SpeedStack

Discussions for scratch builders. For the enthusiast that builds from the ground up. Tell us what you've built and what projects you working on in the space race.

Moderator: Moderators

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

SpeedStack

Postby drew » Tue Jun 02, 2015 1:21 pm

Earlier this year at THUNDA I had a two stage project that I wanted to attempt but wasn't able to get the booster together in time. Instead I flew the sustainer which I call 38s on a shakedown flight to ensure stability. That flight was nominal and had excellent performance so I decided to build a booster that would fly on 4 grain AeroTech motors and attempt a flight at Williams Wildfire Westernationals 2015. The sustainer is only really configured for flying AeroTech J510 motors due to the depth of the motor retainer. Both rockets are single break dual deploy designs with avionics housed in the nose and ISC. Raven 3s are in each rocket, tracking is handled by a TeleGPS in the sustainer and a TeleMini in the booster. Chute deployment is handled by cable cutters and sustainer ignition is programmed on the booster's Raven 3 third channel.

Here's the "SpeedStack" in OpenRocket.
Image

The booster flew on a K1100T and the sustainer a J510. The Raven's third channel was custom configured to fire the ignitor at 0.5 seconds after liftoff and at an altitude above 96ft AGL. Here's the stack on the pad ready to go.

Image


Flight

Scoop caught this excellent liftoff photo.
Image

I was able to catch a reasonable shot as well.
Image

I also had my Canon video camera on a tripod recording the launch. You can see that video here.



Recovery

I went off with Brandon and Tom to recover the sustainer, Joe and Michelle went out looking for the booster. I had good data from the TeleGPS so we set off. We basically drove right to the rocket. Here's how we found it.
Image

I couldn't believe it! it must have fallen in a flat spin from apogee. I carefully picked up the rocket and checked the charges after gently removing the nose cone. Both had fired but the apogee charge (0.5g bp) didn't separate the airframe. Another curious thing was that the motor didn't ignite in the sustainer. :? Confused but pleased with what could be described as a miraculous recovery we proceeded to drive back to camp. Along the way Tom casually spotted the booster. Besides the lack of sustainer ignition it had a nominal flight and recovery, as you can see.
Image


Post Flight Analysis

Keen to diagnose what went wrong after the flight on Saturday I jumped into both Ravens and the TeleGPS to get all the relevant flight data. Here's a brief flight summary.

Sustainer Raven 3 data
Max velocity was 1371ft/sec, or just over Mach 1.2.
Max Altitude was 12,039ft AGL.
Sustainer TeleGPS Data
Max Altitude was 12,152 ft AGL.
Landing site roughly 750 meters away from pad.

Here's the flight statistics from the TeleGPS program.
TeleGPS-summary.jpg

Here's a Google Earth overlay of the TeleGPS flight data.
Image

Booster Raven 3 data
Max velocity was 1,197ft/sec, or just over Mach 1.
Max Altitude was 6,994ft AGL.


Failure Analysis

First off, obviously the sustainer did not deploy the recovery gear. Both charges fired in the sustainer at the appropriate time. I guess either the nomex burrito wrapped chute was a bit tight in the airframe or the 0.5g BP charge wasn't enough to separate the the nose from the airframe. Either way next flight I'll attempt to pack the chute more carefully and load more BP into the apogee charge vial. Another valuable lesson from this failure is that just because you've had a successful flight on the vehicle/motor combo you're attempting to do again doesn't necessarily mean it's a trivial exercise. I suspect I was a little more concerned about the booster's ISC and didn't give the appropriate attention to the sustainer. I'll ensure this doesn't happen again. Here's a couple graphs of the Raven 3 in the sustainer's Axial and Lateral Accelerometer data. That tumble was definitely intense from a lateral G force perspective!

Axial Accelerometer data
Image

Lateral Accelerometer data
Image

Secondly, the obvious failure of the sustainer motor not igniting needs to be resolved. I've had a look at the top grain on the J510, there was zero burning/charring on the grain. I can only assume that the ignitor never fired. When the booster was recovered the ignitor had been cut by the ISC and the booster's Raven confirms that the channel had continuity for 1.7 seconds. My suspicion is that my programming was incorrect on the Raven 3 and the ignitor was never lit. My takeaway from this is that I need to become more knowledgeable about Raven 3 staging configurations.


