Premature Drogue Chute Deployment Diagnosis

[Nebulous1515]

Hello rocketeers!
This is my first post on here so forgive me for my amateur-level knowledge of terminology and the like. That being said let's get right into it.

I have made (my first) two-stage rocket out of a modified Madcow Adventurer. It is made entirely out of fiberglass and Aeropoxy. It has electronic separation, ejection and airstart. It as two AV bays one for each stage. The first stage has a single Raven 3 for separation and chute deployment. The sustainer has a Blue Raven for primary drogue and main chute deployment and airstart. It also has an Eggtimer quark for redundant main and drogue chute deployment. I have attached a picture of the OR model and the OR file itself. One detail that may or may not be important that are not reflected in the model: it has a Runcam2 mounted on the outside of the airframe at the center of pressure of the sustainer. The enclosure is linked here: https://www.additiveaerospace.com/products/runcam2-video-camera-shroud.

It is meant to fly to about 12,000 ft and reach a top speed of Mach 1.3. I think going supersonic may have a role to play in why things went the way they did. I used a Cesaroni I-1299 for the first stage and a J-530 for the sustainer.

I got to launch it twice in the same day. The first time, the airstart failed due to my forgetting to arm the channel on the Blue Raven app (the custom configuration I used to set the conditions for airstart had automatic arming upon power-up set to off :/). It's worth noting that the sustainer worked flawlessly here, with the drogue chute deploying at apogee and the main chute deploying about 500ft from the ground. First stage separation was perfect and the chute deployed a little late but still worked just fine.

The second time around, I made sure to arm the airstart channel and it worked! Almost immediately after 2nd stage burnout, the drogue chute appeared to deploy sending the rocket tumbling. The immense drag on the drogue chute did three things: it zippered the aft section making it unusable (I plan to make a new aft section), it ripped the drogue chute completely off, and it provided enough force to break the shear pins on the main chute and deployed it. The main chute ripped off immediately as well. Thankfully all of the sections were apart so the drag was high enough to prevent the complete destruction of the rocket. Somehow the aft section was the only part that was no longer flightworthy. I attached the video of the ground POV and the Runcam POV along with a picture of the zippering below for more context.

Here's what I think happened and this is what I need help trying to solve:

I've been looking through the Blue Raven flight details and the Runcam footage here's what I found.
Watching the footage:

• Liftoff occurs at 0s
• First burnout occurs at 0.5s
• Stage separation occurs at 2 seconds
• Aistart occurs at 4.1 seconds
• Burnout occurs at about 6 seconds
• At 7.4 seconds, the flight becomes completely unstable, you can see the kevlar chord outside of the rocket.
• At 7.43 seconds, you can see what I believe to be the white strings from the ripped drogue chute near the bottom of the rocket.
• At 7.7 seconds we can see the avionics and forward section without the nose cone, confirming that both the drogue and main chute must have deployed at this point.
• At 8.3 seconds, we can see a red dot below the rocket which I believe to be the main chute ripped off.
• At 11.4 seconds is the first time we see the nosecone.
• We saw both the primary and redundant charges go off at 600 and 500ft above the ground so the Quark set off the main chute charge at the correct time.

Looking at the Blue Raven flight data:

• First burnout occurs at 0.5 sec (confirmed by video)
• Third channel (airstart) fires at 4.1 seconds (confirmed by video)
• Gyro overload occurs at 7.5 seconds (matches with instability seen in video)
• Apogee channel fires at 13.7 seconds (long after both chutes deploy)
• Main channel fires at 65.3 seconds (confirmed by witnesses on ground seeing black powder).

All of this information leads me to believe that the drogue chute was deployed too early by the Eggtimer Quark because of the mach transition. It is clear that the Blue Raven was not responsible for the seperation and the Quark doesn't save the times of deployment (I don't think) so we can only speculate. Although it reads that the "max velocity" was only 948ft/sec, looking at the "Velocity_Up" channel, it peaked at about 1200ft/sec, just over Mach 1. The Quark manual mentions that it takes the increase of pressure and decrease of apparent velocity into account with a prevention system that checks if the " noise-filtered velocity drops below 100 ft/sec for at least 1 second" and re-enables deployments. I also attached the paragraph talking about mach transition in the Eggtimer Quark manual below

My best guess is that because of the two-stage nature of the rocket, and the fact that velocity was in fact decreasing before airstart (all the way to a minimum of 150ft/s according to the Blue Raven), the deployments were re-enabled shortly before airstart and did not turn off again. This caused the mach transition to falsely trigger drogue chute deployment, which caused both the aft section zippering and provided enough force for main chute deployment. Both parachutes were torn from the kevlar.

