Post your Raven FIPa files here

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
So, this past weekend I had a very interesting occurrence with my Featherweight Parrot V2. I'd post the file, but it's 1.7MB and too big for the forum.

The rocket was my 4" LOC Bullet, weighing 2.5 lbs on the pad loaded with an H238-10T motor, Rocketman R3 chute, and the Parrot in a bay between two of the fins. This has flown four times now, all successful flights. All have used motor back up, which has saved the rocket twice now.

1st flight was nominal, with deployment via the Parrot and motor back up.
2nd flight had no altimeter deployment, but a late deployment by the motor (un-drilled Loki delay in an EX 38-240 blue motor, 14s too long). Charge hooked up to the altimeter was still intact. This was my fault, as I did not charge the onboard LiPo prior to this flight.
3rd flight was nominal, with deployment via the Parrot and motor back up (onboard video here)

And now for the 4th flight. Rocket went up really quickly (Parrot shows Max Accel of 20.36 gees), then deployed the parachute perfectly at apogee. I listened for two charges, but only heard one. Rocket drifted down on the R3 chute, then landed about 1,000' away from the pad, 74 seconds into the flight. I grabbed the video camera off the ground that I was using for ground footage, then started my recovery walk.

When I was about 50' away from the rocket, I hear a "THUD" and see a decent-sized puff of smoke emit from the rocket, obviously that of a black powder charge. Based on the Parrot data, this occurred 237 seconds "into the flight," or a full 163 seconds after landing! There is a small blip in the data trace where the charge went off and nudged the rocket a little.

I HIGHLY doubt that this was the motor ejection charge. I inspected the area in front of the parachute chamber upon reaching it for the red cap that was holding in the black powder on the motor. It was not found.

So, why, after 163 seconds (over 2.5 minutes!) of nothing did the Parrot deploy? Settings for apogee deployment are "out of the box" default.

If you'd like me to e-mail you the file for investigation, please let me know. Any opinions are greatly appreciated, as right now I am a little uneasy with depending on this altimeter.

View attachment Bullet_4_13_12_382G_Blue.FIPa

View attachment Bullet_6_9_12_Smokey.FIPa
 

Attachments

  • Screen Shot 2013-03-16 at March 16, 20137.24.48 PM.png
    Screen Shot 2013-03-16 at March 16, 20137.24.48 PM.png
    199.6 KB · Views: 20
Last edited:
So, this past weekend I had a very interesting occurrence with my Featherweight Parrot V2. I'd post the file, but it's 1.7MB and too big for the forum.

I don't need to see the data file because I have seen this before using other altimeters. The problem is a combination of three things:

1) Altimeter inertial (accelerometer based) deployment algorithm is late or something else is early.
2) Something else causes the apogee event. In your case motor deployment.
3) The altimeter is mounted so that after the apogee event, it has the same orientation as at the pad. Mounting between the fins does this.

What happens in this situation is that after deployment the accelerometer is reading almost exactly 1G but the firmware is subtracting almost exactly 1G from that before the integration step. The result is that the integrated velocity changes very little and can delay the altimeter apogee event a very long time.

Look at the velocity data and you will see a slope change at deployment. If there is an error in the 1G offset then it will be something other than a horizontal line. When the rocket lands there will be another slope change. Now velocity is decreasing and it is only a matter of time till the altimeter fires the charge.

Including a timer as a backup will make sure that the charge fires in the air. Depending on charge location having it go off on the ground could be really bad if the rocket lands in dry grass.
 
I don't need to see the data file because I have seen this before using other altimeters. The problem is a combination of three things:

1) Altimeter inertial (accelerometer based) deployment algorithm is late or something else is early.
2) Something else causes the apogee event. In your case motor deployment.
3) The altimeter is mounted so that after the apogee event, it has the same orientation as at the pad. Mounting between the fins does this.

What happens in this situation is that after deployment the accelerometer is reading almost exactly 1G but the firmware is subtracting almost exactly 1G from that before the integration step. The result is that the integrated velocity changes very little and can delay the altimeter apogee event a very long time.

Look at the velocity data and you will see a slope change at deployment. If there is an error in the 1G offset then it will be something other than a horizontal line. When the rocket lands there will be another slope change. Now velocity is decreasing and it is only a matter of time till the altimeter fires the charge.