Moving forward

I plan on attempting another flight on this vehicle stack at our second Williams Launch in September. Many aspects of the flight were incredibly successful and lucky. The ISC performed well under the 30+ G's it sustained during liftoff without bucking under the sustainer weight. Booster drag separation was clean, and realistically I had a highly inefficient boosted dart flight. Separation would have occurred when the vehicle was marginally supersonic and yet it handled that fine. As for the main failure of the sustainer motor not igniting it's obvious I need to spend more time with wrapping my head around Raven 3 Custom configurations for pyro events. Any advice or relevant reading on this topic would be much appreciated. Furthermore I suspect I would be well served by constructing an ignitor that's a little more grunty than the standard AeroTech ignitors. Possibly a combination of ematch+AP+BP. Any advice on this would also be appreciated. Finally I'm tossing around the idea of manually fixing the ignitor with a dab of epoxy or Gorilla glue to the top grain of the sustainer to ensure that when the sustainer ignitor fires I can ensure it's in contact with the top fuel grain. As before, any discussion regarding this would be appreciated.
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015

martymonsta
Engineer
Engineer
Posts: 709
Joined: Mon Oct 13, 2014 11:57 am
Location: Drouin Vic.

Re: SpeedStack

Postby martymonsta » Tue Jun 02, 2015 11:31 pm

Hey Drew. Great looking project and that looked great off the pad. Just a question about staging and the raven3 configuration. There looks to be some strange goings-on with the baro readings. During the assent it looks like there was something that caused the pressure to remain high (equivalent to low altitude) during boost and then drop rapidly at motor burnout and then normalized to the flight profile.

Is there any chance that the static ports were blocked by a chute or something else under the acceleration of liftoff and this cleared with deacceleration at motor burn out? It could provide a reason for staging failure if the port was blocked and the raven sensed that the rocket was under 96ft AGL @ L+0.5 sec.
Many years ago the great British explorer George Mallory, who was to die on Mount Everest, was asked why did he want to climb it. He said, "Because it is there."
Well, space is there, and we're going to climb it

MARS #21
AMRS #41 L2/LCO/RSO/CO

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1889
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: SpeedStack

Postby SpaceManMat » Wed Jun 03, 2015 9:19 am

martymonsta wrote:Hey Drew. Great looking project and that looked great off the pad. Just a question about staging and the raven3 configuration. There looks to be some strange goings-on with the baro readings. During the assent it looks like there was something that caused the pressure to remain high (equivalent to low altitude) during boost and then drop rapidly at motor burnout and then normalized to the flight profile.

Is there any chance that the static ports were blocked by a chute or something else under the acceleration of liftoff and this cleared with deacceleration at motor burn out? It could provide a reason for staging failure if the port was blocked and the raven sensed that the rocket was under 96ft AGL @ L+0.5 sec.


I think that is mach transition you're looking at, except data from the Telemega says it was not that fast. Perhaps from staging of some other event? Drew can you confirm how fast it was simmed and when that peek is?

Never mind, read it again. Mach 1.2 the slower figure is from the GPS. It looks like a spike just before burnout followed by a dip as it drops out of Mach. The Ravens have a lock out so would not be confused by that.
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1889
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: SpeedStack

Postby SpaceManMat » Wed Jun 03, 2015 9:40 am

Any chance that the sustainer separated prior to ignition and cut the wire preventing ignition?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

Lamp
Aged like a good scotch
Aged like a good scotch
Posts: 1098
Joined: Fri May 08, 2009 9:28 am
Location: Melbourne

Re: SpeedStack

Postby Lamp » Wed Jun 03, 2015 10:59 am

Interesting data there Drew, any chance of expanding the time scale so I can look at the first 5 seconds expanded out?
From the look of it in an unexpanded scale, your accelerometer data detected liftoff but you altitude data does not look to have really recorded that. It is possible that the sustainer ignitions were not met by the Raven 3...your recorded altitude at 0.5 seconds looks to be below your programmed minimum requirement.
‘It takes sixty-five thousand errors before you are qualified to make a rocket.’ — Werhner von Braun
TRA 12286 L3
TAP member

Lamp
Aged like a good scotch
Aged like a good scotch
Posts: 1098
Joined: Fri May 08, 2009 9:28 am
Location: Melbourne

Re: SpeedStack

Postby Lamp » Wed Jun 03, 2015 11:10 am

Also, there is something else screwy with the Raven data. It shows the sustainer DECREASING in altitude before separation. This however could be the baro data being unreliable as it drops through the Mach transition back to subsonic speed.
‘It takes sixty-five thousand errors before you are qualified to make a rocket.’ — Werhner von Braun
TRA 12286 L3
TAP member

Lamp
Aged like a good scotch
Aged like a good scotch
Posts: 1098
Joined: Fri May 08, 2009 9:28 am
Location: Melbourne

Re: SpeedStack

Postby Lamp » Wed Jun 03, 2015 11:27 am

PS, I would never ignite the sustainer from the booster avionics (if I have read correctly that is what you have done) because there is just too much that can go wrong. Also, I would personally use a CTI motor in the sustainer because they come up to pressure so quickly, unless you were counting on the delay it takes for an Aerotech to pressurise. Finally, I would definitely beef up the igniter if you are using an Aerotech!
‘It takes sixty-five thousand errors before you are qualified to make a rocket.’ — Werhner von Braun
TRA 12286 L3
TAP member

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

Re: SpeedStack

Postby drew » Wed Jun 03, 2015 4:28 pm

Thanks for all the responses guys. I really appreciate it. Also, good spot on the dip in the booster's baro reading. That said it coincides with the stage separation, not Mach transition. FYI all the avionics of the booster are housed in the ISC so something covering the baro port isn't really a possibility unless something dislodged in the ISC's AV bay (which it didn't).

