(Yet Another) 100k Attempt

[watheyak]

Good stuff! I really like the transition to fiberglass.

 

[robopup]

Next level **** going on here!

Tony

And almost all of it stolen from guys like you Tony!

@watheyak - thanks, it's worked pretty well so far with directional trackers in 29-54mm land but I haven't run it with GPS or telemetry before. I have a bunch of bench testing on my list of things to do the next month or so...

 

[plugger]

Very cool thread, I'm actually building a staging stack quite similar to this at the moment. One question though.

• 4 fins, > 2 calibers of stability through flight profile per Rasaero, 0.1 in G10 with no tip-to-tip

Please correct me if I'm wrong, but I'm reading that as your sustainer fins are 0.1 inch thick G10. Given your sustainer is simulated to hit ~Mach 3.6ish I'd be concerned about those fins not being thick or stiff enough for that flight profile.
And while in general I agree that t2t isn't necessary for the up bit (assuming your fins are adequate thickness with fillets to match) I'm applying it to everything aggressive moving forward as most of my fins are swept back and the risk of popping one on landing is high.

 

[cls]

Outstanding work! Exciting, thanks for your extensive detailed posts and pix. Wow, just wow.

 

[robopup]

Very cool thread, I'm actually building a staging stack quite similar to this at the moment. One question though.

Nice! Any documentation or pictures I can take a look at (read: steal from)?

Please correct me if I'm wrong, but I'm reading that as your sustainer fins are 0.1 inch thick G10. Given your sustainer is simulated to hit ~Mach 3.6ish I'd be concerned about those fins not being thick or stiff enough for that flight profile.
And while in general I agree that t2t isn't necessary for the up bit (assuming your fins are adequate thickness with fillets to match) I'm applying it to everything aggressive moving forward as most of my fins are swept back and the risk of popping one on landing is high.

That's correct, bare 0.1 in G10 on the sustainer, 0.125 in G10 on the boosters. The thickness of the fins is certainly a concern, but it's actually a ways down the list of things I'm worried about on this project (famous last words?). Over the last couple years I've been playing with how thin fins can go, and also more aggressive fin planforms with the trailing edge tip swept behind the trailing edge root. This has been my basic profile, where span ~1.3x tubing diameter and flown with three fins:


Here are a few of the more aggressive successful test flights with this planform, no tip to tip on any of them:

• 38mm J570 - 0.045 in fin thickness, T700 prepreg carbon [-45/0/45/90]s - mach 2.5
• 54mm K250 - 0.045 in fin thickness, T700 prepreg carbon [-45/0/45/90]s - mach 2.25
• 75mm ~M1200 - 0.125 in fin thickness, G10 - mach 2.75 (simulated - to use Tony's phrase, rocket is "misplaced" right now, but I'm confident things held together through boost)

A few other thoughts on the comparison with the two stage, which I've used to convince myself this design is more conservative than my tests:

• planform trailing edge tip is in front of trailing edge root. In various FE models I've run for fin thicknesses and planforms I was considering for this project, this pushed the natural frequencies up ~15%
• 4 fins instead of 3 - this should put less load on any individual fin, and also allows the span to drop to ~1.1x diameter
• Max q on the single stage flights above are all ~2x that of the two stage sustainer. Max speed on the booster is ~mach 1.75
• Kip's Mesos fins were quite thin (they did still use tip to tip)

I've never had a fin break on landing, although I also have never flown a configuration where the fin extends past the bottom of the body tube.

@cls - thanks!

 

[plugger]

Interesting results Rob, and thanks for the reply! Sounds like you've been doing some pretty aggressive testing. Quite impressive to hit Mach 2.5 on a J570! For the record, I don't think your booster is a concern, but as mentioned the M3.5+ combined with thin FG fins caught me off guard. I'm also surprised you've switched from CF to G10. But it sounds like you've done testing and analysis to support your choices. Looking forward to seeing this thing fly!

Nice! Any documentation or pictures I can take a look at (read: steal from)?

No documentation as of yet, but just this weekend I started building.