If this is the issue I can think of two solutions:
1. Just removing the redundant system altogether
2. Airstarting slightly earlier to prevent the velocity from dropping too far and enabling the chutes deployment


Another possible cause that I haven't given too much thought is the the chutes being deployed some other way. Here are a few theories.
1. The drogue chutes were packed quite tightly and secured with two #2 shear screws. Would the deceleration due to burnout be enough to seperate it?
2. The top section was somehow able to wobble enough to catch enough air to bend over and seperate, causing the zipper-like crack and deploy both chutes (I'm skeptical).

I hope I covered everything sufficiently. Please let me know about any confusion

I plan to launch again at ROCStock this weekend so I would very much appreciate quick responses (I know its a lot to digest and consider). Thank you!

 

[cwbullet]

The first post was approved. Welcome to the forum. Sorry for the adverse results of your flight.

 

[Neutronium95]

I'd agree that the Quark is the most likely culprit. I've heard of a few other stories about premature deployments with Eggtimer altimeters. Being able to review all of the flight data is incredibly useful when trying to figure out why something failed, the lack of this capability has put me off of super simple dual deploy altimeters like the Quark and RRC2 that don't save all of the flight data for later analysis.

Hope you have better luck at Rocstock. I hope I'll be able to catch your flight.

 

[rfjustin]

Hello rocketeers!
This is my first post on here so forgive me for my amateur-level knowledge of terminology and the like. That being said let's get right into it.

I have made (my first) two-stage rocket out of a modified Madcow Adventurer. It is made entirely out of fiberglass and Aeropoxy. It has electronic separation, ejection and airstart. It as two AV bays one for each stage. The first stage has a single Raven 3 for separation and chute deployment. The sustainer has a Blue Raven for primary drogue and main chute deployment and airstart. It also has an Eggtimer quark for redundant main and drogue chute deployment. I have attached a picture of the OR model and the OR file itself. One detail that may or may not be important that are not reflected in the model: it has a Runcam2 mounted on the outside of the airframe at the center of pressure of the sustainer. The enclosure is linked here: https://www.additiveaerospace.com/products/runcam2-video-camera-shroud.

It is meant to fly to about 12,000 ft and reach a top speed of Mach 1.3. I think going supersonic may have a role to play in why things went the way they did. I used a Cesaroni I-1299 for the first stage and a J-530 for the sustainer.

I got to launch it twice in the same day. The first time, the airstart failed due to my forgetting to arm the channel on the Blue Raven app (the custom configuration I used to set the conditions for airstart had automatic arming upon power-up set to off :/). It's worth noting that the sustainer worked flawlessly here, with the drogue chute deploying at apogee and the main chute deploying about 500ft from the ground. First stage separation was perfect and the chute deployed a little late but still worked just fine.

The second time around, I made sure to arm the airstart channel and it worked! Almost immediately after 2nd stage burnout, the drogue chute appeared to deploy sending the rocket tumbling. The immense drag on the drogue chute did three things: it zippered the aft section making it unusable (I plan to make a new aft section), it ripped the drogue chute completely off, and it provided enough force to break the shear pins on the main chute and deployed it. The main chute ripped off immediately as well. Thankfully all of the sections were apart so the drag was high enough to prevent the complete destruction of the rocket. Somehow the aft section was the only part that was no longer flightworthy. I attached the video of the ground POV and the Runcam POV along with a picture of the zippering below for more context.

