Clear all

Recycle PLA issue

New Member

Good morning,

I try to recycle PLA from Smartfil since 2 months but i encountered some issue.

Like you can see on the picture form attach file, they look like some material is not proprely melted.
I start to work with original parameters and increasing the temperature until 250°C and nothing change, the end process filament is in the same shape. I also try to use sieve to try to use only particules <4mm, and between 2 and 4mm in another batch, nothing change.

This bad shape is a real problem because they disturb the diameter regulation loop. The diameter oscillates between 1.50 to 2.0 mm, it is not acceptable to try to spool this : they are 100% chance to clog the 3D printer.

I don't know what parameter i should change to improve my quality and try to spool something. If someone have any suggestion, i take it !

Thanks you,


Topic starter Posted : 13/07/2021 7:32 am
Member Admin

Dear @simon_lanis ,

Thank you very much for this nice illustrated explanation. I imagine this has been a frustrating process for 2 months, and I am sorry about this. I will do my best to assist you.

First, I must say that each grade of PLA is unique. I don't believe I have processed PLA specifically from Smartfil - my knowledge is based on the processing of many other PLAs, and polymers in general. 

The main symptom :

It seems that the filament contains bumps. Those bump seem to correspond to unmelted sections (particles that didn't fully melt in time, and therefore come out as large bits of solid).

Potential causes/solutions :  

- non-PLA particles in the batch --> do you think this is possible ? 
- particle size heterogeneity --> unlikely, given your description
- imperfect settings --> increasing the temperatures sometimes is a good plan, but does not always favor the melting. Sometimes, lowering the temperatures results in more viscosity, more shear stress, more friction, and, in the end, better melting. This is especially true in the first half of the barrel (H4/H3) --> it would be worth starting around 180°C again, and going down. I believe this will be our focus.
- inconsistent feeding --> I do not think this would take such a form, but definitely is an important factor in general. 

Conclusion : 

Could you please send me a datalog of your experiment ?

Analyzing data such as the motor current would perhaps give me more answers about the melting behavior.

I would like to share this amazing video with you, where Stefan beautifully explains the steps he went through, in a similar situation :

Have you tested other grades/brands of PLA ?

Do you have any questions ? 
Please let me know if there is anything else I can do for you. My goal is to keep helping until you are satisfied with your results 

Best regards,

Posted : 15/07/2021 5:16 pm
New Member

Dear @louis,

Thanks you for your anwser.

I try to do some work based of what you suggest to me.

I updated the frimware to V1.3.2 and installed devovision and try to do some log about what happend (logs are in attach file).
I started this test with PLA from 2mm to 5mm of medium diameter, with standard PLA 1.75mm configuration.
I was slowly decreasing temperature for H1 and H4 from 180 to 165°C, 5 by 5°C. I can observe better result : the flament was a little bit smoother on surface but they are they was alwayse unmelt particle like the picture in my previous post.

What do you think about it ?

Also, I dont think it can be cause by bad feeding because I used the vibrator system (the same as in the video of CNC Kitchen).

I will try to use another batch of PLA to see if the first one is contaminated by something. We work in a Fablab, they are a lot of dust in air.




Topic starter Posted : 19/07/2021 2:00 pm
Member Admin

Hello @simon_lanis ,

Thank you for keeping me up to date.

I think there was an issue with the attaching of the log files. If you wish to attach multiple log files, the easiest way is to send them my way at 

Settings :

I see that you experimented a bit with the settings ; this is great. 
Decreasing the temperatures does not seem to solve the problem completely. Obsviously, the other way we can improve the melting is by increasing the temperatures. 

It is important to distinguish the 2 halves of the process :

  • what happens inside the barrel --> this determines the pressure, consistency of the flow, and melting. I think this is the issue.
  • what happens in the nozzle and outside the machine (cooling, pulling). This part of the process is not responsible for the melting. The goals are simple :
    • H1 should be high enough that the polymer does not solidify at the tip or inside ; but not so high that the output is super liquid and hard to manipulate
    • the fancooling % should be just high enough that the filament becomes solid before reaching the pulley ; but not so high that it freeezes the nozzle

With this in mind, perhaps there is no need to experiment a lot with H1 : any 'average' value that allows the filament to come out smoothly and be cooled down in time, will do. We should focus on the first half of the process : H4/H3/H2.

It would be interesting to try the following : (H4) 175 195 190 180(?) (H1) , or something similar. The idea is to increase H3/H2 a lot and see if the unmelted particles end up melting. However, increasing the overall temperature too much might annihilate the pressure and consistency. 

Particle size : 

I think this may also be a big problem.

In theory, the max acceptable size is 4mm. Unmelted particles perhas correspond to excessively large particles. 

Do you think you could shred the batch to a homogeneously smaller size ? 

Conclusion :

I know this is a lot of info. I would be glad to look at the datalogs and maybe find out more. I am ready to assist you further !

Best regards,

Posted : 21/07/2021 8:58 am