Post your Raven FIPa files here

the rocketry show
Advertise with us
Onebadhawk
The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
cvanc said:
Hi to all Raven owners (and potential owners)-
Sound good?
Click to expand...
Does it sound like a good idea to post flight data files for all to see and share? This stuff fascinates me and I think there's a lot of learning to be had.

Examples of both good flights, where everything went well, and not-so-good flights are welcome. I think both kinds can be learned from and if we start sharing our experiences we'll all learn faster. Could lead to some interesting discussions.

Hi cvanc; here's mine (dual deployment, one charge per event each with two igniters. One for the Raven 3, one for the G-Wiz LCX backup.

Raven 3 barometric apogee, 570 feet main, LCX acceleration apogee, 470 feet main. Raven 3 apogee measurement was 749 feet; the LCX was 752 feet.
The pressure spike of 0.0026 atm recorded was probably due to a small leak in the avionics bay when my back-up G-Wiz LCX fired on accelerometer. The Raven's igniter survived the deployment charge firing via the G-Wiz (the Raven drew current between the pressure spike and apogee..igniter still intact?). The Raven also registered a positive acceleration and velocity spike consistent with the rather heavy tail being ejected downwards by the G-Wiz.......Does this make sense?
Questions: why is the apogee channel voltage behavior so much different than the main pyro voltage?....The apogee is all over the place but the main dips on the two events only. I'm assuming we are measuring the voltage that each igniter "sees" during firing (drops to zero at firing).....is this correct?
In the Parameter Selection dialog box there are two accelerometer based velocities: one below the temperature selection and the other below the barometric altitude selection. What is the difference ? (they show different values).
The rocket landed without a scratch.
Thanks in advance,
(PS - my rocket is also called the Raven; mere coincidence...)
Paul
 

Attachments

  • The Raven 051813 Flight 2.FIPa
    67.6 KB · Views: 27
Last edited:
Adrian A said:
No charges fired before or during the breakup (based on the recorded continuity voltages). There is a big (20G) lateral impulse at 4.04 seconds, shortly before the breakup. Maybe a coupler failure?
Click to expand...

No vehicle failure. Booster separated from the av bay because of front bulkhead blow by. My questions are; why no fired charges, why the speed increase past the "event" ?

I had a second alt, that's what fired all the charges.
 
GRIFFIN said:
No vehicle failure. Booster separated from the av bay because of front bulkhead blow by. My questions are; why no fired charges, why the speed increase past the "event" ?

I had a second alt, that's what fired all the charges.
Click to expand...

By "front bulkhead blow by" do you mean motor forward closure blow-by?

When the rocket had its high-speed separation, the pieces were facing every which way as it slowed down to terminal velocity. The altimeter assumes that it's always pointed upward when it calculates the velocity, but without gyros, it doesn't know. So when you look at the Velocity (Accel-feet/sec) plot, which is the velocity calculated and used on-board, it follows the real velocity until the breakup, then goes up and down as the av-bay gets jerked around. When it's descending on the main chute it's hanging with a steady-state 1-G, so the velocity doesn't change from the 780 feet/second it had estimated at the end of the tumble. That's too fast for reliable baro measurements, which is why neither charge fired. If the rocket had stayed together, the velocity estimate would have gone down to zero at apogee and stayed there on the way down, and the charges at apogee and main would have fired as expected.
 
Adrian A said:
By "front bulkhead blow by" do you mean motor forward closure blow-by?
Click to expand...

Yes, that's what I meant. The motor is a snap ring motor, I'm used to calling them bulkheads. (truth be told, I forgot to put an eye bolt in the closure. It was the style that had a small hole going through the center....I thought i had one of my solid "puck" style closures installed)

Thanks for the insight on why the charges never fired. I guess that's why it is always important to run redundant electronics. The rocket would have been a 100% loss without the 2nd altimeter. I guess this would be a good time to re-calibrate the Raven. I plan on a Fathers day weekend flight on the same motor. I will post the data and hopefully a more successful flight.

