Hybrids 2024

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Happy to hear this. Hopefully they'll approve the regular canisters.

Agreed, that would be absolutely awesome. There’s a few micro-hybrid flyers here and I don’t think we can get aluminium bulbs. But I've done well over 50 flights now and they’re quite safe. I’ve done all sorts from drilling out to a way oversized injector to putting in completely untested grains with way more pre-heater than needed and found that it’s very hard to break. you can’t cause a runaway overpressure unless you’re using oxidised grains (turbo APCP or bi-hybrid) because the oxidiser flow reduces when chamber pressure increases.

The biggest (only) “failure” I have had is with a printed grain that wasn’t thick enough at the top - the grain regressed enough to expose the o-ring and burned through and allowed hot gasses to pass the injector housing.

On a mono-tube hybrid that would normally be very bad. But since the oxidiser is inside the bulb, those gasses don’t reach the rest of the oxidiser and there is no damage at all. I noted that the nitrous bulb only had soot on the outside up to where the tape was too, which tells me you could probably add an extra safeguard against blow-by from just adding a second o-ring groove halfway up the injector housing (or by making sure there’s enough meat on your printed grain haha)
 
Last edited:
Scott has been helping me along with my experimental hybrid as well. 4' 38mm Skyripper 'clone'. Case has been machined, hopeful to test it out next weekend at our first launch!
Just received my 38mm Skyripper nozzles from China, fit perfectly! I had modeled them up a while ago, finally sent them out for machining. They really nailed the dimensions.

1713321816003.png1713322169253.png1713322404146.png1713322425572.png
 
MZ-G3 is the grade, 70$ for 5
Xinghe County Muzi Carbon

Took about 10 weeks from the order. You have to give them a FULLY dimensioned drawing, they wanted a drawing with dimensions, they are a machine shop not a nozzle company. I have another nozzle i am about to have made by another shop.

1713329588981.png
 
Last edited:
Curious as to what the density of the 3D printed ABS is relative to stock, cast or extruded, ABS.
I havent checked ABS but i do do the calculation for most parts I do, just as a quick reality check.

Part of the calibration do for my machines is calculate the finished part volume (easy with cad) and divide by the actual weight. More of a check on porosity.

Quick example, Nose and Base (thick nose for some nose weight)

1713379861345.png1713381082060.png

208076.46 (+/- 0.073) cubic millimeters for 2 solids or 208.08 CC PET-G has a specific gravity of 1.27 so;

Theoretical 1.27 G/cc X 208.08 CC = 264.3
Measured at 249 grams (dry)
Porosity 6% of volume (theoretical estimate)

This can be lowered by increasing the calibration for the filament feed. If you try and get too close to zero porosity the quality of the print goes down, not sure where the sweet spot is, but I have been down to 3% and still have a good part. Think about it, the material coming out of the nozzle is MUCH hotter than the finished part, and shrinks as it reaches part temperate. Passing through the glass transition temperature adds to the shrinkage.

Some other things to think about is what is the real tolerance of the diameter of the filament, and are you temperature compensated.

Mike (one class short of having a degree in chemistry too) K
 
I havent checked ABS but i do do the calculation for most parts I do, just as a quick reality check.

Part of the calibration do for my machines is calculate the finished part volume (easy with cad) and divide by the actual weight. More of a check on porosity.

Quick example, Nose and Base (thick nose for some nose weight)

View attachment 641111View attachment 641113

208076.46 (+/- 0.073) cubic millimeters for 2 solids or 208.08 CC PET-G has a specific gravity of 1.27 so;

Theoretical 1.27 G/cc X 208.08 CC = 264.3
Measured at 249 grams (dry)
Porosity 6% of volume (theoretical estimate)

This can be lowered by increasing the calibration for the filament feed. If you try and get too close to zero porosity the quality of the print goes down, not sure where the sweet spot is, but I have been down to 3% and still have a good part. Think about it, the material coming out of the nozzle is MUCH hotter than the finished part, and shrinks as it reaches part temperate. Passing through the glass transition temperature adds to the shrinkage.

Some other things to think about is what is the real tolerance of the diameter of the filament, and are you temperature compensated.