Here's my 'kit'


ISC and aft end of the sustainer


I'm very pleased with this, the coupling point is quite rigid


A mate of mine made me the Aluminum ISC, it's tapered internally to match the 54mm CTI Tailcone taper so it seats really well.


And after getting that to fit I figured I may as well start tacking fins. Hopefully on Thursday I'll start laying fillets.


My planned motor config is a M840 to L265. I'll be using HEI to light the L. Sims fluctuate between 60-80k ft depending but until I actually build the thing I'm not worrying too much about the software/sim side.

 

[robopup]

@plugger - Thanks for the pictures, that's excellent, keep us updated and have fun! You have me going back and checking my sims again now... I'm only ~10% more impulse on both the booster and sustainer motors than you.

Regarding CF vs G10:

• Yes I only mention the booster as its max speed at low altitude/high density air is quite a bit lower than some of my other test flights.
• Choice was inspired in part by @watheyak's 54mm and this N5800 flight
• I've found flat panels to be one of the harder things to reliably make.
  • I've flown a few rockets with fins made from wet layup plain weave, but I had a hard time preventing the weave from shearing a little bit. And as soon as it shears, there goes the symmetric layup and you get a little bit of warp in the final product.
  • The flights mentioned above used uni prepreg which is much easier to consistently hit desired angles. I'm still only pulling vacuum on them though - not the kind of pressures I understand G10 is made with - and I'm at least as concerned about the leading edges as the bending and resistance to flutter and probably more so.
• Best of both worlds is probably uni carbon panels manufactured with whatever process G10 is. I've looked a little for these, but pretty quickly got scared off by the price and/or because it's not clear how they're actually made...

These completed sustainer fincans feel much more solid than the fincans I used on those single stage flights with the 'ole (highly scientific) bend-a-fin-by-hand test. We'll find out eventually!

 

[plugger]

@plugger - Thanks for the pictures, that's excellent, keep us updated and have fun! You have me going back and checking my sims again now... I'm only ~10% more impulse on both the booster and sustainer motors than you.

No worries Rob! Same to you, I'll be following your thread closely.

 

[robopup]

Second stage ignition noodling

I've been digging through threads here for the last few months on second stage ignition and I think I have a rough plan of what I want to do now:

• No head end ignition - I want to be able to do a full up electronics test at the pad. Accordingly, I'll be using Jarvis style flat speaker wire down the side of the second stage case
• I'm leaning towards 1-2 pyrodex pellets set into the top grain as opposed to some of the other fancier (at least to me) solutions like KBNO3 pellets. I'm much more comfortable with the pyrodex pellets and have had very reliable single stage ignition the last few years using them. The pellet diameter is something like ~0.45 in and the core of my sustainer is 0.375 in so I'll have to open up the top grain a hair or reduce the pellet OD to place them in. Probably I'll open up the core slightly there for the added benefit of holding the pellets in place.
• I'll ignite the pellet(s) with an ematch, supported with a 1/8 in dowel rod
• Burst disk to support dowel rod and hold up to certain pressure

A rough concept design for this burst disk looks like below. I haven't sized the tabs yet, and haven't nailed down an oring size yet...




A couple questions I haven't tracked down answers to so far:

• Any thoughts on an appropriate pressure to try and contain before the disk fails? I'm considering something like 200 psi right now but this is an absolute roll of the dice between a friend and me.
• Perhaps a noobish two stage question...: I would be shocked if the speaker wire didn't break when the burst disk fails but I had a nightmare where somehow this didn't happen and I ended up with the entrails of the igniter and burst disk dragging behind the sustainer. Accordingly, I've been considering connectors to the aft of the burst disk (but would prefer not to use any). I went through @JimJarvis50 's three stage thread again and found a couple pictures but nothing conclusive. Right now I'm planning to either thread the speaker wire through the burst disk hole or to crimp a short length of shooters wire as soon as the speaker wire wraps around the aft end of the motor.

EDIT: I haven't gone through all of it yet, but finally stumbled on NASA/SP-8051, Solid Rocket Motor Igniters https://ntrs.nasa.gov/api/citations/19710020870/downloads/19710020870.pdf, which seems excellent.