Thanks again, Adrian.



-Jason (who will never forget to put on the eye bolt on ANY motor)
 
GRIFFIN said:
Thanks for the insight on why the charges never fired. I guess that's why it is always important to run redundant electronics. The rocket would have been a 100% loss without the 2nd altimeter. I guess this would be a good time to re-calibrate the Raven. I plan on a Fathers day weekend flight on the same motor. I will post the data and hopefully a more successful flight.

Thanks again, Adrian.
Click to expand...

Actually, the data from the main chute descent shows that you may not need a calibration. You should only re-calibrate it if the G readings with either end up are outside of 0.95-1.05 Gs.

And it's probably not true that the rocket would have been a total loss without the redundant altimeter; the fact that the main chute came out at apogee is why the Raven didn't fire its charges. Why is that?

Take a look at the Velocity Accel-Ft/sec plot, and follow it through the flight:



During the motor burn, the velocity quickly got up over the 400 foot/second threshold that the default settings use to determine when the rocket is going transonic or higher, (too fast to trust the baro readings). That happened at 2.18 seconds. At motor burnout when it had its high-speed separation, the rocket had about a half second of tumbling hash where sometimes the accel was pointed up and sometimes down, and in the net it couldn't tell that the velocity had come down a lot from the 700 feet/second it had before the breakup. Once the rocket stopped tumbling upward and started falling, if the Raven were the only electronics on board, the velocity estimate would have gradually dropped since it would no longer be measuring a steady upward 1 G. You can see this in the period between the two big accel spikes between 5 and 9.5 seconds when the main chute inflated.



From then on, the rocket was hanging upright from the main chute, and the Raven was measuring 1 G just like it was on the pad, so its velocity estimate didn't come much down until landing, when the rocket was on its side. If the main chute didn't come out at apogee, the Raven's velocity estimate would have gone down below the too-fast threshold after about 10 seconds and the Raven would have fired its charges then.
 
OK, back at the Tripoli Vegas Springfest launch (great time for those that missed it) I flew my two stage Wildman with a K815SK staging to a K160CL. The flight was, umm, interesting. First, the rocket weathervaned fairly badly during the booster burn. The stages separated (using the 3rd pyro channel set at 2nd stage at motor 1 burnout) successfully and the sustainer ignited using pyro 4 with a 2 second delay, Velocity > 500ft/sec, Pressure decreasing, and Height above pad > 576 ft. At burnout of the longburn K160, the rocket was more horizontal than vertical. The main deploy at approximately burnout. The apogee charge never burnt.

Now, looking at the data, it doesn't look like I ever made the triggers for the main (Height above pad < 480ft, pressure increasing, and Velocity < 400ft/sec, and after Motor 1 burnout.) Also, it appears there was voltage at the main channel the entire time? That doesn't make sense to me...

This is a version 1 Raven on a Power Perch using the standard LiPo for it. Surprisingly, there was no damage to the airframe, just a couple popped cords on the chute due to the stupid fast deployment.

Thoughts?

Droopy

View attachment Two Stage Wildman K815 K160CL (fail).FIPa
 
Adrian A said:
Actually, the data from the main chute descent shows that you may not need a calibration. You should only re-calibrate it if the G readings with either end up are outside of 0.95-1.05 Gs.

And it's probably not true that the rocket would have been a total loss without the redundant altimeter; the fact that the main chute came out at apogee is why the Raven didn't fire its charges. Why is that?