Including a timer as a backup will make sure that the charge fires in the air. Depending on charge location having it go off on the ground could be really bad if the rocket lands in dry grass.

This is a very interesting hypothesis. My only concern is that this deployment occurred 2.5 minutes after touching down. The rocket did not deploy while going up, closer investigation of the data shows that the motor actually deployed the chute slightly after apogee. I definitely recall the rocket being horizontal when the chute deployed. I'd expect this jostling would be enough to tip off the accelerometer.

For future flights, I plan to change the method of apogee sensing to barometric instead of via the accelerometer. Seeing as this rocket will never fly near the transonic region, I am more comfortable with that method.
 
Many altimeter firmware implementations inhibit or don't check for main event unless the apogee event occured.

This could be remedied by an apogee detect override for baro + accel altimeters. If the barometric system detects that an apogee has likely occured then the main altitude event is enabled even if an accel apogee detect was programmed. In this case of an acceleration integration anomoly that prevents an apogee event the main will still fire as the rocket descends through the main programmed altitude.

There is at least one altimeter on the market that incorporates this implementation.
 
Many altimeter firmware implementations inhibit or don't check for main event unless the apogee event occured.

This could be remedied by an apogee detect override for baro + accel altimeters. If the barometric system detects that an apogee has likely occured then the main altitude event is enabled even if an accel apogee detect was programmed. In this case of an acceleration integration anomoly that prevents an apogee event the main will still fire as the rocket descends through the main programmed altitude.

There is at least one altimeter on the market that incorporates this implementation.

Great idea. The recovery scheme was main parachute apogee deployment via the Parrot, not dual deploy. The motor delay was used as back up and is typically chosen to be much longer than an "accurate" delay. In this case, the delay was chosen because a 10s delay comes standard in 29-180 loads. Oddly enough, it was just about perfect and sparked all of the issues discussed here.
 
Patelldp was kind enough to email me his FIPa file.

I don't know the Parrot, all I have are 2 Ravens. It looks like all 3 of the e-match outputs are added together in the data into a single channel called 'continuity'. (Do I have that right?)

Also I guess Parrots don't log the battery voltage. So the only way to get a sense of battery voltage is indirectly, by looking at the voltage reported under 'continuity'.

Anyway the 'continuity' shows the highest voltage as only 0.4 (4 tenths of a volt)! Could this be indicative of a severely drained battery?

And the [Altitude (Accel - Ft)] data is really weird. It shows the flight ending at -375,000 feet. Which I somehow suspect isn't actually the case LOL. Ground_charge.png

Based on the extremely low voltage reported on 'continuity' I wonder if your battery was nearly dead?
 
Last edited:
Patelldp was kind enough to email me his FIPa file.

I don't know the Parrot, all I have are 2 Ravens. It looks like all 3 of the e-match outputs are added together in the data into a single channel called 'continuity'. (Do I have that right?)

Also I guess Parrots don't log the battery voltage. So the only way to get a sense of battery voltage is indirectly, by looking at the voltage reported under 'continuity'.

Anyway the 'continuity' shows the highest voltage as only 0.4 (4 tenths of a volt)! Could this be indicative of a severely drained battery?

And the [Altitude (Accel - Ft)] data is really weird. It shows the flight ending at -375,000 feet. Which I somehow suspect isn't actually the case View attachment 122811LOL.

Based on the extremely low voltage reported on 'continuity' I wonder if your battery was nearly dead?

All great points. The pyro battery was an aux LiPo that measured 4.2 volts prior to the flight. The onboard battery had flown previously that day, but had only been turned on for a maximum of 10 minutes. I did not measure the voltage of the onboard battery inbetween flights. That battery is used to power the unit's processor.
 
If you email the file to me, I'll check it out. From what I've seen so far, my leading two guesses about what happened are: 1. Motor ejection caused at apogee ejected the chute, which caused high G readings on the way down, which kept the Mach lockout active until after the rocket was on the ground for a while (the Parrot doesn't have the Raven's landing detection). 2. Run-down battery

Thanks.

-Adrian
 
I flew my Raven3 as a passenger on Saturday at a NAR NARHAMS launch at the Mt Airy Maryland Rec Fields. I will post the FIP and see if you can guess what happened.

