Why did my main deploy at apogee?

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

s201

Member
Joined
Apr 30, 2016
Messages
14
Reaction score
0
I flew my first dual deploy rocket for the first time a few months ago. I lost sight of it on the way down, but upon reviewing the altitude data from the on-board log I saw that it showed a steady descent from apogee instead of a fast descent followed by a slow descent which would be consistent with a low altitude deployment of the main. Thus, I concluded that the main deployed at apogee, but I'm unsure as to why, or more importantly: how to prevent it from happening again.

Here's my setup: 2.6" diameter, 54mm motor, forward main, aft drogue. I had custom electronics controlling the ejection charges (the log showed that the charges were fired at the correct time) so as a simple insurance plan I left the motor ejection charge in. FWIW, in the ground tests that I performed beforehand the drogue deployment worked as intended and did not also deploy the main. These are the possible scenarios that I can come up with:


  • The motor ejection charge fired before the electronics controlled drogue charge causing the main to deploy with the drogue.
  • I put too much black powder in the drogue charge resulting in the main also deploying.
  • The drogue charge fired after the motor charge had already separated the rocket causing the main to deploy at this time.
  • I simply got unlucky.

I'm really just grasping at straws here though. My question for those more experienced with dual deployment than myself is: Are any of these scenarios likely? What is the usual cause of main deployments at apogee and how can they be prevented?

Any advice is greatly appreciated. Thank you!
 
Did you retain the nosecone with shear pins or was it friction fit?
 
Important facts omitted:
1) were shear pins used?
2) length of harness in apogee deployed section?
3) vent/bleed hole in airframe sections?
4) weight of upper section vs. drag of lower airframe?
5) did you wire your charges backward?

I got unlucky is really not an available option. Electronic deployment is pretty precise once you master it. Sheer pins and long harnesses are your allies.

--Lance.
 
What was the length of the boosters harness ???

Teddy
 
Thanks for all of the suggestions everyone. Answers to the various questions:


  • Were shear pins used?
    • Yes, three shear pins each on the upper and lower airframes. Three rivets for the e-bay.
  • How much blackpowder was used?
    • A hair over 1g
  • Shock cord length?
    • 15ft, all kevlar
  • ​Vent hole in the airframe sections?
    • Tthe e-bay has two vent holes. None in the upper or lower airframes though.​
  • Weight of upper section vs. drag of lower airframe?
    • ​I'm not sure what this means. Can you elaborate?
  • Charges wired backward?
    • ​It's certainly a possibility. I have a wiring diagram that I checked before along with doing a continuity check so I really hope this wasn't the case though.
  • ​What electronics?
    • ​All custom designed by me.

It seems like the general
consensus is that I may have used too much BP. I'll do a ground test or two to make sure I'm using the correct amount.
 
Seems like your shear pins, harnesses, and ground testing all check out. You never got a good visual of what happened, so maybe it is the data itself.

How big is your drogue chute and main chute? Perhaps the drogue is too large and of similar descent rate as the main. Perhaps your custom data logger does not capture any noticeable changes in rate. Can you post the altitude vs. time plot?
 
Have the electronics flown successfully prior to your launch? If so, how many times?

Greg
 
3 shear pins in the lower airframe for drogue deployment? Likely amount of BP used to shear those 3 pins, coupled with a harness of insufficient length to allow the 2 parts of the airframe to slow down before reaching terminal length, caused sufficient force with the sudden stop to throw the main/recovery harness bundle against the bulkhead of the nose cone with sufficient force to shear the upper pins and shake out the main.

Please post the length of the lower and upper airframe sections that you're pressurizing so that the numbers can be run to help determine if your charges were in the proper range. Measure from the top of the top most centering ring to the edge of the airframe, and from the top of the avionics bay to the top of the airframe, and subtract avionics bay or nose cone shoulder as appropriate.
 
Do you have an OpenRocket file? That would help tremendously for diagnosis. Also, was the flight a normal straight up powered ascent phase and not overly horizontal? We'd need to know the force of deceleration when the drogue deployed. That helps determine the force applied to the upper section holding in your main.

Assuming this is a standard looking L1 airframe, I'm guessing a rocket <7 lbs fully loaded made of 2.6" glassine coated tubing. None of that is very heavy itself. If the drogue deployed at apogee with little horizontal motion, the deceleration force on the upper airframe wouldn't be much. 3 shear pins (assuming nylon 2-56 here) will hold about 60 lbs. The deceleration force wouldn't be anywhere near that on an electronics bay + upper frame weighing only a couple pounds.

My bet is on the electronics side. Wired drogue and main reversed. Wired it so all charges fired. Wiring ok, but homemade electronics were faulty. Something like that.
 
Last edited:
For your bullet #4, there is no such thing as "unlucky" in physics. :wink: Something caused it; the question is what.

Even though you used shear pins, they may not have been able to retain the NC after the apogee deployment. The amount of shear strength of the pins needs to exceed the amount of force being applied to the forward half of the rocket at apogee deployment. IOW, if your payload/NC half received 150 pounds of force, and your number/size of shear pins were only capable of retaining 120 pounds of force, then out comes the NC.

We would need to know the weight of the forward half, the shear pin material (6/6 nylon screws?), size of pins (#2-56?), and amount of acceleration at apogee deployment (from altimeter download) to do further analysis.
 
