I managed to lose a rocket with GPS Telemetry

[tedapke]

Well, technically, it was my son's (Colin's) rocket.

Colin flew his MadCow Tomach on a CTI K590 Dual Thrust on 8/10/14.
He had a Telemetrum on board. The boost was beautiful and telemetry showed the rocket went a little over 20k'.
Unfortunately telemetry then showed that the rocket was coming back down way to fast! We had telemetry right up to 33m off the ground. Data showed the rocket landed about 0.8 miles from the pad, just over a small hill. We went out prepared to do some excavating but there was nothing to be seen!

We walked around for over two hours including searching for the beacon from the Telemetrum, with no luck.

I analyzed the data and went back to the site on Tuesday, 8/12, armed with a metal detector (aluminum capable), a GPS, some rope and a bunch of stakes. The ground cover is very short, shorter than even what we have around the NCR North site, so you can see anything sticking up more than a couple of inches. I marked the last GPS position and then pulled a line 100' to the West. I searched that line, marking both ends. I then moved 10' North and repeated. I did this for 100' and then moved East of the last GPS position and repeated. I then repeated Southwest and Southeast until I had searched, and marked, an area 200' by 200'. Still no sign of the rocket. Note that most places the ground was hard enough that it was difficult to put in a 3/8" stake! There were spots where I could push said stake over 18" without meeting a lot of resistance and there were a lot of prairie dog holes (I looked in them, too). I still had some time before the metal detector needed to be returned so I did 3 laps around the 200' square, each pass moving out more feet. Still no luck.

I've attached the csv file of the telemetrum data, taken from Colin's phone and convedted with Altosui to csv.
Looking at the last few lines of the data you can see that the headings don't seem to match the columns (looks like latitude is 4 columns left of the header). If the same error applies to the right of latitude then hdop (horizontal dilution of precision) was 95. This would indicate a very poor accuracy and could easily account for why I didn't find the rocket.

The telem file (raw telemetrum data, via the teledongle) is available here:
https://drive.google.com/file/d/0ByHmN01gRSfAdmVEWlBXRjJINHc/edit?usp=sharing

Any input will be appreciated. This includes suggestions as to what went wrong and where to look for the remains.
Thanks
Ted Apke
TRA 03726 L3

View attachment 2014-08-10-serial-1326-flight-0001-via-1052-shortened.txt

 

[dixontj93060]

I don't want to be negative but there sure seems to be an inordinate amount of hiccups and negative experiences reported on TRF for Telemetrum units. Maybe I'm wrong (hope I am) but it's pretty evident in a search.

 

[RAHagen]

Hmm. I just got a telemetrun for my Lv. 1 cert. So I hope I don't have the same, very unfortunate , result. Did the telemetrum have a full GPS lock? That's the only thing I can think of. Is it possible that there was a flaw in the system?

 

[RAHagen]

Also, following up on another post I've seen, was the paint a metallic? I've heard that people have had problems with metallic paint completely shielding satellite signal from the GPS module.

 

[tedapke]

I think the telemetrum is being flown a lot lately and they are more complicated than your basic altimeter. I also know of quite a few successful flights and recoveries (I had one this America day). I really don't think the ballistic recovery had anything to do with the telemetrum. Looking at the data you can see the charges appear to fire. For some reason the body didn't separate.

Fortunately I have some data to analyze which is what I really like about flying a telemetrum.

If there are problems, sharing data is how we'll get the fixed.

Thanks
Ted Apke
TRA 3726 L3


Sent from my iPad using Rocketry Forum

 

[tedapke]

If I'm reading the data correctly there were 8 sats being "tracked" during descent.

No paint. Just bare green fiberglass making for a total lack of contrast with the prarie!

Ted Apke
TRA3726 L3


Sent from my iPad using Rocketry Forum

 

[cwbullet]

Telemetrium for level 1 cert? Keep it simple. I would recommend standard recovery and a low and slow approach. It you need a telemetrium for level 1, I would recommend lower flight.

 

[bobkrech]

Are you sure you used the right conversion format for the GPS and or the mapping program? Some units/software uses decimal degrees, or degrees and decimal minutes, or degrees, minutes and decimal seconds.

