Issus
Well-Known Member
- Joined
- Aug 27, 2012
- Messages
- 410
- Reaction score
- 0
I've been struggling for inspiration for a new rocket (mostly due to lack of time/funds). When a friend borrowed one of my 29mm CTI cases to finish the retainer on his rocket- I figured I should probably build a rocket for the cases myself - having had the cases for a year or so and yet no rocket that small 
Rather than building constructing a rocket of usual means, I wanted to do something different. So this rocket is going to be completely 3D printed other than the eyebolt in the nose for the shock cord and recovery harness itself.
I want to see what the limits are of printing in plastic, so I'm planning to fly the rocket on a CTI H399WT, simulating to M1.45, and just over 6000ft altitude. It will carry a Telemetrum, but probably only use motor ejection due to space constraints. The length will be exactly 700mm and will weigh 574g with the H399. The weights are known for each part +/- a couple of percent due to the slicer knowing how much plastic it is extruding.
As with all my builds (not that I post many rocket builds), this is first designed in OpenRocket/RASAero, then completely modelled in SolidWorks while updating the OpenRocket model back and forth with SolidWorks based on design changes/issues in SWX.
The biggest challenges on this are, as I see them:
* Printing in PLA rather than ABS, I have the heated bed off my printer at the moment. PLA has a very low Tg and is not as strong as ABS.
* Cant print flat overhangs.
* 205x205x205mm build volume.
* Fairly low hoop strength... not great for supersonic shockwave collapse.
* FDM prints in layers, the bond between layers is only about 30% as strong as an injection moulded part would be.
* Thin walls create bed adhesion problems when printing tall parts. Several test parts came free and printed a giant hairball rather than a tube.
* Low Tg.. motors and ejection charges are hot.
* Fairly poor horizontal tolerances due to slight variations in filament diameter.
* Very long print times, printing 70mm/s walls, 85mm/s interiors these parts are looking to take 5-8hrs each.
* Poor surface finish, as the layers are all slightly different widths giving a very finely ribbed finish.
Advantages, as I see them:
* Pretty much every part of the rocket can be designed in, meaning that only sections need to be glued together.
* Complex features can be designed in, features I can't make on my lathe or CNC mill.
* Fairly cheap.
* Airfoils can be printed into the fins, as can the fillets... leaving very little finishing work.
* Ribbed surface will bond very well on the tube sections.
* Sands fairly readily, with a bit of wet sanding a polished finish is quite simple to obtain.
So, to combat these I've designed a rocket with a 4.3mm wall, giving me a 38.1mm OD, this will be sliced as two walls, both 0.8mm with about 20% density infill . This keeps the weight down, but also adds a lot of strength over a thin feature - making a torsion box of sorts. The fins are 5mm wide, with an "airfoil" designed into them (couldn't be bothered putting in a proper airfoil, so 0.8mm LE/TE and a 33% MAC curvature.) There are lots of 45 degree angles going up the build direction to eliminate overhangs.
This is the rocket I've come up with. The rail buttons are designed into the rocket, as is the recovery point for the lower tube. Even though I'm trying to go for some performance here, I've gone with 4 fins in order to reduce their semi-span. The layers like to curl up on the long thin fingers of the fin as they are printed, due to contraction which can cause issues if it gets too long. There are ways to combat this, but its easier to just design around it.
The tailcone is the motor retainer, which fits on with a twist lock.. if I have my tolerances right. Hopefully the motor case won't melt into this. The fast burn motors I plan to run this on should reduce the amount of heating it seems.
The Avionics bay is one piece, no sled, just a printed hatch cover which is held in by 3x M3 screws, which will hopefully handle the forces on it. The telemetrum antenna sits forward into the nosecone, with a hole in the top bulkhead of the AV bay for it. It took quite a while to design the supports for the top bulkhead into the avionics bay, to stop the "inside" from sagging on my test prints.
Under the telemetrum a small flat is positioned between the mounting points, with a stop at the bottom for the battery to sit on. I'll just be vecroing a 160mAh 1S Lipo into here, there shouldn't be enough room for anything bad to happen if it comes lose, not that it has enough weight to overcome the velcro on deceleration. I don't plan on doing any electronic ejection, even if it would let me get a lot more height as the motor ejection is much too short, so the small battery should be sufficient for tracking the rocket down.