Here's a better graph from the FIP that displays baro altitude, velocity, and the 3rd pyro channel's continuity. As you can see, top velocity was reached and already declining before the baro dip. My baro port was some length down on the coupler and I've since noticed that it's located nearly directly inline with one of the three ISC motor breathing vents there to ensure there's adequate airflow for the sustainer motor to light. :? Next time I'll ensure to locate these vents as far away from the baro port as possible.

Image

Lamp wrote:From the look of it in an unexpanded scale, your accelerometer data detected liftoff but you altitude data does not look to have really recorded that. It is possible that the sustainer ignitions were not met by the Raven 3...your recorded altitude at 0.5 seconds looks to be below your programmed minimum requirement.


Lamp, you're bang on. Despite my simulation stating I'd be well over 150ft at 0.5 seconds into the flight the actual baro data shows I was at ~50ft at 0.5 seconds into the flight. Given my custom configuration on the 3rd channel was for 0.5 seconds and 96 ft and the vehicle was at only 50 ft I can only assume that since the parameters weren't met at that time the computer's logic determined to not open the channel as by 0.85 sec into the flight I had surpassed the 96 ft threshold leaving ~1 second between that point and loss of continuity on the 3rd pyro channel.

Lamp wrote:PS, I would never ignite the sustainer from the booster avionics (if I have read correctly that is what you have done) because there is just too much that can go wrong.

I think it depends on the flight profile you're attempting mate. I don't really see any difference from lighting via the booster instead of the sustainer in this use case as I'm looking for speed and actively NOT looking for altitude to stay within waiver. Both flight computers will see the same flight profile until stage separation occurs so whether the ignition is done by the booster or sustainer flight computer is somewhat immaterial in this case as the parameters would be identical. Further lighting from the sustainer on a minimum diameter is pretty complicated from a AV Bay location and wiring perspective.

Lamp wrote:Also, I would personally use a CTI motor in the sustainer because they come up to pressure so quickly, unless you were counting on the delay it takes for an Aerotech to pressurise.

Agreed in theory, in practice for 38mm sustainers the CTI product line doesn't appeal as much to me as they don't have separate aft closures that suit coupling via the motor casing. I've got one of the RouseTech 38mm aft closures that's been turned down to remove the knurling and is the same diameter as the casing. This provides a good coupling point in my opinion. In 29mm using CTI is a no brainer as they have an aft tailcone available that's the same diameter as the motor casing. But in 38mm AeroTech is better positioned given the aft closure and availability of the J510 (which coincidentally is a higher Ns motor than the J530IM).

Lamp wrote: Finally, I would definitely beef up the igniter if you are using an Aerotech!

Agreed. Any advice on that front?
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

Re: SpeedStack

Postby drew » Wed Jun 03, 2015 7:03 pm

After thinking a bit more about this data a more likely possibility is that the baro dip could indicate my ISC AV bay isn't sealed appropriately. If so it's possible that after separation the initial rush of air at transonic speeds into the coupling point caused a pressure increase in the AV bay and made the baro sensor think it was well below AGL (-5967ft). Looks like I'm going to need to beef up the AV bay seal inside the rest of the ISC, especially the port where the sustainer ignitor exits the Av bay.
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015

Lamp
Aged like a good scotch
Aged like a good scotch
Posts: 1098
Joined: Fri May 08, 2009 9:28 am
Location: Melbourne

Re: SpeedStack

Postby Lamp » Thu Jun 04, 2015 10:53 pm

All good Drew, I knew you would have a valid reason.

You could always try head end upper stage ignition, Scoop has the "scoop" on that....

A sliver of Blue Thunder should augment your igniter well but seek out Scoop for advice, he might have an even better solution.

I think you might be right on the Av bay leakage through the bulkhead. The booster of my 2 stage at Thunda deployed early, shortly after separation. The data showed unstable altitude readings just before the pyro fired, I decided this was leakage past the forward bulkhead which was in the airflow after staging, making the altimeter think apogee had been reached.

