J record attempt using Loki J1026

[Adrian A]

Things are coming together for a flight on my new sustainer rocket tomorrow, using a Loki J1026 motor.



This stage is designed as a new sustainer for my StratoSpear 3-stage rocket, and most of the build thread is over in the multi-stage sub-forum.

For electronics, I'm using 2 new 29mm diameter Blue Raven altimeters and a Featherweight GPS Tracker, which fit together with 2 batteries into an av-bay with 3" long threaded rods that are also the deployment terminals.

The 2 altimeters, 2 batteries, GPS tracker, 2 magnetic power switches (built into the 29mm Blue Ravens), top and bottom bulkheads, and attachment hardware to the motor and the chute cannon altogether weigh 69 grams. The chute cannon, chute, nosecone ejection piston, and harness combine for 76 grams.



The aft body with the fins is 120 grams, and the scratch-built nosecone is 82 grams, for a grand total of 347 grams empty weight. The motor weighs 1171 grams. The airframe, recovery and electronics come in at 23% of the launch weight.

Below is the full configuration with the av-bay, chute cannon, and the top of the nosecone ejector piston peeking out. You can see the antenna taped to the side of the carbon fiber chute cannon. Not ideal for RF range, but fortunately the Featherweight GPS tracker system has a lot of range margin.


The rocket is 39" long. 24.5" is motor, 11.2 is the curved part of the Von Karman nose cone, so the rocket is 3.3" longer than it could be if the recovery gear were smaller.
The rocket is expected to have about 100 Gs for 1.0 seconds, accelerating to Mach 2.8 in that time, if everything holds together. That's a pretty significant if, because this scratch rocket is all brand new. The fin stock is 0.060 mostly-unidirectional that I haven't used before, though the manufacturer claimed it used high-Tg resin. The carbon tubing is also supposed to use a high-Tg epoxy. The fins are a little delicate-looking to my eye, but at least I have some nice big fillets on them. They feel pretty stiff considering their shape and size. I installed them with a new fin jig I made from aluminum angles, and I think they're quite straight. The reason for the big sweep is that according to RASAero, the limiting case for the sustainer for stability is during the 3-stage configuration when the sustainer is going Mach 4 at 30,000 feet, and at those speeds, the sweep apparently really helps.


The nosecone and the fin fillets used Cotronics 4461 epoxy. Today I post-cured the airframe at 250F for a couple of hours and it came out just as smooth as it went in. This rocket has about 6.3" of stability margin, since it's designed for a sustainer application where after this flight I'll cut off about 3" of the aft body to expose that much motor as a coupler, moving the fins forward by about 3."

The deployments use a chute cannon that is new for this rocket but the same concept I have used before. The nosecone ejection uses a new small-diameter aluminum piston that's designed for reliability at very high altitudes. The ground testing went fine, with no surprises. The av-bay with 2 Blue Raven 29mm altimeters sharing 3 threaded rod/deployment terminals is similar to the setup I used last week in a 29mm rocket. The harness is fresh. The chute is scratch-built but well-tested on 29mm and 38mm rockets.

The rocket sims to a little over the current J record of 23,725 ft, held by Curt von Delius.

 

[tim cubbedge]

Where are you attempting this? Good luck!

 

[pugachu]

Looks great, Adrian!

How close are you to being at the optimal mass for that airframe? It looks like you ditched the tungsten nose, and the rocket is just about all motor.

 

[watheyak]

Awesome! Best of luck!

 

[Adrian A]

Where are you attempting this? Good luck!

The Northern Colorado North site in the Pawnee national grasslands. After years of persistent drought, it’s crazy wet out here. This is the first monthly launch since since early May that wasn’t rained out.

 

[Adrian A]

Looks great, Adrian!

How close are you to being at the optimal mass for that airframe? It looks like you ditched the tungsten nose, and the rocket is just about all motor.

It’s about 0.1 lbs under optimal mass according to RASAero, giving up a bit under 100 feet of altitude, IIRC. I have maybe 10 grams of tungsten powder in the tip where I potted in the harness attachment hardware.

 

[kjhambrick]

Adrian --

I just found your post but I also learned about the TRF 'follow feature'.