Mike (one class short of having a degree in chemistry too) K

Thanks a lot, Mike. I appreciate the detail and the thoroughness of your response! The calculations - Theoretical vs. Measured - in particular. The implied Porosity seems logical!!

My particular interest is in POM (Polyoxymethylene) as a fuel. Rho (Specific Gravity) = 1.41. POM is, commercially, Delrin and Acetal.

Any knowledge of or experience with 3D Printing in Delrin or Acetal?

Steve (about 3 classes short of a BS in chemistry, too!!)
 
Used lots of Delrin and Acetal in the day job, always machined it, a dream to lathe turn or mill. Never printed it, any idea what the Isp is?

Um dumb question, why print it? easy to machine, even easy to cut with a broach or offset drill with a guide if you wanted internal grooves? I looked at printing PP grains until I realized it is much cheaper to just machine it.

My default is Polypropylene, yeah low density but the Isp is around 190 seconds, easy to machine and cheap. About $6/foot for Schedule 80 1" fits well in a 38" motor, 1-1/2" works for 54mm size
Nylon 6/6 is a little higher density, and has an Isp of 205, pretty easy to machine not as cheap.

Mike K
 
Why print?
Finocyl, spiral pathways. Infill % then fill with wax/ isopropyl/ take your pick.
A printer can print everything overnight.
If you print in cheap ABS, the cost is cheaper than machining bar. No wastage.
 
Last edited:
Used lots of Delrin and Acetal in the day job, always machined it, a dream to lathe turn or mill. Never printed it, any idea what the Isp is?

Um dumb question, why print it? easy to machine, even easy to cut with a broach or offset drill with a guide if you wanted internal grooves? I looked at printing PP grains until I realized it is much cheaper to just machine it.

My default is Polypropylene, yeah low density but the Isp is around 190 seconds, easy to machine and cheap. About $6/foot for Schedule 80 1" fits well in a 38" motor, 1-1/2" works for 54mm size
Nylon 6/6 is a little higher density, and has an Isp of 205, pretty easy to machine not as cheap.

Mike K
I, too, have worked with PP, HDPE, and Delrin; and you're right - simply machine these.

Raw material is relatively inexpensive.

I work with GOX, as opposed to NOX, so I don't know what the Isp (with NOX) is likely to be; I have done some modeling, with various free and purchased software.

Most of the work thus far has been actual empirical experiments and tests: PP, PE, Nylon 6, Nylon 6/6, PMMA, HDPE, PVC, ABS, PS, a few others and, my favorite, POM (Delrin/Acetal). Burns residue- and char-free in 100% o2 (Gaseous;GOX) at about 1300 C. Flame is an almost invisible very light blue. Zero smoke. If I recall, the Mp is about 165 C, which is not terrible, but the density - the density - is 1.41-1.42.
 
Why print?
Finocyl, spiral pathways. Infill % then fill with wax/ isopropyl/ take your pick.
A printer can print everything overnight.
If you print in cheap ABS, the cost is cheaper than machining bar. No wastage.
This is a good point!!!

I was thinking of some sexy CNC work on Delrin Right Circular Cylinders, but anything sexy would, I think, be limited to the outer surface. E.g., spiral pathways on/down the outer circumference to get some helical O2 flow/mixing just upstream of the convergent section of the nozzle....

Would you happen to know if you can 3D print in PP, PE, HDPE, or Delrin/Acetal?

(Thanks for mentioning the idea of filling with wax/IPA/or other!!!!)
 
spiral pathways on/down the outer circumference to get some helical O2 flow/mixing just upstream of the convergent section of the nozzle....

Would you happen to know if you can 3D print in PP, PE, HDPE, or Delrin/Acetal?
you can definitely get POM filament for 3d printing but don’t know which manufacturers use homopolymer or copolymer as their filament, if that matters. I don’t know how hard it is to print but from some quick reading it sounds like bed adhesion and layer separation are the biggest challenges, so a heated bed of 100-150c and an enclosure would be preferable I reckon.

Spiral pathways is exactly what I’ve been experimenting with for printing ABS grains to use with my micro-hybrid, and it makes a massive difference in thrust. I don’t have any way of confirming but by visual inspection, it does appear that there is some helical mixing going on, as the regression is greater, and very even throughout the length of the grain compared to bates style ports that seem to regress more in the middle of the grain but very little at each end.