I think you have pretty much covered everything that was sub-optimal with the first flight and I look forward to hearing about the results of flight No.2...and seeing the data. I expect it will all go very well!
‘It takes sixty-five thousand errors before you are qualified to make a rocket.’ — Werhner von Braun
TRA 12286 L3
TAP member

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

Re: SpeedStack

Postby drew » Tue Sep 29, 2015 1:00 pm

Well, I tried this flight again on the weekend. The only difference is that I used a K1103X for the booster instead of a K1100T. Good news is that the sustainer lit this time. Bad news is that the lighting of the sustainer fried the booster's electronics. Combined with the TeleGPS in the sustainer's nose stopping transmitting as soon as the rocket left the pad I didn't find the booster or sustainer. I went back to our Williams launch site yesterday and searched for a number of hours with no success. SPEWIN.

Image

There's a shot of the launch. I've got some video as well but I haven't edited it yet. I'll get to that soon enough. Until then, some lessons.

1) I'm cured of the "bang, bang" staging for speed attempts affliction I've suffered from over the past couple years where I light the sustainer with booster electronics during boost from the interstage coupler in a futile attempt to reach ludicrous speeds. I don't believe this is easily possible with the flight computer in the ISC without roasting the electronics. YMMV, but for me I'm done.

2) It was apparent that the booster went unstable from the sustainer igniton and lost a fin as you could see the tell-tale "corkscrew" from the K1103's smoke grain. Even IF I was able to effectively seal the ISC's electronics from the sustainer ignition I don't believe I'd be able to mitigate the "kick" effects of sustainer ignition on the booster's flight path.

3) Use redundancy for everything. I've never had issue with my TeleGPS until yesterday. That said because it was the only RF device in my Sustainer and my flight profile easily simmed to 20k+ ft I was screwed as soon as it packed it in. In future I plan to use redundant transmitters (minimum 1 GPS + 1 RDF transmitter) and redundant altimeters.

4) Stop fiddling with little sh*t. I don't know about y'all but prepping a 38mm to 54mm two stage minimum diameter stack is not something that's simple or easy. Scaling up will actually decrease the complexity of launch prep.

Next two stage attempt I do will be a 75mm booster to 54mm sustainer. The sustainer will have two Ravens, one to sit on top of the motor to handle sustainer ignition and apogee deploy, the second will be in the nose for apogee and main deploy. I plan on coasting post boost and relying on drag sep to ensure the booster is clear from the sustainer before that motor ignites.

Positives to take away from the launch.

1) I'm cured of my previous staging affliction.
2) I feel more confident with Raven programming for staging.
3) Our season here in WA is nearly at a close so I have some time to absorb this before moving forward.

More soon.
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015

User avatar
SpaceManMat
Astronaut
Astronaut
Posts: 1889
Joined: Mon Dec 15, 2008 9:56 pm
Location: Brisbane

Re: SpeedStack

Postby SpaceManMat » Wed Sep 30, 2015 5:45 pm

Out of curiosity Drew, how many batteries, what type and how were they wired? Also which channel on the raven did you do the air start with?
QRS: 124
AMRS: 32 L2 RSO
Highest Altitude: 10,849 feet
Largest Motor: CTI 1115J530 IM
Current Project: X Wing

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

Re: SpeedStack

Postby drew » Fri Oct 02, 2015 11:43 am

SpaceManMat wrote:Out of curiosity Drew, how many batteries, what type and how were they wired? Also which channel on the raven did you do the air start with?


Booster had a Raven with a 130mAh battery which was turned on by a Featherweight magnetic switch. It also had a 100mW Big Red Bee RDF transmitter running off a 200mAh battery. Sustainer was lit from the 3rd channel on the Raven in the booster.

Sustainer had a Raven with a 130mAh battery and was turned on by a Featherweight screw switch. It also had a TeleGPS with a 400mAh battery.

Transmitter and GPS were activated on the bench, altimeters were enabled on the pad.
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015

User avatar
CATO
Mr Smiley
Mr Smiley
Posts: 1863
Joined: Thu Dec 24, 2009 6:20 pm
Location: Sydney, NSW
Contact:

Re: SpeedStack

Postby CATO » Fri Oct 02, 2015 12:39 pm

Nicely done Drew...
"In thrust we trust"

AMRS 21 L3

Impulse:
2017: 5,973 Ns (13.14% M)
Ns 16: 34,558; 15: 35,955; 14: 6,016; 13: 10,208
PB - Gorilla N2717WC, H: 10,260', S: 1.14M

drew
Southern fried goodness
Southern fried goodness
Posts: 714
Joined: Sat Aug 01, 2009 10:20 pm
Location: Perth

Re: SpeedStack

Postby drew » Fri Oct 02, 2015 4:04 pm

CATO wrote:Nicely done Drew...


Thanks mate. Fingers crossed when I attempt a staging shot again I'll have better results... 8)
Andrew Hamilton
AMRS 28 L3
AMRS Records Committee Chairman
Max Alt AGL - 23,908ft - K300 - Balls 22
Max V - 2,488 ft/s, ~Mach 2.2 - M2250 - THUNDA 2015


Return to “Scratch Built”

Who is online

Users browsing this forum: No registered users and 6 guests