 

[JimJarvis50]

Rob,

The speaker wire approach can work (I've used it many times). But, the connection at the bottom of the rocket can be rather fragile. I would plan to minimize manipulation of the wires to the extent possible. I haven't ever used the technique on 54mm motors. The wires that I use have some width and I suspect that the interstage coupler would need to be designed to accomodate the wires, and then designed so that the sustainer can't turn against the booster.

I'm no expert on the theory of igniting solid motors, but my belief is that you should either go with an instant-on approach with a burst disk or arrange for the propellant to provide the pressure (as we do when lighting motors on the ground). I think the 200 psi is not high enough for instant-on. One strategy for getting sustainers lit is to arrange for some fire over a period of seconds, as opposed to the very short-term pellets. An approach that has worked for me is to pot some magnalite pyrogen in the smoke grain (or in a slug of epoxy if there is no smoke grain). Think a 1/2" diameter by 1" long cylinder containing a couple of grams of pyrogen. It looks like this.

Motor igniter

Jim

 

[robopup]

Hey Jim,

Interesting, thanks for the response!

• Speaker wire - copy that, sounds like I'm overthinking things, don't need connectors at the aft end of the motor. I would prefer to run two pairs of wire down the sustainer - 1 for the igniter and another for the separation charge - so that all the staging logic is off the same altimeter. The speaker wire I got to play around with is 0.25 in width. I attempted to size my 54mm ISC tube correctly to pass a tape sandwich of [cellophane tape/speaker wire/cellophane tape/aluminum tape]. One pair has a slight amount of wiggle and the ISC falls freely if held vertical. Two pairs has no wiggle and the ISC does not fall freely (but is still a smidge looser than I usually do my nosecones).
• Sustainer turn against booster - I put your keyed design in my back pocket and still may do something similar but would prefer not to. I think the aluminum tape does a really good job protecting the speaker wire.
• motor ignition - Perhaps no expert, but you've had a lot of success...! Thanks for the video. I still have a lot of reading and testing to do before I make up my mind on this. Whatever I come up with, I'm planning to use it as a test when I static fire one of my sustainer motors, and fly the another as a single stage.

 

[OverTheTop]

I think your idea for the burst disc has merit. I would be concerned about only some of the tabs breaking, and the remaining piece of plastic momentarily directing the thrust off-axis. That could really upset the flight. Is there a way of maybe having four separate clips holding the disc in, so that when the disc starts to exit one it basically falls out of the others? Or perhaps some other method that would work too.

 

[robopup]

@OverTheTop - yes I'd wondered about that too. Right now I'm planning to use this filament, which has a surprisingly brittle failure mode, especially when it hasn't been heat treated. I may also fiddle with the slicing so that the tabs are printed such that all the filament is printed tangentially. We'll see what testing reveals...!

EDIT - and I may still omit tabs or similar altogether and hold things together with something like aluminum tape. My goal in general is just to make things as repeatable as possible.

 

[OverTheTop]

Good to see you thinking about this. Glad you are considering the minutiae of the 3D printing.

Also, remember, always design for off-nominal-situations .

 

[robopup]

ISC

Analysis:

There's a lot of tribal knowledge that ISCs see some pretty serious loads, so going into this project this was probably my #1 concern structurally. Early on I built several structural FEA models to get a feel for things. I found it fairly tricky to get boundary conditions and contact correct (I'm not much of a nonlinear contact guy) for this problem.

At the time I built these models, I assumed my rolled 54mm and 75mm tubing, connected by several G10 centering rings. I don't have material test data for my carbon fiber product, but have tested coupons with something similar and made an educated guess.

I still haven't convinced myself my methodology was correct, but here's what I did: I made a guess at the loading from Rasaero - I applied the calculated normal force of the sustainer at 4deg + mach 2 (fastest ISC speed) through the sustainer Cp, and also the force of the sustainer at maximum acceleration (these two events obviously don't happen at once, but I wanted to be very conservative).

