L3 Post Mortem - Flight of the Terminator

eggtimer rocketry
shop deals on amazon
The Rocketry Forum

Help Support The Rocketry Forum:

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

Worsaer

Amateur Propulsionist
TRF Supporter
Joined
Jan 11, 2012
Messages
2,083
Reaction score
510
Location
Central Virginia
I finally flew my L3 project this past Saturday at Tripoli Central California. I started the project almost a year ago, and it was ready to fly in May, but the flight was scrubbed due to unforeseen issues with my electronics. Thanks to John Derimiggio for troubleshooting and tweaks to some firmware, I was ready to fly this weekend.

This is a post-mortem because it didn't go well, but as always, I learned something.

First of all, here's some details on the rocket:

Rocketry Warehouse Terminator 5, flying on a CTI M1540 (5 grain IMAX).
Fully redundant Marsa54L's, using Marsa gadgets (MrfBridge, and MrfPyros)
Drogue in forward payload pay; Main in nose cone
Raven 3 as tertiary altimeter
Real Flight Systems GPS system

The up was beautiful. Real-time GPS tracking allowed me to watch it race to apogee, even though the rocket was out of sight. Unfortunately, however, the same real-time GPS showed it losing altitude VERY quickly - it was obvious the drogue didn't deploy.

Here it is at takeoff (Thank you Benno!!!)

29736395053_8fda919db1_o%20-%20Copy_zpshzjisibv.jpg




And Google Earth view of the flight, up and down, using data from the GPS1:

10-15-16_earth_view_zpsdwiirlrm.jpg



The main deployed on cue at 1,000', but the rocket was descending at about 550 fps, instantly shredding the high quality Fruity Chute, but slowing the descent just enough allowing the electronics to survive.

The SIM predicted 12,075'. Here's what was recorded:



Device..... Apogee.....Peak Velocity
Marsa #1.....12,169'.....1060 fps
Marsa #2.....12,183'.....1058 fps
Raven 3.......11,983'.....1317 fps
GPS-1..........12,377'.....1341 fps


10-15-16_zpsujd8edm5.jpg






Stay tuned for more photos, what went wrong, and what I learned.
 
Dang. Hoping to fly my Terminator for the first time in November! I'm sorry for your loss... What were your charges like and with what shear pins if any? Ground tested? Two ematches per charge?
 
FredA said:
It looks like you had a charge fire - too little of a charge???
Click to expand...

What Fred said. Those two shock spikes at apogee and apogee +1 look like drogue charges. Sorry for the loss. At least the up part was beautiful looking, I bet. Luckily, your pricey electronics were spared.

Will you launch the same model again? Or go to a different airframe?
 
Man I hate to hear that, hope your able to learn why the drogue didn't deploy.
 
Sorry to hear about your loss. At least you have telemetry data to look over.
Is anything other than the electronics recoverable?
 
Here's what I know. The nose cone containing the main and an auxiliary AV bay (in the nose cone coupler) was retained using (3) #4 shear pins.
The primary apogee charge was 2 grams, and the backup was 3 grams. Both charges fired, at the right times, at apogee and 1.5 seconds post apogee.

I will post additional photos when back at my PC. I successfully ground tested with 2 grams, and flew exactly as tested. ( I will also post video of the ground test. ). That may not sound like much BP; the volume was 5 inches in diameter, 12 inches in length, containing the drogue and harness. There was actually very little unoccupied space in that section of airframe.

Based on inspection of the forward payload bay containing the drogue and harness, my working hypothesis is the amount of BP was sufficient, but the residue in the payload bay indicates it was not sufficiently contained. I built the AV bay with aluminum charge holders, and packed the BP firmly with dog barf, sealed with tape. (More photos to come.)



Here's a drawing of the project that depicts how the drogue and main are situated in relation to each other.


scale_zpsyktrsn6f.jpg
 
Oh my gosh. I had that kind of ballistic curve from a Parrot with a modified Talon 2. I thought there was a malfunction on the pad and stopped to check the
ematches. I disconnected and reconnected the ematches but since this was before cataract surgery, I missed one leg of the apogee ematch. Landowner found the remains and I had found the main chute and nosecone that was sliced by a fin the day of the flight. Put a new battery in the Parrot 18 months later and surprisingly it showed the ballistic curve. I got 84 G's longitudinal and 45 lateral on impact. Of course that exceeded the limits of the accelerometer but interesting nonetheless. Too bad. Perhaps use larger apogee charges next time.