My theory is (and I hope I word this right) that creating riffling or helical channels might also be giving the oxidiser some angular momentum, moving the boundary layer closer to the surface of the grain.

I completely overlooked Delrin as a printable fuel, and I haven’t read this yet, but here’s a paper I found talking about it as a fuel for hybrids in my quick search!
https://etda.libraries.psu.edu/catalog/16189git5063/
 
I completely overlooked Delrin as a printable fuel, and I haven’t read this yet, but here’s a paper I found talking about it as a fuel for hybrids in my quick search!
Printing Delrin seems to be really not worth it, since it has a ton of drawbacks as a 3D printing material.
 
you can definitely get POM filament for 3d printing but don’t know which manufacturers use homopolymer or copolymer as their filament, if that matters. I don’t know how hard it is to print but from some quick reading it sounds like bed adhesion and layer separation are the biggest challenges, so a heated bed of 100-150c and an enclosure would be preferable I reckon.

Spiral pathways is exactly what I’ve been experimenting with for printing ABS grains to use with my micro-hybrid, and it makes a massive difference in thrust. I don’t have any way of confirming but by visual inspection, it does appear that there is some helical mixing going on, as the regression is greater, and very even throughout the length of the grain compared to bates style ports that seem to regress more in the middle of the grain but very little at each end.

My theory is (and I hope I word this right) that creating riffling or helical channels might also be giving the oxidiser some angular momentum, moving the boundary layer closer to the surface of the grain.

I completely overlooked Delrin as a printable fuel, and I haven’t read this yet, but here’s a paper I found talking about it as a fuel for hybrids in my quick search!
https://etda.libraries.psu.edu/catalog/16189git5063/
You could print your fuel grain, mould in silicone and then cast into that. Acrylic can be cast, but the solvent catalyst isn't nice.
Casting urethane with aluminium powder in it could be interesting. If casting urethane into silicone, use the specific spray on release agent as urethane attacks silicone and this limits the number of uses you get. 20+ should be good. That's more than most of us would need.
You could also print a shell and fill it with urethane.
 
Last edited:
Printing Delrin seems to be really not worth it, since it has a ton of drawbacks as a 3D printing material.

There’s a lot of off-the-shelf machines that can handle most of those problems now. Bed adhesion would still be a big challenge… and I guess the formaldehyde too… if you like breathing…

You could print your fuel grain, mould in silicone and then cast into that.
has crossed my mind but you lose the benefit of “just hit print and come back later”, and I reckon removing the original print from the silicone mould with something like a twisted finocyl port could be difficult. Depends on how complex the shape is and how aggressive the twist is, I guess. Printed shells for casting would work if they’re printed that you can cast in and be dissolved though.

Complex geometry for helical mixing AND additives would definitely be interesting. Additives currently can’t be done with printed grains unless you find a filament that has the additives you want in the quantities you need - for now… pellet extruders are possible, not that much more of a stretch for printing with powder mixes!
 
There’s a lot of off-the-shelf machines that can handle most of those problems now. Bed adhesion would still be a big challenge… and I guess the formaldehyde too… if you like breathing…


has crossed my mind but you lose the benefit of “just hit print and come back later”, and I reckon removing the original print from the silicone mould with something like a twisted finocyl port could be difficult. Depends on how complex the shape is and how aggressive the twist is, I guess. Printed shells for casting would work if they’re printed that you can cast in and be dissolved though.

Complex geometry for helical mixing AND additives would definitely be interesting. Additives currently can’t be done with printed grains unless you find a filament that has the additives you want in the quantities you need - for now… pellet extruders are possible, not that much more of a stretch for printing with powder mixes!
Twisted finocyl is easy to get out. I've pulled an apes skull out. No issues. ..... :)
 
Delrin is a difficult material to print. The point at which it melts and can be deposited is very close to the point where it starts to degrade, so you need very precise temperature control (i.e. an enclosure), and even if you have that it prints at a relatively high temperature. UTK has used Delrin in the past, it's kind of nice in that when you burn Delrin it continues to smolder and release formaldehyde, so you can actually do re-starts with it. Down side is that the performance is garbage and the regression rates are low enough that UTK used Delrin as a liner in their 75mm and 152mm liquid motors. Video of them testing re-starts with a Delrin grain in a hybrid, regression rate was ~0.5 mm/s or so and specific impulse like 120 seconds:



In general, ABS is going to offer the best performance of any of the 3D printable materials. And having done hollow channels filled with wax / powdered coal / etc. they "regress faster" but most of the time the fill material just gets blown out the nozzle without really doing anything.
 