Take a look at the Velocity Accel-Ft/sec plot, and follow it through the flight:



During the motor burn, the velocity quickly got up over the 400 foot/second threshold that the default settings use to determine when the rocket is going transonic or higher, (too fast to trust the baro readings). That happened at 2.18 seconds. At motor burnout when it had its high-speed separation, the rocket had about a half second of tumbling hash where sometimes the accel was pointed up and sometimes down, and in the net it couldn't tell that the velocity had come down a lot from the 700 feet/second it had before the breakup. Once the rocket stopped tumbling upward and started falling, if the Raven were the only electronics on board, the velocity estimate would have gradually dropped since it would no longer be measuring a steady upward 1 G. You can see this in the period between the two big accel spikes between 5 and 9.5 seconds when the main chute inflated.



From then on, the rocket was hanging upright from the main chute, and the Raven was measuring 1 G just like it was on the pad, so its velocity estimate didn't come much down until landing, when the rocket was on its side. If the main chute didn't come out at apogee, the Raven's velocity estimate would have gone down below the too-fast threshold after about 10 seconds and the Raven would have fired its charges then.
Click to expand...

Adrian,

I appreciate all your input on these flights. I would never be able to diagnose these issues. Thank you again, great customer support.
 
droopy said:
OK, back at the Tripoli Vegas Springfest launch (great time for those that missed it) I flew my two stage Wildman with a K815SK staging to a K160CL. The flight was, umm, interesting. First, the rocket weathervaned fairly badly during the booster burn. The stages separated (using the 3rd pyro channel set at 2nd stage at motor 1 burnout) successfully and the sustainer ignited using pyro 4 with a 2 second delay, Velocity > 500ft/sec, Pressure decreasing, and Height above pad > 576 ft. At burnout of the longburn K160, the rocket was more horizontal than vertical. The main deploy at approximately burnout. The apogee charge never burnt.

Now, looking at the data, it doesn't look like I ever made the triggers for the main (Height above pad < 480ft, pressure increasing, and Velocity < 400ft/sec, and after Motor 1 burnout.) Also, it appears there was voltage at the main channel the entire time? That doesn't make sense to me...

This is a version 1 Raven on a Power Perch using the standard LiPo for it. Surprisingly, there was no damage to the airframe, just a couple popped cords on the chute due to the stupid fast deployment.

Thoughts?

Droopy

View attachment 132971
Click to expand...

Good job on having a nice, safe airstart program. I always like to see an altitude check on high powered airstarts, and the velocity check helps too.

The data shows that you had the sustainer airframe separate before the Raven fired any charges. About the only event I see leading up to the separation is an increase in the baro pressure. Howerver, the accel spike at 14.55 seconds, at the front of the separation data, is really high and sharp though, like a deployment charge. Were there any other electronics on board that might have fired a charge? The Raven fired the main charge when it was passing through 500 feet on the way down, 400 seconds into the flight.
 
Any one know why a Raven would only arm upside down?

I am not sure that there was a specific orientation required on any of the ravens.

Thanks in advanced, this should be a quick answer to my dumb question.

I did a little searching and came up with nothing.
 
Maybe the calibration is bad? There's a procedure in the documentation for recalibrating the accelerometer. I'd suggest trying that.

-Kevin
 
GRIFFIN said:
Any one know why a Raven would only arm upside down?

I am not sure that there was a specific orientation required on any of the ravens.

Thanks in advanced, this should be a quick answer to my dumb question.

I did a little searching and came up with nothing.
Click to expand...

troj said:
Maybe the calibration is bad? There's a procedure in the documentation for recalibrating the accelerometer. I'd suggest trying that.

-Kevin
Click to expand...

If the calibration is off far enough, it's out of the range where the Raven knows it's upside-down. A re-calibration should take care of it. You should have 0.95-1.05 Gs in the live reading for either orientation.
 
GRIFFIN said:
Anyone (Adrian) care to shed some light on what may have happened here at 63.75 (time)
Click to expand...

One guess is that your upper and lower sections were running into each other. Was this a drogueless setup? Did you see any dents or damage? There seems to be a lot of action on the way down to main deployment.
 