I take it the Raven was along for the ride and not wired to charges? I believe some folks use a larger charge load in their backups sometimes assuming the first
charge didn't have enough oooomphh to get the laundry out. If it's out, the backup blows harmlessly out the tube.

Two matches per charge warrants bench testing with the deployment electronics to be certain they can handle the lower resistance presented by two parallel ematches although some may argue to wire in a series. Again, sorry about the loss. Kurt
 
Thanks Kurt. Yes, the Raven was pretty much along for the ride, also serving as a tertiary backup for the nose cone main in the event the MrfPyros lost communication. I had a long list of risks, with corresponding mitigants - belt, suspenders, bailing wire and duct tape. Murphy has a way of teaching us lessons just the same.

I won't debate the dual matches per charge thing here, as it wasn't relevant to the failure. My original design did include dual ematches per charge, but after discussing with both TAPS, I removed them. I came to the conclusion that it added unnecessary complexity. Dual ematches mitigates risk of a defective ematch, so I instead managed that risk by testing them in advance.
 
Last edited:
Worsaer said:
Here it is at takeoff (Thank you Benno!!!)

29736395053_8fda919db1_o%20-%20Copy_zpshzjisibv.jpg
Click to expand...

I think that is my shot; Benno copies my cards at the end of launches because I can be so slow to share afterwards.

We barely got into view before takeoff so we shot blind, and mine was overexposed quite a bit. I do have a raw file though and will see if I can adjust for the better.
 
watermelonman said:
I think that is my shot; Benno copies my cards at the end of launches because I can be so slow to share afterwards.

We barely got into view before takeoff so we shot blind, and mine was overexposed quite a bit. I do have a raw file though and will see if I can adjust for the better.
Click to expand...

Then thanks to you also Eric. The sun wasn't at our back, so a lot of photos look like silhouettes. I'm please to get a few shots of the up portion of the flight.
 
Bill, sorry for the mishap.

The Raven data shows a diminished shock and less nosecone pressurization on the second apogee charge even though it was 50% larger. This indicates to me that the nosecone might have been at least partially deployed after the first event. I think losing tape containment of both charge cannisters in a single flight is a low probability, unless the tape adhesion was really bad which you would have noticed.

The fact that you were able to deploy the main supports that the nosecone was either off or nearly off the rocket when the main charges fired.

My hypothesis was that the 3 #4 shear screws plus any additional friction was a little too strong for the charges this time.
 
Thanks John.

I also recognize the possibility that I reversed connection the two drogue charges, firing them in the wrong sequence.

The amount of what appears to be unburned BP in the airframe leads me to believe that the charge fired, spewing unburned BP before it could fully ignite. I will be doing some interesting testing, measuring containment in the same charge wells with different types of tape, and wells of greater length, optimizing for full burn of the BP.

I'm reminded of muzzle loaded rifles designed for BP. Barrel lengths were quite long, giving more time for the BP to burn before being expelled.
 
Worsaer said:
....I will be doing some interesting testing, measuring containment in the same charge wells with different types of tape, and wells of greater length, optimizing for full burn of the BP. ....
Click to expand...

Been there done that. I empathize. I was doing a shakedown flight of my L3 wanna be 4" Broken Arrow fiberglass kit with a large L motor. It went in completely ballistic, never figured out why. I'm sure it was operator error. Never found nose cone or drogue. Did save one altimeter. the fins, the motor mount and the main chute that deployed on impact. Sorta like giving me the finger.

Anyway by the time I rebuilt it with spare parts, and tested deployment I was using 3 1/2 to 4 grams of BP for both drogue & main. Also upon recommendation of TAPS and others I taped my charge wells with electricians tape, 3-4 cross-ways with an additional wrap around the well securing the cross tapes. I certified L3 with my Broken Arrow 2.0 this summer.

Best of luck.
StanO
 
Worsaer said:
The amount of what appears to be unburned BP in the airframe leads me to believe that the charge fired, spewing unburned BP before it could fully ignite.
Click to expand...