Dick Moran
NAR 6306L1
Tri 14074L1

View attachment FC3.FIPa
 
I flew my Raven3 as a passenger on Saturday at a NAR NARHAMS launch at the Mt Airy Maryland Rec Fields. I will post the FIP and see if you can guess what happened.

Dick Moran
NAR 6306L1
Tri 14074L1

That's a puzzle. Wild guess: Was the static port immediately behind an abrupt transition, with the aft side narrower?
 
!!!!!!!!!!!!!!!!!!!!!!!!!!!!____Spoiler Alert Spoiler Alert____!!!!!!!!!!!!!!!!!!!!!!!! Do not read this if you are still thinking about Post 193........
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
 

Attachments

  • IMG_0578 (Medium).jpg
    IMG_0578 (Medium).jpg
    331.9 KB · Views: 24
So it appears there was deployment at apogee, but you had separation?
 
I know Adrian navigates this thread once in a while, so I would like to repost a question that was asked in the middle of another thread:

"Now, this issue is something a little more interesting. I am not fully up to snuff on this theory, actually not up to snuff at all, so I will mention it and hope that the educated individuals make themselves known!

From what I understand, a magnetic field is generated by a rocket motor as it burns. Strength is determined by the amount of mass flowing through the motor (I think) and it becomes more severe as motor size increases.

I think that I recall Adrian of Featherweight state that the Delta that they flew a few years ago actually experienced toggling of the magnetic switches as a result of this phenomenon.

Butalane, Daveyfire, MClark, and Adrian will have more knowledge on the subject."

In my particular situation, I am using 2 Ravens, each mounted on the Power Perch. They are located around the motor mount tube, hence pretty close to the motor. Is this a real issue? Should I consider using a different approach, and is there a way to use the Power Perch with an external switch instead of the magnetic switch?

The thread is located here
 
A magntic field is not generated by a solid rocket motor when it burns, so the conbustion of the motor will not effect the magnetic switch.

There can be a percent of 2 of ions present in the combustion, but there is no net charge since the number of positive ions equals the number of electron so the net effect is charge neutral. You are confusing what could happen to the ions if a strong magnetic field were present along the axis of the motor. In that case the positive ions would travel in one direction and the electrons would travel in the opposite direction, but that is not the case.

Bob
 
A magntic field is not generated by a solid rocket motor when it burns, so the conbustion of the motor will not effect the magnetic switch.

There can be a percent of 2 of ions present in the combustion, but there is no net charge since the number of positive ions equals the number of electron so the net effect is charge neutral. You are confusing what could happen to the ions if a strong magnetic field were present along the axis of the motor. In that case the positive ions would travel in one direction and the electrons would travel in the opposite direction, but that is not the case.

Bob

I didn't want to say anything because I didn't know for sure, but there's no reason a solid motor should ever exhibit any non-negligible amount of charge separation and motion of said separated charges.
 
A magntic field is not generated by a solid rocket motor when it burns, so the conbustion of the motor will not effect the magnetic switch.

There can be a percent of 2 of ions present in the combustion, but there is no net charge since the number of positive ions equals the number of electron so the net effect is charge neutral. You are confusing what could happen to the ions if a strong magnetic field were present along the axis of the motor. In that case the positive ions would travel in one direction and the electrons would travel in the opposite direction, but that is not the case.

Bob

I didn't want to say anything because I didn't know for sure, but there's no reason a solid motor should ever exhibit any non-negligible amount of charge separation and motion of said separated charges.

Thank you for the clarification. I had never heard about a magnetic field around the motor during burn, but I wanted to have a second ( and third) opinion. :)
 
I made the mistake of using only engine deployment in this GPS radio test. I did not have enough time to install the new Raven3 and the ejection wiring. The 3.5 lb rocket flew sort of nice on a CCI G78-15 but the whole flight from start to finish was 15 seconds. The motor ejection blew .5 sec after "landing". The whole rocket went in just as you see. The Raven3 was still working and the GPS radio had a broken battery wire or it would have still been sending. The only loss was the battery, and the av bay case which had to be cut in half to remove the hardware. I really lucked out. Once the video is cut down I will post on YouTube. The camera survived also. The video shows the whole flight looking at the back of the rocket which was looking down at the beginning and up at the sky through the remainder of the flight. As the rocket hit earth you can see the ejection charge puff out the rear of the rocket .5 seconds later. If the front had not been buried it may have separated the rocket while stuck in the ground.
Dick