Adrian A said:
One guess is that your upper and lower sections were running into each other. Was this a drogueless setup? Did you see any dents or damage? There seems to be a lot of action on the way down to main deployment.
Click to expand...


That's what i thought. Yes, it was droguless. I have a bog "crack" in the G12 booster. I thought they must have collided at main deployment. Hard to imagine that they hit that hard while tumbling. Time to stat using a drogue. I have had a dozen flights on that rocket with no drogue and no issues......lesson learned.
 
Here is a "successful failure" of a 29mm DD initial test flight with an AT F27-8R. Early motor eject seems to have jarred the battery connection loose.

View attachment 137125

The idea was to test my cable cutter setup in flight with a dummy burrito, but actually deploy the main at apogee using the motor ejection charge as backup. The motor delay was 4.9 s instead of 8 s, so the main deployed at 70 mph going up (example of how redundancy and testing can bite you a little even if they are best practice). Looks like the motor eject popped the Raven 3 lipo battery connection loose followed by 2.5 s of data powered by super cap. No cable cutter event this time, but I learned that the battery mounting needed work. It is a little tricky to get this right since the active bulkhead has to seal the AV bay and apply a snug battery fit against passive bulkhead. My AV bay setup uses a longer coupler tube to accommodate a GPS tracker that extends into the nose cone above the Raven 3. The passive bulkhead is sanded down to fit inside the coupler and supported by hex nuts on the forward side. These hex nuts have to be ground down on the outer side to fit inside the coupler, so they are not very user friendly when it comes to aligning things. In addition to being more careful on bulkhead spacing, my solution includes a cable tie around the shoulder of the lipo battery through the PCB hole. Not so tight as to damage battery or board, but snug enough to provide additional support. The assembly before slipping it into the AV bay is shown below.
View attachment 137126
 
Yes, the av-bays need the bulkheads to be spaced apart pretty precisely in order to prevent the battery from coming loose. Putting a 2.0" coupler between the bulkheads is a good way to get the right spacing every time.
 
I just bench tested the Raven with an inert igniter (hand made, nichrome bridgewire, no pyrogen.)
All 4 channels can bring the nichrome to a nice orange glow at the appropriate time in the test routine.

From the FIPa data, it appears that the Velocity was "stuck" and never dropped below 400ft/sec required to fire the charges.
In fact, it shows that apogee was at 3333 MPH. I assure you this would have been a shred at that speed, not to mention in low earth orbit at 490k'

Why is the accelerometer so off? On the bench, the live data shows Axial and Lateral Gs at near 1.0 as expected for each position.

Madcow-SuperDX3-Aug10-13.CTI-970J394-13A-Green.png
 
Last edited:
KBlack said:
I just bench tested the Raven with an inert igniter (hand made, nichrome bridgewire, no pyrogen.)
All 4 channels can bring the nichrome to a nice orange glow at the appropriate time in the test routine.

From the FIPa data, it appears that the Velocity was "stuck" and never dropped below 400ft/sec required to fire the charges.
In fact, it shows that apogee was at 3333 MPH. I assure you this would have been a shred at that speed, not to mention in low earth orbit at 490k'

Why is the accelerometer so off? On the bench, the live data shows Axial and Lateral Gs at near 1.0 as expected for each position.

View attachment 141690
Click to expand...

It looks like the accelerometer readings were junk during the flight. Since the accel reads correctly on the bench, either we're looking at an intermittent failure, or there was something in your configuration that shorted an accelerometer trace or something. When you did the bench test that worked, was that with the altimeter installed in your av-bay the way it was for the flight?

When you look at the live data readings and you flip the altimeter over 180 degrees, so you see -1G until the sign convention is flipped? When it's on its side, does it read near 0?

Looking at your data more closely, I see that in the first part of the flight, the data looks sort of plausible except that during the coast period, there is a steady +6-+7G reading. This is what you get when the accelerometer's test mode is activated, which is possible if some conductive material touches the wrong pin. The hash we see later in the flight could be from contact with adjacent pins.
 