That's a beautiful rocket !

Good luck on the flight -- I am looking forward to your usual thorough launch report

-- kjh

 

[Rschub]

I am curious about the use of unidirectional material for the fins. What is your thinking on the advantages/disadvantages compared to bidirectional? I did read the build thread and saw the fiber orientation.
Good luck with the launch!

 

[Adrian A]

The flight started out well. The boost was straight, about 2 degrees from vertical. It rolled only about 2 revs during the motor burn and then rolled the other way back to zero in the next second. The max Gs were 118. The max velocity was 3107 feet/second, vs 3170 predicted. After the burnout, the drag deceleration was 20.5 Gs, compared to the predicted 22.5 Gs, so it was on track to break the altitude record.

But then the fins started fluttering. Even as the velocity went down, the axial drag deceleration went up to 23-28 Gs, and the lateral acceleration was +/- 20 Gs.

The apogee was right around 17k.

At main deployment, the main chute was tangled so it landed at 60 feet/sec and broke 2 of the 3 fins off.

Here you can see the creases in all 3 fins where from the fluttering.

More data tomorrow when I have more bandwidth.

 

[Sandy H.]

The flight started out well. The boost was straight, about 2 degrees from vertical. It rolled only about 2 revs during the motor burn and then rolled the other way back to zero in the next second. The max Gs were 118. The max velocity was 3107 feet/second, vs 3170 predicted. After the burnout, the drag deceleration was 20.5 Gs, compared to the predicted 22.5 Gs, so it was on track to break the altitude record.

But then the fins started fluttering. Even as the velocity went down, the axial drag deceleration went up to 23-28 Gs, and the lateral acceleration was +/- 20 Gs.

The apogee was right around 17k.

At main deployment, the main chute was tangled so it landed at 60 feet/sec and broke 2 of the 3 fins off.

Here you can see the creases in all 3 fins where from the fluttering.

More data tomorrow when I have more bandwidth.
View attachment 596109

Given your background, I assume you consider this a successful test flight, as you have found pretty good validation of some of your modelling and have found an area for improvement. That's how I read it, at least (but in all fairness, it did come out of your wallet and it was your time spent. . . ).

Thanks for the update and I look forward to your future attempts!

Sandy.

 

[tsmith1315]

Here you can see the creases in all 3 fins where from the fluttering.

Very interesting to see the fins creased, and consistently, but still in one piece.

 

[pugachu]

Sorry about the damage, but at least this failure mode happened now, rather than on your planned 3 stage flight. It looks like the body tube is still in good shape.

I look forward to reading your detailed flight report tomorrow. It would be interesting to compare the differences between the data on the redundant Ravens

 

[0011001100]

But then the fins started fluttering. Even as the velocity went down

Aeroheating causing the epoxy to weaken? That would be my guess as to why it happened after motor burnout.

 

[Sandy H.]

Aeroheating causing the epoxy to weaken? That would be my guess as to why it happened after motor burnout.

My gut feeling (0 qualifications, so zero validity) is that when it went through a natural frequency, fairly quickly while under power, not too much bad happened, but as it slowed after main power was gone, the transition lasted longer and remaining in that natural frequency range (with elevated temperature as you said) caused the failure.

Either way, I look forward to Adrian's analysis, as it will help educate all of us on the reality of what happened!

 

[VTS SkunkWorks]

Adrian
Awesome flight, and great data! Do you have the velocity profile for the flight?
There is an optimum velocity profile to get the most altitude from a given NS, with a vertical flight (The Goddard Problem: he was a smart guy)
What other J motors did you sim?

 

[OzHybrid]

4 fins can significantly make them stiffer as they'll be smaller with a small weight/drag penalty.
Good luck.
Norm

 

[Adrian A]

Everything from my fin alignment jig to my tower is made for 3 fins but I appreciate the argument for 4 fins a lot more now.

I’m currently thinking of rebuilding with extra plies in the +/- 45 direction for torsional stiffness and maybe also a tip-to-tip layup of Uni fiber rather than changing the fin profile from the swept delta to a clipped delta. For stability at Mach 4, the clipped delta fins would need to be so large that they would force the other two stages to have larger fins also. And since the fin flutter didn’t start until it was starting to slow down, I feel like what I had wasn’t that far from working.

