StratoSpear: 38mm 3-stage for complex L record attempt

Adrian A

Sponsor
TRF Sponsor
TRF Supporter
Joined
Jan 21, 2009
Messages
2,561
Reaction score
881
Location
Lakewood, CO
I prepped to fly today at Balls, but the prep took too long and my tracker and 2nd stage tracker batteries got down to about 3.6 and 3.7V. I like to fly with them higher than that, so I decided to scrub to recharge and try again tomorrow. The weather here couldn't be more perfect today. I hope it will be good tomorrow too.
 

Adrian A

Sponsor
TRF Sponsor
TRF Supporter
Joined
Jan 21, 2009
Messages
2,561
Reaction score
881
Location
Lakewood, CO
I made the flight attempt on Saturday, and only the first stage lit. The good news is that I got everything back, but the bad news is that because of a number of mistakes on my part, not everything is flyable again. I'll try to document all the key events and problems now while it's fresh in my memory, for reference later and so if anyone else is interested in wading through the details maybe they can learn too:

First some photos:
IMG-9854.jpg

My launch tower the evening before the flight, at the 1500' pad area. The Black Rock Desert is a spectacular place for rocketry. Everyone who is intensely into the rocketry hobby owes it to themselves to fly there at least once. It could be a long time before the weather was as good as Balls 30 though, with clear skies and almost no wind at any altitude for all 3 days.


IMG-9859.jpg
The Strato-Spear rocket and its builder, happy to have the prep finally done.

IMG-9861.jpg

Stratosphere in the tower. The tower poles are 6' tall, for reference.

The standard battery for the Featherweight GPS tracker is 400 mAhr, and it fits nicely into my combined Blue Raven/GPS av-bay:

IMG-9830.jpg
The av-bay has separate batteries for the tracker and Blue Raven prototype, and the Blue Raven is turned on and off with a magnetic switch built into the round circuit board at the bottom, which is the active bulkhead for the Featherweight 38mm av-bay kit. There is also a screw switch for arming the airstart. The tracker, however, just has a manual slide switch on the board to turn it on and off. It has accommodations to add an external switch, but I elected not to do that for this av-bay because of space and access constraints, and to keep the build simpler. However, the bad part of that decision is that once I turn on the tracker, the clock starts running because very little of the rest of the prep can happen until the av-bay is installed. On the 400 mAhr battery, the tracker has several hours of capacity before and during launch, and then after landing if it doesn't hear from the ground station or Bluetooth from the phone it goes into a "lost rocket" mode, where it can last for 16 hours. But with trackers in my sustainer and 2nd stage, both of which have a lot of prep after the av-bay is buried, the prep becomes a race against time, which is something I want to avoid the next time I try something like this.

Here's my sustainer prep checklist:
IMG-9828.jpg
I forgot to print this out while I was at home, so I took a photo of my laptop screen and referred frequently it on my phone.

Here's what the bottom of my chute cannon looks like after step 14, before I screwed down the chute cannon:
IMG-9832.jpg

The bare wire that's unlabeled is the tracker antenna. I'm glad I took this photo because of a problem I discovered when reviewing the flight data later.
 

Adrian A

Sponsor
TRF Sponsor
TRF Supporter
Joined
Jan 21, 2009
Messages
2,561
Reaction score
881
Location
Lakewood, CO
I'm using head end ignition and the igniters are installed with the electronics from the start. In addition to the switch controlled by the Blue Raven, there is a magnetically-activated switch that disconnects the battery from everything, and on top of that there is a screw switch that disconnects the airstart igniter even when the altimeter is powered on and the deployments are armed. That makes 3 inhibits for 2-fault tolerance. But to avoid any doubts about safety, I also keep the motor propellant grains out of the case until I get out to the pad. This makes for a less useful RSO check, since they have to take my word about the fully-loaded center of gravity, but after some good questions and discussion, and a nice distant pad area assignment from Deb Kolombs, I passed the RSO check and went out to the pad with Paul, a Wyomingite who nonetheless was part of our Colorado contingent. (also I want to give a shout-out to Larry Haynes and Vic Davis, who dragged all my stuff across the west so I could fly into Reno and provided the best food and accommodations on the playa).

