 5Kg spools. If you do, make sure it spins very freely on your holder. To add to it the Makeshaper roll someone  bought was wound horribly. The only reason we are using one is because its left over from the previous employees.  The spool will sit a long time in the atmosphere before you can use it and risk absorbing moisture. Even printing things 100%  every day it takes a long time to use it up. With Ultimakers nice material sensors we have changed 1k spools many times without affecting the print when they run out.

[Ishy 8]

On 11/11/2018 at 9:41 PM, kmanstudios said:

If the switching arm (the lever that raises or lowers extruder 2) is facing the front of the printer, then it is Material one that is giving the error.

 

If it is facing the back of the printer then it is Material 2 that is the culprit.

 

Are you using the app?

 

And, I bet that the culprit is the extruder with PVA. Why? Because it is the usual suspect.

My S5 has recently started doing this and according to the above it's the CPE not the PVA. Not connected to a network and it's a pretty new roll of filament.

Does Ultimaker have an official fix for this yet? Reading this thread I only see workarounds and maybe's.

[Ishy 8]

Interestingly the problems were with an AA 0.25 head. Yesterday I changed to a 0.4 and it's just finished a 15hr print without a single problem.

[Ishy 8]

Does nobody have any input on this?

[cboath 0]

It seems as if people are getting this error for different reasons. So, unless someone else had your same issue/problem it may be hard to get direct input on the situation.

 

Maybe a warmer head temp? The implication of .4 working vs .25 is that it's not getting through a smaller opening as easily....

 

Just to cover your bases, make sure the materials can feed freely as well and aren't getting caught. I have seen that issue. I've actually had that happen on the material that wasn't giving me problems.

Edited July 22, 2019 by cboath

[Ishy 8]

Interesting thought, I'll try a slightly hotter temperature next time I use the .25. Thank's for the input.

[Giorgos 0]

Hello,

 

I have the same problem with my Ultimaker S5.

I tried all the above but the problem still, remains. 

I used PVA material as support, but although I removed it and I worked with only one nozzle with PLA material, I cannot finish my print.

I do not know, if that is the cause of the problem but it started after I performed a XY offset calibration. 

I use the current firmware 5.4.27.

The problem is serious and make the printer useless

Is there any solution?

 

Thank you

[PKlemm 0]

In my case, the filament diameter was often the cause. It may be that even small grains in the filament tube prevent the filament from being transported further. I was able to determine this by loosening the transport tube on the printhead and the lever on the filament feed, and tried to pull the filament out by hand. If it is difficult to pull, it is the cause.