https://www.youtube.com/watch?v=qe7eG_vYhvA
 

Attachments

  • IMG_0581 (Medium).jpg
    IMG_0581 (Medium).jpg
    245 KB · Views: 20
  • IMG_0579 (Medium).jpg
    IMG_0579 (Medium).jpg
    245.8 KB · Views: 22
Last edited:
I made the mistake of using only engine deployment in this GPS radio test. I did not have enough time to install the new Raven3 and the ejection wiring. The 3.5 lb rocket flew sort of nice on a CCI G78-15 but the whole flight from start to finish was 15 seconds. The motor ejection blew .5 sec after "landing". The whole rocket went in just as you see. The Raven3 was still working and the GPS radio had a broken battery wire or it would have still been sending. The only loss was the battery, and the av bay case which had to be cut in half to remove the hardware. I really lucked out. Once the video is cut down I will post on YouTube. The camera survived also. The video shows the whole flight looking at the back of the rocket which was looking down at the beginning and up at the sky through the remainder of the flight. As the rocket hit earth you can see the ejection charge puff out the rear of the rocket .5 seconds later. If the front had not been buried it may have separated the rocket while stuck in the ground.
Dick

Is that a 38mm motor? If so why didn't you adjust the delay? :confused: A quick simulation with any of the free programs would have gotten you close. I would have been better than breaking things. :facepalm:
 
Yes a 38mm CSI. I am used to flying with an H or I motor at our MDRA field in Price MD on the Eastern Shore. There I use my Raven 3 to fire charges. This may be the 1st time I have ever used motor ejection on a G and it was because the field limit is a G. Anyway I do not normally use motor ejection so I did not invest in the "tool" used to shave down the delay that Animal Motor works sells. I had made some changes to my av bay to lighten it up and ran out of time to wire up the Apogee and Main charges. Hind sight is 20x20 and it was a learning lesson for me. I have flown about 12 or so HPR flights and this is the first real failure. I may do the sim next time that I need to before using motor ejection.








Dick Moran
NAR 6306L1
Tri 14074L1
USAF [4yr2m1d]

Inspect 003 (Medium).jpg

Inspect 002 (Medium).jpg

Inspect 001 (Medium).jpg
 
Last edited:
A magntic field is not generated by a solid rocket motor when it burns, so the conbustion of the motor will not effect the magnetic switch.

There can be a percent of 2 of ions present in the combustion, but there is no net charge since the number of positive ions equals the number of electron so the net effect is charge neutral. You are confusing what could happen to the ions if a strong magnetic field were present along the axis of the motor. In that case the positive ions would travel in one direction and the electrons would travel in the opposite direction, but that is not the case.

Bob

Funny you should mention this. It's something I've always wanted to try - for no good reason. Like... would there be any net change in the direction of thrust...? Would one propellant vector one way and another anther way? Depends on the relative masses of the +/- species.
 
I know Adrian navigates this thread once in a while, so I would like to repost a question that was asked in the middle of another thread:

"Now, this issue is something a little more interesting. I am not fully up to snuff on this theory, actually not up to snuff at all, so I will mention it and hope that the educated individuals make themselves known!

From what I understand, a magnetic field is generated by a rocket motor as it burns. Strength is determined by the amount of mass flowing through the motor (I think) and it becomes more severe as motor size increases.

I think that I recall Adrian of Featherweight state that the Delta that they flew a few years ago actually experienced toggling of the magnetic switches as a result of this phenomenon.

Butalane, Daveyfire, MClark, and Adrian will have more knowledge on the subject."

In my particular situation, I am using 2 Ravens, each mounted on the Power Perch. They are located around the motor mount tube, hence pretty close to the motor. Is this a real issue? Should I consider using a different approach, and is there a way to use the Power Perch with an external switch instead of the magnetic switch?

The thread is located here

In the end I decided that it was my own error in arming due to trying to distinguish between too many sources of beeps. The Power Perch doesn't have a way to turn it on except by the magnetic switch.
 
Funny you should mention this. It's something I've always wanted to try - for no good reason. Like... would there be any net change in the direction of thrust...? Would one propellant vector one way and another anther way? Depends on the relative masses of the +/- species.
Do a search on MHD thrusters (magnetic hydrodynamic thrusters)

Also found 1 reference on enhancenent of burnrate of propellant in a strong magnetic field. Didn't read article (journal) but I'm skeptical it would be practical.

