Tests on PA12 powder
I'm glad to hear that the log results work normally now. Could you please share with me the log file? (In Devovision, click 'log view' on the left-side menu, and then download your log.) This would give me more insight, to help give some tips for finetuning.
Also, in this article you will find some tips on how to improve the filament thickness consistency: https://support.3devo.com/filament-thickness-deviation-inconsistent-diameter/
I hope you are doing alright after the holidays.
As you ask in your previous message, I attached the log file to this message.
The quality of the data is not perfect. I will try to do another run with a more reliable data log.
I hope that you will be able to find some info.
I read somewhere that you could run a test if we provide some material.
Do you still offer this service? If so, do you have a price?
If we can't find the optimal settings by ourselves, it could be a great solution to give you some powder to run a test for us.
Apologies for the late response, and thank you, all is well here! I hope you are doing good as well.
Thank you for sending over the datalog, this gives me more insight in the process. I also noticed that at one point you had the puller speed set to 'manual' which disabled the automatic puller function to control the diameter, but perhaps this was to experiment the flow.
From the log it looks a bit like the material is unable to build up good and consistent pressure, this could be due to issues such as bridging and ratholing, but if you placed the feeder as in the photo in the attachment, it is probably something else.
Things that could be happening:
- The hopper is not filled enough, which causes a drop in pressure buildup in the extruder. Especially with powders, the effect from how much material is in the hopper can play a bigger role.
- Heater 4 is set either too high (or too low). If a material melts too early in the extruder, it makes it more difficult to build up (consistent) pressure.
I hope this helps, and I'm looking forward to your next message!
As you suggested, I increased the quantity of powder in the hopper, and I reduced the temperature of the heater H4.
It seems to work.
To find the right parameter for the RPM and the diameter, I followed the following step:
- I increase the RPM progressively to augment the quantity of PA12.
- I didn't put the puller wheel in automatic to be sure I had enough material.
- Once the filament diameter was close to 2.85 mm, I changed the puller wheel in automatic to let the machine stabilise the filament.
The problem is that once the puller wheel is automatic, everything falls apart. If the filament starts to be too large, the puller wheel reduces speed.
The consequence is that the filament begins to move sideways, and then the sensor can't measure the dimension and give the value 0, and the puller wheel stops pulling.
The other problem I'm facing is that the RPM start to deviate from the fixed value.
The fact that the RPM where high (10 at the maximum) may play a role in this, but you may have another explanation.
On another subject, could you tell me how the puller wheel work with the diameter sensor?
- Does the extruder take an average value of the diameter during X second, and then the puller wheel work with it?
- Or does it work with the instantaneous value?
It may have an impact on the extrusion.
Anyway, I will continue to do some tests and hopefully have better results.
Good morning @timo,
I'm writing again to add some thought to the discussion.
After some research, I was able to find an interesting document about the process of extrusion.
Although being heavy to read, I was able to find some info that caught my attention.
- They suggest having a low temperature on the first part of the screw. It helps create pressure inside the barrel.
Should the temperature be under the melting temp (Approx. 195°C for Pa12 powder).
- They speak almost only about pellets, and we use powder.
Does the fact that we use powder can affect the way the plastique melt (see picture)?
I would think that the powder melts faster than the pellet, but I may be wrong.
I hope the message is clear.
I'm looking forward to your message.
Thank you for being so elaborate, it definitely helps troubleshooting, although I had to think for a while and I still might not have the perfect answer for you unfortunately. I'm glad to see however that the output flow improved significantly.
Regarding the melting process of PA12 powder:
Indeed, usually with extrusion the feed section is set below the melting point of the material. The trick is to melt the material gradually in the extruder, with the material completely molten near the exit of the extruder. This way you have the best pressure buildup. Powders do have more surface than pellets, which means there is more heat transfer happening in the feed section. This is why for processing powders the temperature should be even lower in that zone to make sure the material does not melt too quickly.
About the other issue - the puller mechanism:
The optical sensor indeed takes an average, and based on an algorithm it changes the puller speed to make sure the diameter stays as close to the 'set diameter' as possible. If there are several 0 readings from the sensor, the algorithm cannot properly regulate the speed anymore.
What is causing the waving of the filament? Is it molten/soft material collapsing under its own weight, or is it already solidified there? The cooling fans have a major influence here, but materials with a low viscosity tend to suffer more often from this phenomenon. This is because they are either too liquid (or already solidified when decreasing the temperatures) to retain in the extruded shape, which is a straight filament in this case. If the waving is minor, it often helps to put something on the sensor to make sure it stays within its range. An example in the picture below.
I hope this helps!
Hello again @germainl 😉
Thank you for your other participation https://support.3devo.com/community/filament-maker-issues/tests-on-pa12-powder/
I understand it would perhaps be more practical for you to communicate directly with @francoisnicol
I will send a private email to him, from email@example.com, disclosing the email adress attached to your forum account, with also a link to this very conversation.
Please let me know if there is anything else I can do to help