Last edited:
Adrian A said:
When you did the bench test that worked, was that with the altimeter installed in your av-bay the way it was for the flight?
Click to expand...

Mounted to the Madcow 4" av bay sled, connected to the power perch.

Adrian A said:
When you look at the live data readings and you flip the altimeter over 180 degrees, so you see -1G until the sign convention is flipped? When it's on its side, does it read near 0?
Click to expand...
Near zero Axial and lateral when laying flat, face up on the desk. On the sides Lateral is near 1, with axial near 0., flipping to the other side sees the values go negative and stay there (USB side down.)
With green phoenix terminals up and USB down, Axial near 1, Lateral near 0.
Terminals down, USB up, Axial goes to near -1 then +1, lateral near zero.

Adrian A said:
Looking at your data more closely, I see that in the first part of the flight, the data looks sort of plausible except that during the coast period, there is a steady +6-+7G reading. This is what you get when the accelerometer's test mode is activated, which is possible if some conductive material touches the wrong pin. The hash we see later in the flight could be from contact with adjacent pins.
Click to expand...

AV bay interior is clean, no residues and no uninsulated wires or splices in the AV bay.

There was a BigRedBee GPS transmitter in the upper section of the rocket, about 20" above the Raven. TPO of the BRB-GPS set to 12dBm, which I don't expect would cause an issue, unless the ignitor leads were by odd luck close to a multiple of a quarter wavelength of the ~433MHz I was transmitting at. If RF from the tracker is an issue, perhaps a shielded AV bay, or looping the leads through a ferrite core, or some RF bypass capacitors on the outputs would help, but I don't think that at this power level there would be interference.

I don't see any solder joints that I would suspect, but the LGA mount devices make their inspection impossible. I could attempt reflowing with my hot air rework station.
 
I took another look tonight for more clues. The lateral accel looks normal throughout the flight, so I don't think the problem is with the microcontroller internally, or a problem with something like the power feed to the accelerometer. That also suggests that the accelerometer's self-test mode isn't getting activated, since that would show up in the lateral accel data also. My best guess is that there is an intermittent solder problem somewhere between the microcontroller pin and the axial lead of the accel.

If you want to ship it back, I'll repair or replace it. PM me for return address information. If you want to have a go yourself, see if you can reproduce any bad behavior in the live data by manipulating or bending the board. The trace in question is at the top center of the accel part as you look at the board with the R025 current sense resistor in the lower right.
 
Something is wrong with primary Raven 3. Accelerometer reports 1.67 s burn for 4.9 s M1520. Baro is flat during burn, then goes negative (-1525') before under reporting apogee by ~3000' vs secondary. Primary main pyro voltage drops out at 87 s. Then primary stops recording after 131 s versus 157 s for secondary as if Raven 3 was fooled into thinking it had landed or had powered down during descent.
 
Jimmy F. said:
Two Ravens on board, same flight, dramatic difference in data
primary is Raven3 70g
secondary is Raven2 250g
clarification on high gees upon landing-main chute did not entirely inflate
Click to expand...

The flight files you posted are from two different flights. If you have been using both of these Ravens together for more than one flight, I think you'll find that at least one of the posted flights has a match still saved in the data on the other altimeter.
 

Similar threads

Adrian A
Altimeter MOSFET deployment switches turning on when power is applied
2
Replies
30
Views
2K
rocket_troy
R
N
Premature Drogue Chute Deployment Diagnosis
Replies
7
Views
350
cwbullet
cwbullet
C
Hello!
Replies
8
Views
417
Hobie1dog
Hobie1dog
J
Aerotech HV Arcas Build Thread
Replies
29
Views
1K
JayB
J
K
My Level 2 Attempt - a Diurnal Flight of "Nocturnal Missions"
Replies
5
Views
629
kjhambrick
K

Latest posts

Back
Top