I had that same issue when ground testing my MPR. I had BP wrapped in foil with the ematch. It wasn't well enough contained to allow for full burning of the BP. Instead, I got a puff and no airframe separation. Happened about one in five times. I think I posted about it recently again. Now that I switched to putting BP in eppendorf tubes and taping those shut, I've ground tested 10 with 100% success rate. Also used 4 of those charges in 3 MPR flights and my L1 cert flight successfully.
 
I was going to say something about the installation but since you had John on your side I would have been comfortable with that. (ie. John Derimiggio makes the MARSA electronics.)
That charge test above looked nice. A non-stressed deployment but something was up. Perhaps friction fit might have been adequate vs. "weaker" shearpins. I've heard of folks using 2-56 (#2) nylon screws where
I would think 4-40 (#4) would be more appropriate and they have nominal flights. Kurt
 
More photos to come, perhaps tonight. In addition to understanding what went wrong, I will also detail several things that worked well. Sharing information in the spirit of helping others be successful.


For anyone with too much time on their hands, a copy of my L3 project document can be downloaded here:

https://s3-us-west-2.amazonaws.com/12294/WRiley_Terminator_L3-Ver_1.9.pdf
 
StanO said:
Been there done that. I empathize. I was doing a shakedown flight of my L3 wanna be 4" Broken Arrow fiberglass kit with a large L motor. It went in completely ballistic, never figured out why. I'm sure it was operator error. Never found nose cone or drogue. Did save one altimeter. the fins, the motor mount and the main chute that deployed on impact. Sorta like giving me the finger.

Anyway by the time I rebuilt it with spare parts, and tested deployment I was using 3 1/2 to 4 grams of BP for both drogue & main. Also upon recommendation of TAPS and others I taped my charge wells with electricians tape, 3-4 cross-ways with an additional wrap around the well securing the cross tapes. I certified L3 with my Broken Arrow 2.0 this summer.

Best of luck.
StanO
Click to expand...


Thanks Stan.

For round two, I will either do another Terminator ("I'll be back"). That's the stubborn side of me saying I won't fail on a project, it just may take longer than expected. However, my second choice is the 5" version of the Broken Arrow. The upside is that it can also handle 98mm hardware, giving me more flight options over time...

I'm in no hurry - for me designing & building them is as much fun as flying.
 
mccordmw said:
I had that same issue when ground testing my MPR. I had BP wrapped in foil with the ematch. It wasn't well enough contained to allow for full burning of the BP. Instead, I got a puff and no airframe separation. Happened about one in five times. I think I posted about it recently again. Now that I switched to putting BP in eppendorf tubes and taping those shut, I've ground tested 10 with 100% success rate. Also used 4 of those charges in 3 MPR flights and my L1 cert flight successfully.
Click to expand...

This is probably asked elsewhere, but it's germane to this subject, so I'll ask here: how many successful ground tests is "enough" to be confident that the flight will be nominal?
 
soopirV said:
This is probably asked elsewhere, but it's germane to this subject, so I'll ask here: how many successful ground tests is "enough" to be confident that the flight will be nominal?
Click to expand...

If it extends the shock cord completely, one should do it. I use Triple Seven not BP, so I'm used to tightly containing the charge, but it's probably a good idea to do the same with BP as well to prevent unburned powder from getting blown out of the charge well.
 
Hi Bill,

I'm sorry your L3 didn't go as planned. I know how that feels.

What size were your charge wells? The one's I use on my Punisher 4 are .75" ID x .875" D. I'm using 2g for the drogue. I pour the powder in, set the ematch so the head is centered in the BP, pack that with dog barf and then seal it with aluminum duct tape. All four of my charges have gone off without a problem on 3 consecutive flights.
 
ChrisAttebery said:
Hi Bill,

I'm sorry your L3 didn't go as planned. I know how that feels.

What size were your charge wells? The one's I use on my Punisher 4 are .75" ID x .875" D. I'm using 2g for the drogue. I pour the powder in, set the ematch so the head is centered in the BP, pack that with dog barf and then seal it with aluminum duct tape. All four of my charges have gone off without a problem on 3 consecutive flights.
Click to expand...

Chris, my home made T6061 containers are 0.5" ID, and roughly 1.75" tall. After adding BP and the ematch, I pack firmly with dog barf and seal with masking tape (criss-cross, then around the perimeter good and tight.
My gut tells me I should have used electrical tape or duct tape, several layers.