As a bonus, here is a firing with a 3D printed matrix (concentric rows of rings that were later packed with powdered anthracite coal. It had some of the highest performance seen in any of the tests we ran (196 seconds), but given you can see the black cloud of coal at startup I'd venture to guess all of it blew out of the channels at ignition and the performance would have actually been higher without it:



Best I've managed to come up with so far is a plastisol-aluminum based fuel, which has delivered ISP on the order of 211 seconds in a 5 inch motor and in-theory could still go higher than that. It's extremely dense, the optimum OF ratio is like 4 or so so for the same amount of nitrous you can consume more fuel and so there is more total propellant mass and more impulse for the same motor. Add to that the fact plastisol regresses at speeds similar to paraffin and the UT guys flew it on their 5 inch Spaceport rocket to ~29,000 feet back in 2022.

 
you can definitely get POM filament for 3d printing but don’t know which manufacturers use homopolymer or copolymer as their filament, if that matters. I don’t know how hard it is to print but from some quick reading it sounds like bed adhesion and layer separation are the biggest challenges, so a heated bed of 100-150c and an enclosure would be preferable I reckon.

Spiral pathways is exactly what I’ve been experimenting with for printing ABS grains to use with my micro-hybrid, and it makes a massive difference in thrust. I don’t have any way of confirming but by visual inspection, it does appear that there is some helical mixing going on, as the regression is greater, and very even throughout the length of the grain compared to bates style ports that seem to regress more in the middle of the grain but very little at each end.

My theory is (and I hope I word this right) that creating riffling or helical channels might also be giving the oxidiser some angular momentum, moving the boundary layer closer to the surface of the grain.

I completely overlooked Delrin as a printable fuel, and I haven’t read this yet, but here’s a paper I found talking about it as a fuel for hybrids in my quick search!
https://etda.libraries.psu.edu/catalog/16189git5063/
Thanks for this! There are some cool (complex; regression-rate enhancing) geometries that I've read about, and would like to try, that would likely require them to be 3D printed - hence the interest. I imagine that the printing itself would lead to a reduced effective Density - post printing.

You're certainly right to bring up Homo and Co-polymers; some Delrin/Acetal is labeled, some isn't! The PE component of the Delrin Copolymer can be as high as 1 - 1.5%; based on what I recall of my polymer chemistry, I imagine that you could have significantly higher (~10%) PE, in a custom formulation - if, of course, that was advantageous to, say, Isp.

I have read a good deal about spiral / helical pathways - ports. I'm may be mistaken here - Please Do Correct Me!!! - but, in doing this work - these tests - in a purely empirical fashion, if all conditions are identical, test to test, it seems to me that the Regression Rate is the single most important indicator of better - worse. The 'evenness' you've seen, throughout the length of the grain is, I think, a meaningful result!!!

The idea that some angular momentum imparted to the Oxidizer via your helical channels (external channels? On the outer circumference; the periphery? Or internal? Or both, for that matter!?) could move the boundary layer closer to the surface of the grain is interesting! I know that I'm certainly not a good enough Physicist, Thermochemist, or Thermodynamicist to even guess intelligently!!!

It's been a while, and I'm embarrassed because I've forgotten which Software I used to 'study' the Thermochemistry of Delrin - GOX, but I believe that I saw, at ~ 100 PSI Pc, Isps close to 200 sec.

Thanks so much for all of this, and for the paper re Delrin as a fuel for Hybrids!!!!
 
Why print?
Finocyl, spiral pathways. Infill % then fill with wax/ isopropyl/ take your pick.
A printer can print everything overnight.
If you print in cheap ABS, the cost is cheaper than machining bar. No wastage.
My comment was why print delrin, it is a hard to print materials that machines VERY well and is available in many sizes. I do know what you can do with a printer (I have three 3d printers, lots of 3d printed designs of mine on here), my comment is that for Delrin it may not be worth it. The other comment is for many materials, off the shelf schedule 80 is a simple way to make a grain.
 
