Hey everyone! Hope all is well. My friend and I are college students working on designing and building a 3D printer of our own and want to figure out what problem in the 3D printing world we want to address. We have a decent amount of experience with 3D printing ourselves, but we figured the best way we could get some ideas is by going to people with a lot of experience!
We essentially are looking for something that is missing in 3D printers (FDM printers specifically) or something that could be improved - it doesn’t really have to be anything major. We would just like to use this project of ours to do something to improve the 3D printing world!
If there were a way to measure the filament really accurately, recycled filament with varying width would be easier to use. Anything to do with making recycled filament more easily or using it more easily would help. A sensor of that sort would help with making recycled filament as well, but I'm pretty sure those systems have the feature. The rigs for making filament are expensive regardless, other than the new open source bottle to filament machine Polyformer (https://www.reiten.design/polyformer) which seems to also make the filament width consistent since the input into the filament maker seems pretty easy to make consistent (a continuous flat strip).
The Polyformer and perhaps other devices making filament from strips don't fully melt the filament so there is a hole in the middle which adds a complication to calculating the extrusion multiplier. The width wouldn't matter as much as the thickness of the strip in that case so maybe just measure thickness and width and get area from that, then convert area to equivalent diameter to get the multiplier. Using this type of recycled filament is more accessible (far cheaper and simpler) but adding a self-correction feature for this type of filament may involve too many steps for your project.
What I mean by the first part is: Make the printer (something with enough processing power such as Klipper) able to change the extrusion multiplier on the fly as the dodgy filament changes width. It would need to know the target width you want to emulate (such as 1.75) and the distance between the sensor and the nozzle.
E3d makes a nice tool changer, unfortunately each head requires its own extruder motor. Plus, each motor needs a controller, expanding the number of drivers required. Which in my opinion is redundancy, as only one head is used at a time.
If the main carriage had a single motor that "plugged" into the selected head for direct drive extrusion, it would be a big improvement over the current system available.
Instead of just the open loop friction feed using a stepper motor, add some kind of feedback loop. Put an encoder actually measuring the filament feed. Not directly attached to the stepper motor but a separate roller on the filament driving an encoder. If the filament slips or the drive gears are worn, the system will automatically correct the stepper motor to give the exact amount of filament. It needs to be true closed loop servo control and not just a monitoring of the feed. This would go a long ways towards minimizing or even eliminating the need to calibrate the filament feed. It could detect clogged nozzles. If the measured amount of filament was too far out of tolerance for what was being commanded then it could immediately shut the system down before a bad print even occurred.
An 'Arming' system for the extruder heater. Some sort of series switching arangment that can be automatically disconnected in the event of thermal control failure. The common motherboards have only one failure point to cause a fire, leaving it up to the consumer to provide a backup for fire suppresion. I didn't arange enough protection so my family is looking forward to moving back into our house after repairs from a structure fire in March of this year. The printer was in there and nobody was close enough to hear the SKR board alarming as the heater went up in flames. They're reconnecting the power and gas tomorrow.
Everyone got out ok but it has been an overly traumatic experience for all involved. My new printers have automotive relay power control to the motherboard with oversight by RPis. The second any alarm triggers the printer loses all power.
Hey everyone! Hope all is well. My friend and I are college students working on designing and building a 3D printer of our own and want to figure out what problem in the 3D printing world we want to address. We have a decent amount of experience with 3D printing ourselves, but we figured the best way we could get some ideas is by going to people with a lot of experience!
We essentially are looking for something that is missing in 3D printers (FDM printers specifically) or something that could be improved - it doesn’t really have to be anything major. We would just like to use this project of ours to do something to improve the 3D printing world!
Thank you for your time!
If there were a way to measure the filament really accurately, recycled filament with varying width would be easier to use. Anything to do with making recycled filament more easily or using it more easily would help. A sensor of that sort would help with making recycled filament as well, but I'm pretty sure those systems have the feature. The rigs for making filament are expensive regardless, other than the new open source bottle to filament machine Polyformer (https://www.reiten.design/polyformer) which seems to also make the filament width consistent since the input into the filament maker seems pretty easy to make consistent (a continuous flat strip).
The Polyformer and perhaps other devices making filament from strips don't fully melt the filament so there is a hole in the middle which adds a complication to calculating the extrusion multiplier. The width wouldn't matter as much as the thickness of the strip in that case so maybe just measure thickness and width and get area from that, then convert area to equivalent diameter to get the multiplier. Using this type of recycled filament is more accessible (far cheaper and simpler) but adding a self-correction feature for this type of filament may involve too many steps for your project.
What I mean by the first part is: Make the printer (something with enough processing power such as Klipper) able to change the extrusion multiplier on the fly as the dodgy filament changes width. It would need to know the target width you want to emulate (such as 1.75) and the distance between the sensor and the nozzle.
E3d makes a nice tool changer, unfortunately each head requires its own extruder motor. Plus, each motor needs a controller, expanding the number of drivers required. Which in my opinion is redundancy, as only one head is used at a time.
If the main carriage had a single motor that "plugged" into the selected head for direct drive extrusion, it would be a big improvement over the current system available.
https://e3d-online.com/pages/toolchanger
Closed loop control of the filament feed.
Instead of just the open loop friction feed using a stepper motor, add some kind of feedback loop. Put an encoder actually measuring the filament feed. Not directly attached to the stepper motor but a separate roller on the filament driving an encoder. If the filament slips or the drive gears are worn, the system will automatically correct the stepper motor to give the exact amount of filament. It needs to be true closed loop servo control and not just a monitoring of the feed. This would go a long ways towards minimizing or even eliminating the need to calibrate the filament feed. It could detect clogged nozzles. If the measured amount of filament was too far out of tolerance for what was being commanded then it could immediately shut the system down before a bad print even occurred.
An 'Arming' system for the extruder heater. Some sort of series switching arangment that can be automatically disconnected in the event of thermal control failure. The common motherboards have only one failure point to cause a fire, leaving it up to the consumer to provide a backup for fire suppresion. I didn't arange enough protection so my family is looking forward to moving back into our house after repairs from a structure fire in March of this year. The printer was in there and nobody was close enough to hear the SKR board alarming as the heater went up in flames. They're reconnecting the power and gas tomorrow.
Everyone got out ok but it has been an overly traumatic experience for all involved. My new printers have automotive relay power control to the motherboard with oversight by RPis. The second any alarm triggers the printer loses all power.