Results - Displacement looked qualitatively reasonable, and I was getting contact in all the expected places. I was fairly surprised how low the fiber direction strains were, even on the (what I believe is an) extreme case described above I was at less than ~50% max strain on both the carbon and G10. Having said that, while I feel comfortable with the trends I was seeing, I don't feel great about, or have any expectation, to be able to nail failure on this problem the first go around, so take all this with a grain of salt...

Design:

With the data from above, I did feel okay to replace the G10 with a fully 3D printed structure.

High level, the design ended up being pretty straightforward. I rolled pieces of 54mm (5 wraps of 6oz plain weave, two more than I usually do for 54's) and 75mm tubing (my standard 5 wraps of 6oz plain weave). These are connected by a 3D printed part. I usually use motors as my couplers to nosecones to keep things short and it's the same idea here with the ISC. Here's the general design:


I had everything set up in CAD for the ISC a few months ago, but then realized I could sneak the ebay in between the 54mm and 75mm tubing - this shaves a couple inches off the ISC. My altimeter doesn't quite fit unless I leave the terminal blocks off, so I'll be soldering ematches (or probably another wire, then twist the ematches) directly to the board:



Build:

It was sneaky difficult for me to nail all the tolerances just right once I decided to incorporate the ebay on the side of the 54mm tubing. Here's all the test pieces, failed pieces and hatch practice pieces (at least they were printed at 5% so not actually too much filament) it took:


All components before bonding. The hatch was cut from the 75mm tube using a dremel. The 3D printed piece is from PET. I varied the density axially so that essentially there are three 0.5in thick 100% density centering rings at the top, middle and bottom of the part as well as 100% density where the diameter changes. The rest is printed at 30%.


After bonding the 54mm tube into the printed part. I even remembered to get my eyebolts in with some kevlar attached so I don't have to go fishing later!


Bonding to the 75mm tube. I ran out of E-60HP so I used standard JB Weld for this step. I had no confidence I'd be able to nail the hatch cutout with my dremel, and I didn't want to build a jig (although John Coker has an excellent article on his site how to do it), so I ended up buying some PVA filament to print a replica hatch. I attached this after bonding to the 75 mm tube, then filled all the cracks in with epoxy and dissolved the PVA in water after epoxy cured. It looks pretty messy here:


But sanded surprisingly clean - just a little more work to get everything completely flush:


Ebay:


Looks awful right now, but actually pretty smooth and just about ready to be primed and painted:

 

[MClark]

This is a 30 year old picture. 1990 tech, simple timers.
Timer assembly retained by a 0.06” disc of G10, not seen here. Disc is trapped between snap ring and nozzle washer. Motor was not sealed but throat area reduced by half.
Electronics are destroyed at ignition.

if a sealed burst disc of an mass is used the pressure goes up high and fast. A static test to determine failure pressure is pretty much useless.

 

[robopup]

@MClark thanks for the comments. Interesting, I'd been pretty locked in on pressure for ignition but hadn't given time any thought. I think I have more reading to do...

 

[MClark]

With head end ignition I have used really heavy duty aluminum foil RTV-ed to inside of nozzle, flat across. Essentially zero mass.
When assembling motor ignitor is not in so it can vent as closure goes in.

 

[MClark]

And another thing……
Depending on how much pressure drop there is when the disc bursts it can cause the motor to snuff.

 

[ZEDL1]

This is a 30 year old picture. 1990 tech, simple timers.
Timer assembly retained by a 0.06” disc of G10, not seen here. Disc is trapped between snap ring and nozzle washer. Motor was not sealed but throat area reduced by half.
Electronics are destroyed at ignition.

if a sealed burst disc of an mass is used the pressure goes up high and fast. A static test to determine failure pressure is pretty much useless.

View attachment 576648

 

[ZEDL1]

Thank you for this photo Mark. I remember it very well.
I really, really miss Frank.
Jim

 

[StreuB1]

A good means of providing a frangible means of building pressure in the sustainer motor is before you assemble the motor. Take a ball of aluminum foil and form it into the convergent section of the nozzle with a little passing to the divergent section and then press it into place. This will blow out easily, still will breathe a little and become altitude compensating, but still provide enough of a blockage that will allow the motor to pressurize a bit.