Last edited:
My comment was why print delrin, it is a hard to print materials that machines VERY well and is available in many sizes. I do know what you can do with a printer (I have three 3d printers, lots of 3d printed designs of mine on here), my comment is that for Delrin it may not be worth it. The other comment is for many materials, off the shelf schedule 80 is a simple way to make a grain.

Thanks for this!! Entirely ture - based on my experience, anyway - that it's really a lovely/easy material to machine; and I mean machine in any available and and number of ways!!

I've recently read of some work in which the oxidizer is injected at the fore end (top; towards the nose), it flows through many small holes (ports) aft (towards the converging section of the nozzle; the 'end' of the grain - if a solid polymer can be called a 'grain'! - and is ignited at, and burns there. Each and every port (there could be dozens) burns individually - from its end upwards, with a very rich oxidizer supply. All ports burn at roughly the same rate (for a little while, at least!), and the reported data shows that this greatly increases the Regression Rate.

Drilling, say, 3 dozen 1 mm holes, parallel to each other, 3" in length, and evenly distributed radially, over, say, a 1.75" Delrin (or any Solid) Right Circular Cylinder is, perhaps, doable, but I thought that perhaps is would be 'a piece of cake' to 3D print that!!
 
I have read a good deal about spiral / helical pathways - ports. I'm may be mistaken here - Please Do Correct Me!!! - but, in doing this work - these tests - in a purely empirical fashion, if all conditions are identical, test to test, it seems to me that the Regression Rate is the single most important indicator of better - worse. The 'evenness' you've seen, throughout the length of the grain is, I think, a meaningful result!!!

The idea that some angular momentum imparted to the Oxidizer via your helical channels (external channels? On the outer circumference; the periphery? Or internal? Or both, for that matter!?) could move the boundary layer closer to the surface of the grain is interesting! I know that I'm certainly not a good enough Physicist, Thermochemist, or Thermodynamicist to even guess intelligently!!!
Regarding this regression rate is only important to an extent, the extent being that you can have a compact and volumetrically efficient grain that achieves the optimum O/F ratio. With nitrous at least you're generally talking about optimum O/F ratios between 7 to 9, so regression rates don't have to be terribly fast (~1.5 mm/s for ABS or HTPB is enough to make great motors). What is arguably more important is the volumetric efficiency of the grain (do you need a wide open port with an extremely high L/D ratio to hit the target O/F ratio?) and O/F shifting throughout the burn (for a constant oxidizer flow rate regression exponents greater than 0.5 tend to go oxidizer rich as the burn progresses, less than 0.5 go fuel rich, an exponent of 0.5 is constant O/F, in blowdown nitrous systems an exponent of ~0.57 will minimize O/F shift).

It's been a while, and I'm embarrassed because I've forgotten which Software I used to 'study' the Thermochemistry of Delrin - GOX, but I believe that I saw, at ~ 100 PSI Pc, Isps close to 200 sec.
Regarding the 200 second ISP you should always keep in mind the software is calculating equilibrium combustion properties, which generally don't happen in reality. A typical hybrid motor may have combustion efficiency on the order of 70% or so, adding a post-combustion chamber may push that over 80 and a mixer plate like is used by TU Delft and Stuttgart can get combustion efficiency over 90%. As a consequence most hybrid rockets on an amateur scale will see specific impulse in the range of 140 to 180 seconds or so, with the best motors only approaching or slightly exceeding 200 seconds (motors with well tuned fuel grain geometries, post combustion chambers or mixers, proper atomization from injectors, optimized nozzles, etc).
 
Delrin is a difficult material to print. The point at which it melts and can be deposited is very close to the point where it starts to degrade, so you need very precise temperature control (i.e. an enclosure), and even if you have that it prints at a relatively high temperature. UTK has used Delrin in the past, it's kind of nice in that when you burn Delrin it continues to smolder and release formaldehyde, so you can actually do re-starts with it. Down side is that the performance is garbage and the regression rates are low enough that UTK used Delrin as a liner in their 75mm and 152mm liquid motors. Video of them testing re-starts with a Delrin grain in a hybrid, regression rate was ~0.5 mm/s or so and specific impulse like 120 seconds:



In general, ABS is going to offer the best performance of any of the 3D printable materials. And having done hollow channels filled with wax / powdered coal / etc. they "regress faster" but most of the time the fill material just gets blown out the nozzle without really doing anything.