This is my first attempt to fly with snap-ring motors, so after I installed the motor grains into the 2nd stage motor, I made made the comical mistake of trying to install the nozzle backwards. It promptly stuck, and so I had to go back to our prep area to take it apart to reverse it. To push the nozzle out from the forward end, I needed to take out the 2nd stage av-bay and a cross-bar that is drilled across the forward closure. I got that done and took the rocket back to the tower with my tools, and installed the nozzle the right way this time, and got the snap ring in. In the meantime what I didn't realize is that when I re-installed the cross-bar, I failed to thread it back through the little loop of kevlar of my 2nd stage recovery harness.

Back out at the pad, after installing the 2nd stage and nozzle, I installed the upper stage grains and nozzle. The rocket stages had to be assembled within the tower, and with Paul's help that went o.k. Then I armed the booster deployments, the 2nd stage deployments, and the sustainer deployments, and then armed the sustainer airstart and the 2nd stage airstart. In the meantime the guy on the C pads next to us with a scale AMRAAM had been ready to go since before I realized the backwards nozzle, so we snapped a couple of pictures and quickly departed. On the way back I realized I hadn't double-checked the tower straightness from all angles like I normally would. But it seemed at the time it was straight. In hindsight I should have just gone back to LCO to get the AMRAAM off and then gone back to the tower to re-check everything. But I was pretty frazzled by this time and started to get worried about the sustainer tracker battery, which was now down to 3.7V after about 3 hours of prep. So yeah, I've got to admit a caught a case of go fever.

If anyone has a video of the launch, I'd like to see it. The K1127 motor lit and had a nice purple flame, but the rocket left the tower tilted noticeably to the left. After motor burnout a lot of stuff happened fast, and from 1500' I couldn't quite follow what happened. But the booster and second stage parachutes appeared quickly, and the Featherweight GPS told me that sustainer apogee happened at a little over 12,000 feet. Then the sustainer came down fast. Too fast. The last packet showed it came down at 293 feet/second a little over a mile away, and it was time for recovery.

When we walked in the direction of the sustainer tracking, the booster and its chute were visible right away, and I could see another piece off to the northeast. That piece turned out to be the second stage sans av-bay or chute:

IMG-9864 (1).jpg

Sadly, 2 fins were broken on the impact. I figured that the 2nd stage kevlar harness had broken, but later I found the intact harness loop, which told me about the mistake I made re-threading the 2nd stage crossbar. Paul and I met up again and I was happy to see that the booster was recovered intact and as designed. I went off to get the sustainer by myself. The last tracker position was about a mile away at this point, and as I walked in that direction and my range got down to about 1/2 mile, I saw a pickup truck stop about where I thought the rocket was, while I watched them with binoculars. It was a pretty helpless feeling watching someone do something to my rocket but be too far away to waive them off. Note to people at launches: Please don't touch rockets you found unless you know there's almost no chance the owner is still at the event. Instead, just note the location and let the LCO know where you saw it. When I got to the rocket it looked like this:

IMG_9868.jpg

The unburned motor, with head-end igniter installed, was out of the rocket, with the the head-end igniter wires broken. I have been assuming that the motor was pulled out of the rocket by the people who found it, but now that I think about it, if the motor were inside the sustainer when it landed, I think it would have been buried much more deeply. More likely it fell out when the main chute deployed at high speed, and the people I watched just found it separately and set it next to the rocket.

The main chute had shredded, indicating that my apogee deployment failed to eject the nosecone, but the main chute deployment blew the chute out and the nosecone. The nosecone ejector, piston, and main chute piston were scattered around the vicinity of the rocket and I picked them up, but the rest of the chute was nowhere in sight. Fortunately there was no significant damage to the chute cannon which was buried in the mud, but the front 1" or so of the nosecone broke off.

I'll show a more detailed look at the recorded data later, but the initial look at the sustainer data showed that when the apogee charge fired, there were 8-10 Amps of current but no acceleration, indicating that the leads had shorted, which explains the failed apogee deployment. The apogee wire has a bit more un-stripped insulation in image 9832, but I'm not sure whether the problem was there or somewhere else.