If shear pins are loose, they shear more easily. Make sure they are a tight fit in the holes.
 
2.6" Airframe,,, 54 mm motor mount...
How heavy is the nose cone ?? is there tracking equipment in it ??
My vote...
15 Ft is not long enough for the boosters harness..
The payload section maybe,, but not the booster's..
The 1 gm charge isn't overly aggressive,,
but with such a short harness you'd have to be exactly spot on with that charge,,
zero room for the slightest error..
At the apogee event the forward section reached the end of the boosters harness and dislodged the nose cone...

You can prove me wrong pretty easily...
Just fly it again with everything exactly the same but with a 20 ft booster harness...

Teddy
 
I'd do what Teddy says and also add an 1/8" vent hole to both booster & payload section.
 
... IOW, if your payload/NC half received 150 pounds of force, and your number/size of shear pins were only capable of retaining 120 pounds of force, then out comes the NC.

You have to take out the force that was absorbed and used to shear the pins on the lower half. The remaining force will accelerate the upper part. This then gets slowed down by the shock cord. When all that's done, the remaining force will jerk the nosecone along with the drogue also pulling as it opens. That's the leftover deceleration times the mass that is the force that needs to be held back by the shear pins. It's really not that much if you deploy at apogee on a relatively straight up flight. I use 2 x 2-56 nylon pins to hold in a 1lb electronics bay plus about another 2.5 lbs of frame, nosecone, main, cord after drogue deployment, and it's fine.
 
I'd do what Teddy says and also add an 1/8" vent hole to both booster & payload section.

Thank you for sure for the vote of confidence Tim...

If you make more then one change you won't know which one was the problem..
These two changes will both effect the exact same thing,,
the amount of kinetic energy the same charge put into the two separating parts of the rocket...

One change at a time,, keep every other parameter as close to identical as possible...

Teddy
 
My vote(s):
- Insufficient main-bay retention to survive the apogee deployment forces.
- Early motor eject causing significant stress on the NC due to high velocity deployment of the drogue.
- mixed up wiring....see this often when there is a rats nest of monochrome wire - don't know if this is the case, but it is typical.
 
Color-coded wiring helps a lot, and so does marking both sides of the sled and the AV bay so it's more difficult to install upside-down. Better yet, use removable connectors between your sled and the AV bay lids, and use different genders on each lid; you can't reverse them. Of course, you can still install the AV bay upside-down...
 
I have had my main deploy at apogee twice. Once because I was not paying attention and inserted my AV bay in backwards, getting old. Luckily this was only a 3,300 AGL flight so a relatively short walk. To eliminate this going forward I just keyed my AV bay so it can only go in one way.

The other apogee deploy was a bit more of a mystery to solve and your experience reminds me of it. The second apogee main deploy was @ 16,000' AGL and resulted in a 3.45 mile walk, but I need the exercise. After I recovered the rocket and went over the data from both altimeters and GPS, I could clearly see that, like you concluded, it was a nice peaceful slow steady decent. In my case one of the altimeter's log filled up and it stopped recording somewhere around 5,000, but the other altimeter showed no difference, beyond the typical gradual slow decent due to atmospheric changes, but nothing to indicate a main @ 700AGL. Once I completed my analysis I concluded that my apogee charge was too energetic, which caused my nosecone and payload bay to come to an abrupt halt at the end of my 20 Kevlar harness, and the mass and velocity was enough to shear my nosecone pins and release my main.

To mitigate this going forward I ended up getting a 25' shock cord from Teddy, mainly because the extra 5' could fit and I figured it would give me a bit more margin, and I also performed more ground testing to dial in the apogee charge.

Good luck.
 
Answers to questions:


  • Altitude vs time log: This is the full flight report generated from my log file (see also the 3d flight path below the altitude vs. time graph): https://shanet.github.io/Osprey/
  • Length of upper & lower airframes: 12" & 20" respectively
  • Weight of upper airframe + nosecone + e-bay + recovery harness: 1063g (fiberglass airframe)
  • Parachute diameter: 30" main / 14" drogue
  • Shear pin details: These are the exact ones that I used: https://www.mcmaster.com/#91010a019/=14o34uq
  • Electronics flown successfully before: I did numerous test flights on mid-power rockets and on a few on H motors. All of these were motor ejection though. However, the log showed that apogee was detected correctly and the charges were fired at the correct time for each test flight.

Here's what I'm thinking then:


  1. No motor ejection charge to eliminate that as a potential point of failure
  2. Do a ground test to check the force on the upper airframe during drogue deployment
  3. Depending on the outcome of step 2, adjust the BP amount and/or add another shear pin to the upper airframe to prevent premature main deployment.

Does that sound reasonable? Unfortunately, I'm planning on flying it this weekend so I don't have time to get a longer shock cord. I didn't think I'd be able to get my hands on a motor in time, but I did so now I'm behind a bit in addressing this issue.

Thanks again everyone! This is incredibly helpful.
 
I having difficulty zooming in on the graph but it looks like an early deployment to me. Your graph abruptly changes from going up to going down. There should be a nice smooth curve at the top for a few seconds.
 
Are you using an acelloromter to calculate altitude? Also is the sample rate on the graph 1/10 sec?
 
Perhaps the lids were reversed, or the wiring to them - the charges fired but the main came out at apogee .

Kenny
 
Back
Top