Drew,

This is really cool!!! No comments yet, but I thank you for summarizing the details, construction, and performance!

Was the Oxidizer NOX or GO2 (or LOX)???

Agree re 120 sec is, well, not great.

Thanks for the note re ABS. I have read that, as you say, 'most of the fill material just gets blown out..."
 
Used lots of Delrin and Acetal in the day job, always machined it, a dream to lathe turn or mill. Never printed it, any idea what the Isp is?

Um dumb question, why print it? easy to machine, even easy to cut with a broach or offset drill with a guide if you wanted internal grooves? I looked at printing PP grains until I realized it is much cheaper to just machine it.

My default is Polypropylene, yeah low density but the Isp is around 190 seconds, easy to machine and cheap. About $6/foot for Schedule 80 1" fits well in a 38" motor, 1-1/2" works for 54mm size
Nylon 6/6 is a little higher density, and has an Isp of 205, pretty easy to machine not as cheap.

Mike K
Mike,

I just re-read this. Re your default of PP, and your Nylon 6/6, Isps, are these with NOX? (I presume so, but don't want to presume anything!!!)

If so, may I ask the ~ Pchamber?

Thanks, Steve
 
Regarding this regression rate is only important to an extent, the extent being that you can have a compact and volumetrically efficient grain that achieves the optimum O/F ratio. With nitrous at least you're generally talking about optimum O/F ratios between 7 to 9, so regression rates don't have to be terribly fast (~1.5 mm/s for ABS or HTPB is enough to make great motors). What is arguably more important is the volumetric efficiency of the grain (do you need a wide open port with an extremely high L/D ratio to hit the target O/F ratio?) and O/F shifting throughout the burn (for a constant oxidizer flow rate regression exponents greater than 0.5 tend to go oxidizer rich as the burn progresses, less than 0.5 go fuel rich, an exponent of 0.5 is constant O/F, in blowdown nitrous systems an exponent of ~0.57 will minimize O/F shift).


Regarding the 200 second ISP you should always keep in mind the software is calculating equilibrium combustion properties, which generally don't happen in reality. A typical hybrid motor may have combustion efficiency on the order of 70% or so, adding a post-combustion chamber may push that over 80 and a mixer plate like is used by TU Delft and Stuttgart can get combustion efficiency over 90%. As a consequence most hybrid rockets on an amateur scale will see specific impulse in the range of 140 to 180 seconds or so, with the best motors only approaching or slightly exceeding 200 seconds (motors with well tuned fuel grain geometries, post combustion chambers or mixers, proper atomization from injectors, optimized nozzles, etc).

"With nitrous at least you're generally talking about optimum O/F ratios between 7 to 9, so regression rates don't have to be terribly fast (~1.5 mm/s for ABS or HTPB is enough to make great motors)." I really like the way you've expressed this!!!

If I'm correct, with GOX, the O/F Ratio would drop - significantly. Does that imply that, with the lower Ratio, the regression rate needs to be faster? (Or, do I have it ass backwards!?!?)

I do understand the concept of O/F shifting throughout the burn.

Re 'the 200 sec Isp,' Yea. Point Taken!!!!

Drew, I can't thank you enough for sharing and detailing your experience and knowledge re everything that you've written about here!!!! Yes. Of course I wish it were re LOX or GOX, but there are so many points you make, and details that you share, that are valuable and educational - to me, in any case!!!

Thanks Again!!!
 
If I'm correct, with GOX, the O/F Ratio would drop - significantly. Does that imply that, with the lower Ratio, the regression rate needs to be faster? (Or, do I have it ass backwards!?!?)
For the most part this seems to be the case. For something like paraffin you usually have the opposite problem as Paraffin's ideal O/F ratio with nitrous for example is like 9:1, but it regresses extremely quickly. For Oxygen, the regression rates are usually faster than with N2O so it somewhat offsets the lower required O/F ratio.
 
Back
Top