If I think in terms of burn surface area, my next ones will likely be larger diameter. I will be doing several tests to quantify the difference.
 
Wait a minute,,
First,,
The way you described the way you packed your charge wells is the way I always pack mine..
Either copper pipe with a soldered on cap or Doghouse PVC,, put the match in,,
pour in the FFFF fill with barf, pack tight, 2 criss cross layers of masking tape, and a band of masking tape around...
I've never seen this fail..
Charge size,, when you said 2 g primary and 3 g secondary I thought,, that should be good..
The way I get to a final amount is the ground test...
I like the separating parts to almost come to the end of the harness,, but not quite...
Your ground test vid confirmed to me a good / correct amount of FFFF....
I use Marsa altimeters..
I have dealt with Dermiggio extensively...
I put a lot of stock in what he says anyway, but I agree that loosing the containment of both or even one charge is highly unlikely,,
I contain the same way and I've never seen that happen...

I have a question Bill,,
How was the harness packed ??

I have found the importance of neatly "Z" folding the harness and restraining each "Z" folded bundle
with a properly placed band of masking tape or a child's braces rubber band...
When I first started flying my 4" glass scratch "SteamPunk" I wasn't Z folding the payload sections 7/16" Tubular Kevlar harness below the piston....
It's about a 7 ft harness in a tight space, below the glass piston.. I would just haphazardly coil the harness around 4 fingers and stuff it in under the piston..
When the charge went off the harness would knot up or ball up as opposed to unfurling to allow the piston to move upward..

Because we see that both charges went off,, and they were appropriately sized and packed correctly....
Is it possible that the harness balled up instead of unfurling and restrained the separating parts ....

Teddy
 
Man, am I sorry you didn't get your level 3 Bill,,
I thought it was in the bag for sure..

Teddy
 
onebadhawk said:
Wait a minute,,
First,,
The way you described the way you packed your charge wells is the way I always pack mine..
Either copper pipe with a soldered on cap or Doghouse PVC,, put the match in,,
pour in the FFFF fill with barf, pack tight, 2 criss cross layers of masking tape, and a band of masking tape around...
I've never seen this fail..
Charge size,, when you said 2 g primary and 3 g secondary I thought,, that should be good..
The way I get to a final amount is the ground test...
I like the separating parts to almost come to the end of the harness,, but not quite...
Your ground test vid confirmed to me a good / correct amount of FFFF....
I use Marsa altimeters..
I have dealt with Dermiggio extensively...
I put a lot of stock in what he says anyway, but I agree that loosing the containment of both or even one charge is highly unlikely,,
I contain the same way and I've never seen that happen...

I have a question Bill,,
How was the harness packed ??

I have found the importance of neatly "Z" folding the harness and restraining each "Z" folded bundle
with a properly placed band of masking tape or a child's braces rubber band...
When I first started flying my 4" glass scratch "SteamPunk" I wasn't Z folding the payload sections 7/16" Tubular Kevlar harness below the piston....
It's about a 7 ft harness in a tight space, below the glass piston.. I would just haphazardly coil the harness around 4 fingers and stuff it in under the piston..
When the charge went off the harness would knot up or ball up as opposed to unfurling to allow the piston to move upward..

Because we see that both charges went off,, and they were appropriately sized and packed correctly....
Is it possible that the harness balled up instead of unfurling and restrained the separating parts ....

Teddy
Click to expand...



Thanks Teddy,

The harness is Z-folded, using lightweight rubber bands on both ends. I've done countless flights this way and it always works well. When I disassembled the payload pay, the harness was still flight ready, and the drogue was still in the nomex protector. From my perspective there was insufficient pressure to shear the pins, unlike the video of my ground test.
 

Similar threads

mperegrinefalcon
L3 Build
2 3 4
Replies
98
Views
12K
mperegrinefalcon
mperegrinefalcon
C
75mm MD L3 Build
2 3
Replies
63
Views
9K
watermelonman
watermelonman
A5tr0 An0n
98mm Min Dia L3 Build
3 4 5
Replies
147
Views
37K
TRFfan
T
babp1011
L3 Build - Project Icarus
Replies
9
Views
3K
babp1011
babp1011

Latest posts

Back
Top