The AV Bay also has a token vent hole printed into it, however this is likely to be useless as the seal on the hatch will be non existent and allow much more interaction.
Rather than building constructing a rocket of usual means, I wanted to do something different. So this rocket is going to be completely 3D printed other than the eyebolt in the nose for the shock cord and recovery harness itself.
I want to see what the limits are of printing in plastic, so I'm planning to fly the rocket on a CTI H399WT, simulating to M1.45, and just over 6000ft altitude. It will carry a Telemetrum, but probably only use motor ejection due to space constraints. The length will be exactly 700mm and will weigh 574g with the H399. The weights are known for each part +/- a couple of percent due to the slicer knowing how much plastic it is extruding.
As with all my builds (not that I post many rocket builds), this is first designed in OpenRocket/RASAero, then completely modelled in SolidWorks while updating the OpenRocket model back and forth with SolidWorks based on design changes/issues in SWX.
The biggest challenges on this are, as I see them:
* Printing in PLA rather than ABS, I have the heated bed off my printer at the moment. PLA has a very low Tg and is not as strong as ABS.
* Cant print flat overhangs.
* 205x205x205mm build volume.
* Fairly low hoop strength... not great for supersonic shockwave collapse.
* FDM prints in layers, the bond between layers is only about 30% as strong as an injection moulded part would be.
* Thin walls create bed adhesion problems when printing tall parts. Several test parts came free and printed a giant hairball rather than a tube.
* Low Tg.. motors and ejection charges are hot.
* Fairly poor horizontal tolerances due to slight variations in filament diameter.
* Very long print times, printing 70mm/s walls, 85mm/s interiors these parts are looking to take 5-8hrs each.
* Poor surface finish, as the layers are all slightly different widths giving a very finely ribbed finish.
Advantages, as I see them:
* Pretty much every part of the rocket can be designed in, meaning that only sections need to be glued together.
* Complex features can be designed in, features I can't make on my lathe or CNC mill.
* Fairly cheap.
* Airfoils can be printed into the fins, as can the fillets... leaving very little finishing work.
* Ribbed surface will bond very well on the tube sections.
* Sands fairly readily, with a bit of wet sanding a polished finish is quite simple to obtain.
So, to combat these I've designed a rocket with a 4.3mm wall, giving me a 38.1mm OD, this will be sliced as two walls, both 0.8mm with about 20% density infill . This keeps the weight down, but also adds a lot of strength over a thin feature - making a torsion box of sorts. The fins are 5mm wide, with an "airfoil" designed into them (couldn't be bothered putting in a proper airfoil, so 0.8mm LE/TE and a 33% MAC curvature.) There are lots of 45 degree angles going up the build direction to eliminate overhangs.
This is the rocket I've come up with. The rail buttons are designed into the rocket, as is the recovery point for the lower tube. Even though I'm trying to go for some performance here, I've gone with 4 fins in order to reduce their semi-span. The layers like to curl up on the long thin fingers of the fin as they are printed, due to contraction which can cause issues if it gets too long. There are ways to combat this, but its easier to just design around it.
The tailcone is the motor retainer, which fits on with a twist lock.. if I have my tolerances right. Hopefully the motor case won't melt into this. The fast burn motors I plan to run this on should reduce the amount of heating it seems.
The Avionics bay is one piece, no sled, just a printed hatch cover which is held in by 3x M3 screws, which will hopefully handle the forces on it. The telemetrum antenna sits forward into the nosecone, with a hole in the top bulkhead of the AV bay for it. It took quite a while to design the supports for the top bulkhead into the avionics bay, to stop the "inside" from sagging on my test prints.
Under the telemetrum a small flat is positioned between the mounting points, with a stop at the bottom for the battery to sit on. I'll just be vecroing a 160mAh 1S Lipo into here, there shouldn't be enough room for anything bad to happen if it comes lose, not that it has enough weight to overcome the velcro on deceleration. I don't plan on doing any electronic ejection, even if it would let me get a lot more height as the motor ejection is much too short, so the small battery should be sufficient for tracking the rocket down.
The AV Bay also has a token vent hole printed into it, however this is likely to be useless as the seal on the hatch will be non existent and allow much more interaction.
