Tight Tolerance Filament Via Use Of A Laser Micrometer With Extruder

Has anyone messed around with a laser micrometer to help improve the diameter accuracy of filament they have produced? I'm looking to use a dual axis Keyence laser micrometer to control the speed of a filament puller via an Arduino to try and get a combined diameter tolerance of 0.01mm or better. Basically I'm just looking to see if anyone has messed around with this idea so I have a starting point, otherwise I will go from the start and see what happens.

The aim of this project is to eventually be able to extrude more exotic materials to a high accuracy level with the possibility of selling the finished filament along with a complete read out of the diameter of each unique roll of filament along with the min & max diameters.

What does a laser micrometer system cost?
What is the interface? usb? serial?
Is this for printing or making filament?

Measuring what is exiting an extruder hotend during printing
would really be great.
Not much room there to get measurement?

The system I have on the way will be around $3k by the time I get it all to Australia, pay customs fees etc. This is for a second hand system so if it was new it would be $10k+.

It uses a RS-232C connection to transfer the information.

It is for making filament. My aim is to get the filament as dimensionally exact as possible to see how it effects print quality and to be able to offer this higher quality and more specialised filament for around the same price as the Chinese filament. I'm sick of paying $80+ for a roll of Polycarbonate filament when I can buy the resin for $6/kg. The filament will also be run through the laser micrometer again when it is spooled so that each individual roll will have it's own listed minimum & maximum diameters and possibly even a full readout of the filament's dimensions. This will help check the quality and help people tune in their printers to the individual filament for the best print results.

This is basically going to be my first attempt at messing around with an Arduino though so that is why I am looking to see if anyone has use it before for this purpose or if there are any better suggestions.

I realize this may be very late, but just registered and found this. Be sure to read the details on the specifications for the laser micrometer you are considering. Often accuracy and repeatability are specified as +/-2 sigma or +/- 3 sigma. In other words, you don't know how bad the "flyers" might be. If your total diameter tolerance is .01mm (approximately +/-.0002"), you are pushing the limits of most laser scan micrometers out there. Which model Keyence are you considering (or which did you buy)? What is the diameter of the filament?

If you are way past this point, how did things work out?

I'm revisiting this project now that I have more time. The system I am looking to incorporate is the Keyence LS7000 series. The current plan is to set the system up to use the two sensors to get a X & Y cross section of the filament & use the reading from the system to increase or decrease the speed of the puller to affect the sizing of the filament. So any thoughts on this process? Any thoughts or details that people have on how to interface with the laser system or the best hardware to use to control the puller speed etc are all appreciated.

Cheers

... if you get good values for the X/Y crosssection, then you can calculate the average diameter, subtract the result from the ideal value and use it as multiplicator for the filament feeder (extruder motor).

As this feeder setting is only given per line segment, what could be a complete diagonal line through the object, you should do this 'online' with a separate controller for the extruder motor ... this needs a firmware-update or a special connection between the controller board and the extruder driver+motor ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Might be interesting for you to read the lyman thingiverse comment section about puller control and width sensor. Seems that for the moment the most successful setup is a double die with two sensors where the first sensor sits between the two dies and the second sensor more down the line where the filament is already cooled down enough. This setup differs from the normal lyman setup as the sensors control the extruder motor whereas the normal setup controls the puller motor.

That said I think that x and y reading is not really needed as the die will make sure the filament is round. Only incorrect cooling or puller rolling on the too warm a filament will create flat sides?

Edited 1 time(s). Last edit at 06/23/2015 09:06AM by Frans@France.