 

[kjhambrick]

Adrian --

Something I read years ago is that fins where the thickness tapers from root to tip can defeat resonance.

This is one reason, without trying too hard for attaining exact ratios, why I have been shooting for chord::thickness ratios of 20::1 across the span.

I'lll see If I can find the paper in my old stuff.

Maybe a similar taper can be accomplished with judicious application of your tip-to-tip material ?

-- kjh

 

[dhbarr]

I've idly speculated from time to time about intentionally doping tips with a little extra mass. Just noodling, never done anything with it.

 

[Adrian A]

Here are two frames at 11 fps, taken by George Barnes:




D'oh! The upper half of the tower is 6' tall, so just the part of the flame visible in the frame is already longer than the rocket.

Ryan Southworth took the following video. Excellent crowd reaction:


View attachment IMG_5292.MOV

Here are the accelerations measured by one of the Blue Ravens:

 

[kjhambrick]

Holy cow !

That acceleration -vs- time plot is one for the book of fingerprints.

Or maybe the same plot with an expanded time axis.

Amazing that there were any fins remaining at all !

-- kjh

 

[SolarYellow]

4 fins can significantly make them stiffer as they'll be smaller with a small weight/drag penalty.
Good luck.
Norm

In my OR experiments, it has gone the other way. Because the CP of the smaller fins is further aft, you can make them a little smaller still. So you come out ahead. It's a small advantage, on the order of a couple percent, if that. But it is in the favor of four fins.

If you actually read Hoerner and look at the right charts in the chapter on interference drag, it's basically zero at any value of thickness/chord relevant to our rockets and appears to even go negative with a small enough t/c. Which is all at subsonic speed, BTW. So the argument that the interference drag of the fillets will make the four-fin rocket inferior doesn't hold up when the sources its based on are examined.

There are arguments about four fins being less susceptible to coning due to inducing less roll during a long flight. Adrian is collecting the data that would prove or disprove that, at least in his rockets.

With the slightly lesser drag of four fins, you can make them a little thicker and therefore stiffer, still further increasing their resistance to flutter.

Building a four-fin rocket fin can is ~4/3 of the labor of building a three-fin rocket. If I was building low-speed, whoosh-pop rockets for competition, I'd probably just go with three fins. You build a lot of rockets, and the time spent on that can matter. The advantages of four fins are only partially able to be realized with low speed, short flights, and the possible remaining advantage is well inside the RMS variation of motors and most of the other factors that matter.

 

[OzHybrid]

In my OR experiments, it has gone the other way. Because the CP of the smaller fins is further aft, you can make them a little smaller still. So you come out ahead. It's a small advantage, on the order of a couple percent, if that. But it is in the favor of four fins.

If you actually read Hoerner and look at the right charts in the chapter on interference drag, it's basically zero at any value of thickness/chord relevant to our rockets and appears to even go negative with a small enough t/c. Which is all at subsonic speed, BTW. So the argument that the interference drag of the fillets will make the four-fin rocket inferior doesn't hold up when the sources its based on are examined.

There are arguments about four fins being less susceptible to coning due to inducing less roll during a long flight. Adrian is collecting the data that would prove or disprove that, at least in his rockets.

With the slightly lesser drag of four fins, you can make them a little thicker and therefore stiffer, still further increasing their resistance to flutter.

Building a four-fin rocket fin can is ~4/3 of the labor of building a three-fin rocket. If I was building low-speed, whoosh-pop rockets for competition, I'd probably just go with three fins. You build a lot of rockets, and the time spent on that can matter. The advantages of four fins are only partially able to be realized with low speed, short flights, and the possible remaining advantage is well inside the RMS variation of motors and most of the other factors that matter.

I don't think building a 4 fin rocket is 4/3 of the labour of building a 3 fin rocket. The extra work of 1 fin is not 33% of the TOTAL build. While it's never too late to read a good book, the chances of me reading Hoerner are slim. I was suggesting that the height/thickness ratio for a rocket with 4 fins would be higher and therefore the fin would be stiffer(not the material) as the 4 fin would be a lower height than the 3 fin height. Lower height, less bending moment. Lower height, stiffer.
Looks like Adrian is going for thicker(stiffer) fin reinforced with a different fiber orientation. So it's moot.