Bob
 
Do a search on MHD thrusters (magnetic hydrodynamic thrusters)

Also found 1 reference on enhancenent of burnrate of propellant in a strong magnetic field. Didn't read article (journal) but I'm skeptical it would be practical.

Bob

Thanks, Bob.

Only tangentially related, but interesting.
 
Here are a few interesting ones
 

Attachments

  • Mega DarkStar P8000.FIPa
    621.6 KB · Views: 26
  • Space Cowboy L2300 TS4.FIPa
    344 KB · Views: 29
Ouch

Hard lesson, but a good one. I run just about every flight these days several time with Open Rocket or my old version 6 Rock Sim. On numerous occasions this has saved me from making a mistake. I am glad your electronics survived. I would suggest running a redundant recovery method with them for a few flight to ensure there was no damage unseen or internal.

Cheers mate

Yes a 38mm CSI. I am used to flying with an H or I motor at our MDRA field in Price MD on the Eastern Shore. There I use my Raven 3 to fire charges. This may be the 1st time I have ever used motor ejection on a G and it was because the field limit is a G. Anyway I do not normally use motor ejection so I did not invest in the "tool" used to shave down the delay that Animal Motor works sells. I had made some changes to my av bay to lighten it up and ran out of time to wire up the Apogee and Main charges. Hind sight is 20x20 and it was a learning lesson for me. I have flown about 12 or so HPR flights and this is the first real failure. I may do the sim next time that I need to before using motor ejection.








Dick Moran
NAR 6306L1
Tri 14074L1
USAF [4yr2m1d]
 
Here are a few interesting ones

Beautiful flight on the Space Cowboy; very clean. Looks like it temporarily got into a spin with the drogue on the way down at 7000 feet.

Do you know what caused the breakup of the Mega DarkStar at 16,000 feet? Since it had already slowed down and I don't see anything like fin flutter before that, my guess is that you had a premature separation due to differential pressure. The sudden decrease in pressure during the event would be consistent with that. Were all of your sections shear-pinned?
 
Beautiful flight on the Space Cowboy; very clean. Looks like it temporarily got into a spin with the drogue on the way down at 7000 feet.

Do you know what caused the breakup of the Mega DarkStar at 16,000 feet? Since it had already slowed down and I don't see anything like fin flutter before that, my guess is that you had a premature separation due to differential pressure. The sudden decrease in pressure during the event would be consistent with that. Were all of your sections shear-pinned?


We didn't see any problem like that with the Mega. The main came out at apogee, but that was all. All sections were bolted together except the nose section, it was pinned.

This is why I like it when you read these files. You really have a better understanding of them. I wish you would "diagnose" all of theses and give a little incite.
 
We didn't see any problem like that with the Mega. The main came out at apogee, but that was all. All sections were bolted together except the nose section, it was pinned.

This is why I like it when you read these files. You really have a better understanding of them. I wish you would "diagnose" all of theses and give a little incite.

Thanks.

At 21 seconds into the Darkstar flight, a little forward bump shows up in the accelerometer, followed by a drop in baro pressure and then a lot of hash on both accel axes as the rocket sections tumbled chaotically at 150 feet/second. The continuity voltages show that no charges went off until later. There were a couple of other blips in the accel data before then that may have been related.

At 16,000 feet, there is only about half the air pressure that you get at sea level, so if there isn't any venting, you would have about 7.5 psi pushing your airframe sections apart, not including any differential drag forces. A mega darkstar is 12" diameter, if I'm not mistaken, so that's 36*3.14*7.5, or over 800 lbs, pushing the airframe apart if all the compartments are sealed. Vent holes will reduce that significantly, but even 1 PSI difference between outside and inside will put push the sections apart with over 100 lbf for that diameter. I learned about this when my own 38mm rocket, without vents, came apart at 10,000 feet and Mach 1, and the recorded data showed a big drop in internal pressure right at the start of the the event. (the rocket survived the chute-less descent, and the Beeline, Raven, and 38mm av-bay holding them were still powered on and mostly intact after a 900G lawn dart)
 
Last edited:
Back
Top