Here's what I think happened and this is what I need help trying to solve:

I've been looking through the Blue Raven flight details and the Runcam footage here's what I found.
Watching the footage:

• Liftoff occurs at 0s
• First burnout occurs at 0.5s
• Stage separation occurs at 2 seconds
• Aistart occurs at 4.1 seconds
• Burnout occurs at about 6 seconds
• At 7.4 seconds, the flight becomes completely unstable, you can see the kevlar chord outside of the rocket.
• At 7.43 seconds, you can see what I believe to be the white strings from the ripped drogue chute near the bottom of the rocket.
• At 7.7 seconds we can see the avionics and forward section without the nose cone, confirming that both the drogue and main chute must have deployed at this point.
• At 8.3 seconds, we can see a red dot below the rocket which I believe to be the main chute ripped off.
• At 11.4 seconds is the first time we see the nosecone.
• We saw both the primary and redundant charges go off at 600 and 500ft above the ground so the Quark set off the main chute charge at the correct time.

Looking at the Blue Raven flight data:

• First burnout occurs at 0.5 sec (confirmed by video)
• Third channel (airstart) fires at 4.1 seconds (confirmed by video)
• Gyro overload occurs at 7.5 seconds (matches with instability seen in video)
• Apogee channel fires at 13.7 seconds (long after both chutes deploy)
• Main channel fires at 65.3 seconds (confirmed by witnesses on ground seeing black powder).

All of this information leads me to believe that the drogue chute was deployed too early by the Eggtimer Quark because of the mach transition. It is clear that the Blue Raven was not responsible for the seperation and the Quark doesn't save the times of deployment (I don't think) so we can only speculate. Although it reads that the "max velocity" was only 948ft/sec, looking at the "Velocity_Up" channel, it peaked at about 1200ft/sec, just over Mach 1. The Quark manual mentions that it takes the increase of pressure and decrease of apparent velocity into account with a prevention system that checks if the " noise-filtered velocity drops below 100 ft/sec for at least 1 second" and re-enables deployments. I also attached the paragraph talking about mach transition in the Eggtimer Quark manual below

My best guess is that because of the two-stage nature of the rocket, and the fact that velocity was in fact decreasing before airstart (all the way to a minimum of 150ft/s according to the Blue Raven), the deployments were re-enabled shortly before airstart and did not turn off again. This caused the mach transition to falsely trigger drogue chute deployment, which caused both the aft section zippering and provided enough force for main chute deployment. Both parachutes were torn from the kevlar.

If this is the issue I can think of two solutions:
1. Just removing the redundant system altogether
2. Airstarting slightly earlier to prevent the velocity from dropping too far and enabling the chutes deployment


Another possible cause that I haven't given too much thought is the the chutes being deployed some other way. Here are a few theories.
1. The drogue chutes were packed quite tightly and secured with two #2 shear screws. Would the deceleration due to burnout be enough to seperate it?
2. The top section was somehow able to wobble enough to catch enough air to bend over and seperate, causing the zipper-like crack and deploy both chutes (I'm skeptical).

I hope I covered everything sufficiently. Please let me know about any confusion

I plan to launch again at ROCStock this weekend so I would very much appreciate quick responses (I know its a lot to digest and consider). Thank you!

Did you have a pressure relief hole in the bottom of the sustainer? If so what size and where was it located?

 

[Nebulous1515]

Did you have a pressure relief hole in the bottom of the sustainer? If so what size and where was it located?

I did not have pressure relief holes in the sustainer. I doubt the fit between the aft section and AV bay was airtight. I also doubt this was enough to break the shear screws the but you have a great point and there’s no reason not to implement that.
Thank you!

 

[waltr]

Pressure relief is also dependent on how fast the rocket goes up. If the external air pressure drops faster that the bay can 'leak' then it could push apart the sections.

 

[Nebulous1515]

Pressure relief is also dependent on how fast the rocket goes up. If the external air pressure drops faster that the bay can 'leak' then it could push apart the sections.

This makes a lot of sense. The combination of the tight packing in addition to the quick acceleration to Mach 1 could make that case likely.

 

[cwbullet]

I did not have pressure relief holes in the sustainer. I doubt the fit between the aft section and AV bay was airtight. I also doubt this was enough to break the shear screws the but you have a great point and there’s no reason not to implement that.
Thank you!

Pressure causing the separation definitely possible then. A lot of flyers Lear that the hard way. I try to catch when I RSO the bird.