Next time however, if the fins don't flutter and provide a lot of braking to the rocket, the fins may be in a high mach zone for longer and subject to friction heating. A region they've not yet been fully tested in. Going for records is tough.

 

[Adrian A]

In this case it would be more than 4/3 because it would also mean rebuilding a tower, a fin jig, and 1-2 other stages.

Max Haynes sent me another video that shows almost the whole boost:

View attachment IMG_0108.mov

 

[Adrian A]

One thing I have wondered is about comparing heating rates from lower speed at lower altitude to higher speed at higher altitude.

The rocket drag times the velocity is a measure of the power lost to drag. This is straightforward to calculate from simulation results, but I’m not sure it translates directly to a relative measure of aero heating of the body because the relative fraction heating the air might not be the same. For example, blunt bodies have less of an aeroheating problem than a sharp pointy object because more of the energy gets transferred to heating the atmosphere and less to heating up the body. But if I have the same pointy object going faster through thinner air, with the same total drag power as the object going slower through thicker air, would the two cases heat up the body at the same rate?

 

[Adrian A]

Holy cow !

That acceleration -vs- time plot is one for the book of fingerprints.

Or maybe the same plot with an expanded time axis.

Amazing that there were any fins remaining at all !

-- kjh

I’ll have more plots tomorrow.

 

[OzHybrid]

One thing I have wondered is about comparing heating rates from lower speed at lower altitude to higher speed at higher altitude.

The rocket drag times the velocity is a measure of the power lost to drag. This is straightforward to calculate from simulation results, but I’m not sure it translates directly to a relative measure of aero heating of the body because the relative fraction heating the air might not be the same. For example, blunt bodies have less of an aeroheating problem than a sharp pointy object because more of the energy gets transferred to heating the atmosphere and less to heating up the body. But if I have the same pointy object going faster through thinner air, with the same total drag power as the object going slower through thicker air, would the two cases heat up the body at the same rate?

Historically, slow and steady seems to have yielded better record results. When I was simming a hypothetical motor, slow through thick atmosphere and accelerating after 30kft gave good results. Of course that's a difficult profile to get from a solid propellant motor. I suppose you could do it with a core burner for take off going to an end burner for thick atmosphere going to core burner, but I don't think anyone has been crazy enough to do that and the nozzle wouldn't be matched for all of those profiles.
@Chuck Rogers RasAero might be able to assist with heating profiles for your flight.
You're going into velocity regions where few dare to go.

 

[VTS SkunkWorks]

I have NEVER attmpted an altitude record so this is not experience, just math.
In fact my favorite saying is "In theory there is no difference between theory and practice, but in practice there is".

When I ran the calcs for the To100k two stage rocket, I discovered that sub-sonic at lower altitudes provided the most altitude for a given impulse motor.
Look at column W in the spreadsheet and find the row with the lowest total impulse loss for that period, then follow left to column Q for the velocity that provides the lowest loss. This is for the sustainer when at 1000 meters altitude. This all assumes a vertical flight!
The graph shows the results for multiple spreadsheets.

 

[Adrian A]

Historically, slow and steady seems to have yielded better record results. When I was simming a hypothetical motor, slow through thick atmosphere and accelerating after 30kft gave good results. Of course that's a difficult profile to get from a solid propellant motor. I suppose you could do it with a core burner for take off going to an end burner for thick atmosphere going to core burner, but I don't think anyone has been crazy enough to do that and the nozzle wouldn't be matched for all of those profiles.
@Chuck Rogers RasAero might be able to assist with heating profiles for your flight.
You're going into velocity regions where few dare to go.

If a commercial motor manufacturer ever makes an end-burning 38mm full J, I’m pretty sure the J altitude record would go over 30,000 feet. But It would need one heck of a liner. In the meantime, the Loki J1026 sims the highest of anything out there that I have seen, despite being a little longer than the J510 (current record) and having a super short burn time.

 

[Chad]

Did you determine flutter from the data or from inspecting the fins after the flight? I wonder how flutter would show up in data.