Marks comment about time is critical here and is one that you cannot miss. Pressure alone means nothing unless you know the time with which that pressure is acting upon. This is your dt from calculus.

 

[stealth6]

Just to note.......I'm really enjoying this thread.

s6

 

[robopup]

Thanks again @MClark and @StreuB1 for the comments - I still have a lot to think about on this sustainer ignition stuff... I'd also really prefer to avoid anything that requires taking the motor apart at the pad (and I still plan on not using head end ignition for this project).

Accordingly, I'm most interested in something where I can connect wire from altimeter to igniter leads and either screw the assembly into tapped holes in a nozzle washer or retain it with the aft snap ring, so on the divergent side of the nozzle.

I've also been rereading all @jsdemar 's excellent advice around here on igniters and second stage ignition, in particular this thread. I have a few sticks of 38's sitting around I've been meaning to burn. Some igniter+burst disk testing seems like a good use of them.

 

[plugger]

Thanks again @MClark and @StreuB1 for the comments - I still have a lot to think about on this sustainer ignition stuff... I'd also really prefer to avoid anything that requires taking the motor apart at the pad (and I still plan on not using head end ignition for this project).

You can still use HEI and not have to take the motor apart at the pad by leaving the rear closure unattached before assembling the full stack on the pad. At least that's what I'm planning to do.

 

[robopup]

You can still use HEI and not have to take the motor apart at the pad by leaving the rear closure unattached before assembling the full stack on the pad. At least that's what I'm planning to do.

Absolutely a valid way to do things. I'm still making friends with two stagers though, and as a hobbyist doing this once or twice a year, am pretty sold on @JimJarvis50 's preference to do a full up electronics test immediately before an igniter goes into a motor. May or may not stick with this method, but I have a ways to go to be comfortable enough around multi stagers to go that route. To a lesser extent, I would also prefer to not be messing with closures at the pad - I'm sure I'm going to have enough to worry about at T-20 minutes on these early flights.

 

[robopup]

(franken) Tower:

A good friend and rocketry mentor insists on flying out of towers. For everything; Estes, sport flights, min diameter, everything. After 15 or 20 years of that kind of indoctrination, I'm drinking the kool-aid as well.

The couple catches for this project are that 1) this is four fins, everything I had was set up for three and 2) I wanted something a bit longer in an attempt to keep things straight off the pad.

My starting point was my tower below. Three basic components, a base, rails and triangle. I've flown D's to N's out of this.


I started out with a bunch of stuff ordered online, mostly brackets and rails from McMaster. I've slowing been drilling and tapping a ton of holes the last few weeks.


Tower base:


With mounts for rails. One unintended (but very appreciated) result of holding onto the rails this way is that I've found if a motor catos, the rails just slide out of the base rather than bending.


I converted my top triangle to a square (it can now be used in either configuration - that's what the unused brackets are still doing there) and built an additional one to support things midway:


All 14 feet - I'm stacking my original 8 ft rails on top of new 6 ft rails. In the future I'll probably keep the top square ~2-3ft below the ends of the rails. I also need to either move my stakes out of get a second set for the top square ropes. I figured I'd stick a rocket in it too while I was working on things...


The full up test revealed one oversight - the only thing holding the rails to the small angles at the base are 1/4-20 rods running axially into the rails. This works fine with the single triangle, but just too much weight with two squares.


All of that packs back up into this (and my triangle base still fits in the backpack as well!). I need to get some more pvc pipe for the new rails.


And one gratuitous shot of both rockets before I start priming and painting. I haven't cut my cases down to size yet...


Hopefully this pad works better than SpaceX's recent test!

 

[rharshberger]

(franken) Tower:

A good friend and rocketry mentor insists on flying out of towers. For everything; Estes, sport flights, min diameter, everything. After 15 or 20 years of that kind of indoctrination, I'm drinking the kool-aid as well.