The mystery of why the rocket took off at an angle was solved once I went back to the pad. One of the three tower poles was kicked out to the right. It was one I apparently forgot to tighten down before we left the pad last time. That one stings.

Why both separation charges went off right after the first motor burnout was explained by taking a closer look at my deployment settings. It turns out that I had accidentally set the 2nd/sustainer separation to go off after a motor burnout count of 1 rather than 2. User error. Interestingly, without the sustainer attached, the 2nd stage has so much motor sticking out the back that it is stable going backwards, so after the sustainer separation, it immediately did a 180 flip and continued upward facing backwards until it ejected the av-bay at apogee.

More to come later, but first I need to unpack, etc
 

G_T

Well-Known Member
Joined
Dec 29, 2011
Messages
2,772
Reaction score
731
A notable horizontal speed component at apogee would be expected with a fair bit of tilt to the trajectory. That tends to shred things.
 

Adrian A

Sponsor
TRF Sponsor
TRF Supporter
Joined
Jan 21, 2009
Messages
2,561
Reaction score
881
Location
Lakewood, CO
Finally getting around to sharing some data from the Blue Raven prototype that was flying on the 2nd stage and the sustainer.

Here is accelerometer data from the sustainer, with a variety of zooms into different areas. First an overview of the accelerometer data:

1665625142652.png

The big negative pulse around 45 seconds is the chute shredding after a late deployment, followed by the sustainer impacting the playa.

The boost part is pretty normal but you can see both of the stage separation charges. The first one is the 1st/2nd separation charge which was set to go immediately after burnout, and the second one is the 2nd stage/sustainer charge, which was set for an additional 0.5 second delay after the burnout counter reached the threshold. The threshold should have been 2, but was 1, so it separated after the first stage burnout:

1665625418404.png

The slight noise in the accels shortly before liftoff may be from the beeper. You can se the rocket bouncing around in the tower in the first 0.2 seconds of the flight. When the acceleration exceeds 32 Gs, the Blue Raven automatically switches over from the 32-G range accels to the 400 Gs range accels, which are noisier.

The first separation charge was farther from the sustainer and was pushing the empty booster away from the 2 full stages, so the full stages only recorded about 18 Gs. The second separation charge was closer to the accels in the sustainer, and the two parts were about equal mass, so the sustainer got an 80 G bump forward.

Here is the whole sustainer timeline with the vertical axis zoomed in to about +/- 1.5 Gs.:

1665626035653.png

Before launch the X axis was reading right at 1 G, and after impact, it was reading right about -1 G. In between, you can see the effect of drag on the X axis and spin/wobbles on the Y and Z axes. There is a transition in the drag right around 12 seconds and 33 seconds. The rocket was going about 320 feet/second at those points, pretty far from Mach 1, so that's a little surprising.

So why wasn't there an apogee deployment? The current and voltage recorded shows the problem:

1665626715637.png
When the apogee charge was supposed to fire, the Blue Raven tried its best, sending over 10 Amps out that channel, but because of a short circuit, the charge didn't fire. The main channel firing at 48 seconds is more typical, with the ematch opening up too quickly to measure the current while it was firing.

At impact, the accel data is a little interesting.
1665627020609.png

The high-range accelerometer is has a +/-400 G range, but to use flash recording space efficiently, I only record +/- 327.68 Gs. A value over that threshold rolls over and appears negative. In this case we would expect a very large negative pulse for the impact, so it looks like the actual measured value was 267.41 - 655.36 = -387.95 Gs.

Next time I'll get into the gyro and angle data, and why the 2nd stage didn't ignite, but this is long enough for now.
 
Last edited:

Neutron95

Well-Known Member
TRF Supporter
Joined
Jan 17, 2010
Messages
650
Reaction score
562
The high-range accelerometer is has a +/-400 G range, but to use flash recording space efficiently, I only record +/- 327.68 Gs. A value over that threshold rolls over and appears negative. In this case we would expect a very large negative pulse for the impact, so it looks like the actual measured value was 267.41 - 655.36 = -387.95 Gs.
I am even more excited for the Blue Raven now. I have a minimum mass I1299 design that I've been thinking about for a while that's been on hold for a while since there aren't any commercial altimeters that can record 200+ G accelerations.
 
Top