Adding internal heat conduction disk to barrel
Posted by damonb
|
Adding internal heat conduction disk to barrel June 12, 2010 09:17AM |
Registered: 16 years ago Posts: 41 |
While still struggling to get my heater barrel work reliably, I was thinking about the thermal conductivity of ABS.
Doing a simulation in ansys with a 450F heat source surrounding a piece of 1/8 inch ABS made me realize that it can take up to 10 seconds for the center of the ABS to be become molten. This could also be something that could limit the ultimate extrusion speed of a reprap, because the cooler and possibly non molten center could form a higher viscosity plug right at the output of the extruder barrel.
One thought that occurred to me was to add a small disk to the to inside of the barrel about a 1/6 to 1/8 of an inch long with holes or slots cut around the edges. What this would do is improve the thermal conductivity to the center to heat the middle of the abs filament. Also if the slots are at the edge , then only molten abs would be able to pass through to the output of the nozzle.This is because we already know the outside of barrel is hot.
Of course a possible drawback may be that the reduced area through the barrel, could increase the force needed to push the filament through. However this may be balanced out by having only molten abs getting past and better heat transfer to the center. So it is hard to say if something like this would help or hurt.
I plan to give this a try, but I wanted to find out of anyone else had looked into this.
Doing a simulation in ansys with a 450F heat source surrounding a piece of 1/8 inch ABS made me realize that it can take up to 10 seconds for the center of the ABS to be become molten. This could also be something that could limit the ultimate extrusion speed of a reprap, because the cooler and possibly non molten center could form a higher viscosity plug right at the output of the extruder barrel.
One thought that occurred to me was to add a small disk to the to inside of the barrel about a 1/6 to 1/8 of an inch long with holes or slots cut around the edges. What this would do is improve the thermal conductivity to the center to heat the middle of the abs filament. Also if the slots are at the edge , then only molten abs would be able to pass through to the output of the nozzle.This is because we already know the outside of barrel is hot.
Of course a possible drawback may be that the reduced area through the barrel, could increase the force needed to push the filament through. However this may be balanced out by having only molten abs getting past and better heat transfer to the center. So it is hard to say if something like this would help or hurt.
I plan to give this a try, but I wanted to find out of anyone else had looked into this.
|
Re: Adding internal heat conduction disk to barrel June 12, 2010 10:12AM |
Registered: 14 years ago Posts: 7,616 |
Some interesting thoughts, damonb.
I can't see however, what's wrong with an partially unmolten center. The barrel is some sort of heating chamber an if you can't melt the ABS quickly enough my first guess would be to make this heating chamber even longer. The filament has more time to become fluid, then.
If you put some gate at the beginning of this chamber, there's no reason to have this chamber at all.
I can't see however, what's wrong with an partially unmolten center. The barrel is some sort of heating chamber an if you can't melt the ABS quickly enough my first guess would be to make this heating chamber even longer. The filament has more time to become fluid, then.
If you put some gate at the beginning of this chamber, there's no reason to have this chamber at all.
| Generation 7 Electronics | Teacup Firmware | RepRap DIY |
|
Re: Adding internal heat conduction disk to barrel June 12, 2010 10:54AM |
Admin Registered: 17 years ago Posts: 7,881 |
Suppose you are extruding 0.5 filament at 40mm/s. The filament feed is 36 times slower, so in 10 seconds it advances 11mm, so as long as your hot section is longer than that it will have time to melt. If it doesn't then the front face of the filament meets the cone of the nozzle so it will then melt very quickly anyway.
I have a very short hot zone on HydraRaptor and I generally find it a benefit in that it starts and stops more sharply. I expect that the force required to extrude will rise sharply when the extrusion speed is too fast for the filament to melt, but that speed is quite high for the majority of heater designs.
[www.hydraraptor.blogspot.com]
I have a very short hot zone on HydraRaptor and I generally find it a benefit in that it starts and stops more sharply. I expect that the force required to extrude will rise sharply when the extrusion speed is too fast for the filament to melt, but that speed is quite high for the majority of heater designs.
[www.hydraraptor.blogspot.com]
|
Re: Adding internal heat conduction disk to barrel June 12, 2010 06:20PM |
Registered: 16 years ago Posts: 41 |
Did a quick test with my current test barrel. I cut 2 1/16 inch slots about 1/32 inch deep into a piece of 1/8 inch brass.
The length was roughly 3/16 of an inch. I probably should have made it 1/8 of an inch. Unfortunately the the 1/8 inner diameter barrel is a few thousands larger than the rod, so the fit was not a press fit. So there is not complete thermal contact between the rod and barrel. Still a few thousands of an inch should present a negligible thermal resistance.
I was able to get extrusion, but I still have the intermittent extrusion behaviour. I think there is some potential with this option, but I will have to do some experimentation.
Nophead,
My overall barrel length is 3/4 of an inch. The heated area is wound with about about 10 turns of 32 gauge nichrome over 1/4 of an inch and as close to the tip of the barrel as I can get. The insulation is mica paper. When I pull out the filament after things stop, I observe about a 1/8 inch plug. The nozzle diameter is .3 mm. I could bump that up to .5mm, but I doubt I will see much improvement.
The barrel is 1/8 inch ID stainless with .01 wall thickness. By my calculations, this provides a negligible thermal resistance radially and vastly improved thermal resistance along the axis compared to brass.
I will get a set of fish scales and do a pull test similar to what you did and verify that my drive is developing enough force to push the material through before slipping. I think it is, but I will get some hard numbers to verify that. I certainly can not cause a slip by hand.
The length was roughly 3/16 of an inch. I probably should have made it 1/8 of an inch. Unfortunately the the 1/8 inner diameter barrel is a few thousands larger than the rod, so the fit was not a press fit. So there is not complete thermal contact between the rod and barrel. Still a few thousands of an inch should present a negligible thermal resistance.
I was able to get extrusion, but I still have the intermittent extrusion behaviour. I think there is some potential with this option, but I will have to do some experimentation.
Nophead,
My overall barrel length is 3/4 of an inch. The heated area is wound with about about 10 turns of 32 gauge nichrome over 1/4 of an inch and as close to the tip of the barrel as I can get. The insulation is mica paper. When I pull out the filament after things stop, I observe about a 1/8 inch plug. The nozzle diameter is .3 mm. I could bump that up to .5mm, but I doubt I will see much improvement.
The barrel is 1/8 inch ID stainless with .01 wall thickness. By my calculations, this provides a negligible thermal resistance radially and vastly improved thermal resistance along the axis compared to brass.
I will get a set of fish scales and do a pull test similar to what you did and verify that my drive is developing enough force to push the material through before slipping. I think it is, but I will get some hard numbers to verify that. I certainly can not cause a slip by hand.
Sorry, only registered users may post in this forum.