The couple catches for this project are that 1) this is four fins, everything I had was set up for three and 2) I wanted something a bit longer in an attempt to keep things straight off the pad.

My starting point was my tower below. Three basic components, a base, rails and triangle. I've flown D's to N's out of this.
View attachment 577439

I started out with a bunch of stuff ordered online, mostly brackets and rails from McMaster. I've slowing been drilling and tapping a ton of holes the last few weeks.
View attachment 577440

Tower base:
View attachment 577441

With mounts for rails. One unintended (but very appreciated) result of holding onto the rails this way is that I've found if a motor catos, the rails just slide out of the base rather than bending.
View attachment 577442

I converted my top triangle to a square (it can now be used in either configuration - that's what the unused brackets are still doing there) and built an additional one to support things midway:
View attachment 577443

All 14 feet - I'm stacking my original 8 ft rails on top of new 6 ft rails. In the future I'll probably keep the top square ~2-3ft below the ends of the rails. I also need to either move my stakes out of get a second set for the top square ropes. I figured I'd stick a rocket in it too while I was working on things...
View attachment 577444

The full up test revealed one oversight - the only thing holding the rails to the small angles at the base are 1/4-20 rods running axially into the rails. This works fine with the single triangle, but just too much weight with two squares.
View attachment 577445

All of that packs back up into this (and my triangle base still fits in the backpack as well!). I need to get some more pvc pipe for the new rails.
View attachment 577446

And one gratuitous shot of both rockets before I start priming and painting. I haven't cut my cases down to size yet...
View attachment 577448

Hopefully this pad works better than SpaceX's recent test!

You do realize that Bob is a really bad influence........pair him and Marty and your hooked for life....

 

[plugger]

Absolutely a valid way to do things. I'm still making friends with two stagers though, and as a hobbyist doing this once or twice a year, am pretty sold on @JimJarvis50 's preference to do a full up electronics test immediately before an igniter goes into a motor. May or may not stick with this method, but I have a ways to go to be comfortable enough around multi stagers to go that route. To a lesser extent, I would also prefer to not be messing with closures at the pad - I'm sure I'm going to have enough to worry about at T-20 minutes on these early flights.

I'm not entirely across Jim's pad electronics testing methodology but from my perspective I plan to use Blue Ravens in both rockets to allow me to remotely arm both charges per rocket as well as the HEI staging channel in the sustainer. In fact, my 2x BRs just arrived in the mail today!

 

[robopup]

I'm not entirely across Jim's pad electronics testing methodology but from my perspective I plan to use Blue Ravens in both rockets to allow me to remotely arm both charges per rocket as well as the HEI staging channel in the sustainer. In fact, my 2x BRs just arrived in the mail today!

Nice! Before I started doing my own electronics, I was big into Featherweight (I think I still have an operational parrot or two hanging around). Pretty cool to watch what Adrian has put together on this new one.

I must confess - I haven't been following the dev thread closely enough so I'm not really clear on what "remote arming" of the outputs means and the actual hardware involved in making it happen. Personally, I wouldn't feel comfortable giving an altimeter power by screw switch (even if outputs weren't armed) unless it were still backed up with these switches behind the altimeter fet. These could be wireless (and if I flew HEI, these are probably the switches I'd make wireless), mine are just screw switches. I may still make them wireless, with the same traditional ignition setup I have - I have some extra room in my ebay.

Edit - I suppose another alternative is to have a wireless switch turn on the altimeter itself.


From a design perspective, my altimeter which controls second stage ignition gets bolted onto the front of the motor like below. Then a piece of airframe and coupler tubing slides over the top. I would like to have the forward section of the rocket buttoned up before I take it to the pad, while also not having an igniter in the motor any longer than necessary. This would require that all above happens at the pad (I'm using a snap ring motor), which I'm not thrilled about.


I have done a couple full mockups with my casing, fincan, payload and burst disks with the speaker wire method. So far I've been pretty happy with how things assemble.

Lots of folks have been safe and successful with HEI. It's just not something I'm personally interested in atm, or find necessary as a solution to some design/configuration problem.