A ham friend of mine was on the recovery team of a balloon flight using the balloon tracking app the owner setup to convert the balloon data to lat/lon coordinates. He drove as close to the location has he could and then walked into the woods to the location the app lead him to, but didn't see or hear the payload beeping, nor could he pick up the ARPS tracker signal.

After a couple of minutes of head scratching, the software guy realized they had assumed an incorrect data format, and then determined that the true location was 3000' away. After a 10 minutes they were able to drive to within a couple hundred yards of the predicted location and as they approached the new location they started to pick up the ARPS tracker signal and then heard the audible beeper. It was located 70' up a tree right where the corrected GPS signal said it would be....

This may not be the case in your flight, but just in case it should be investigated.

Bob

 

[Bat-mite]

Very sorry you lost your rocket. That stinks. I had to pull mine out of a pond today, but thankfully, my RRC2+ and BRB900 are still working.

 

[RAHagen]

Telemetrium for level 1 cert? Keep it simple. I would recommend standard recovery and a low and slow approach. It you need a telemetrium for level 1, I would recommend lower flight.

Sorry, that was worded wrong. I got the telemetrum for my smaller rockets. I've lost a lot of rockets and I'm hoping the telemetrum will help change that. I just plan to put it in my Lv. 1 cert to see how well it works. My cert flight is simmed to go around 1100 feet.

 

[keithp]

We walked around for over two hours including searching for the beacon from the Telemetrum, with no luck.
If the same error applies to the right of latitude then hdop (horizontal dilution of precision) was 95. This would indicate a very poor accuracy and could easily account for why I didn't find the rocket.

The DOP values in your firmware version are scaled by 10 (I missed the u-blox documentation that defined the units for that value); with some newer AltosUI bits, I'm seeing a PDOP of about 1.6, which is pretty darn good and should represent a positional accuracy of less than 20m.You can see that the GPS height values are pretty close to the barometric height values during descent, so the lat/lon should be very close as well. I've helped find a number of ballistic returns, and the GPS data is usually pretty much spot on if you've got data from that close to the ground. About the only source of error I can imagine would be a mistake transcribing the coordinate value or a position representation mismatch, but I suspect you checked both of those several times...

The map shows the last reported position about 185m due S of the windmill by he stock tank, and about 510m west of the road that goes by the flight line. I've buried rockets in that dirt, but there's always something sticking above grade that you can see from quite a distance away.

 

[Lowpuller]

I agree with Bob K. very common mistake.


So many rockets, so little time, and money.

 

[tedapke]

Thanks Keith.
Do you know which column has the hdop in that file I posted? Is in 2.something or 9.something or something all together different?

Your interpretation of where the rocket should have "landed" matches where I searched. Colin first used a Garmin and I didn't see the values he entered but we were searching the same areas I looked at on Tuesday. I used my iPhone (man is that hard to see in the sunlight) and had it set to decimal degrees which seemed to match the file.

Just to double-check, voltage on main/drogue going from 4 to 0 is an indication that the match fired and has gone open, correct?

Thanks again. As you know, I love the Telemetrum and am very surprised that I can't find this rocket!

Ted Apke
TRA2726 L3


Sent from my iPad using Rocketry Forum

 

[troj]

The challenge may be that you're looking for a hole, instead of something above ground.

Years ago, I had a rocket go about 14,000 up, and 14,006 feet down. This was in Argonia, where the ground is rock hard.

We searched for hours, assuming we were looking for something above ground. Finally, after about 4 or 5 hours, I went back to the last GPS point and started again...only to come across a 6" diameter hold in the ground with three slots 120 degrees apart, about 150 feet away from that point.

Go back to your last GPS coordinate, and start looking for holes.

-Kevin

 

[bobkrech]

Before all of the modern GPS location systems existed, White Sands used Missile Dogs to find lost rockets...with a 96% recovery rate...

https://www.wsmr-history.org/Dogs.htm

Bob

 

[keithp]

Do you know which column has the hdop in that file I posted? Is in 2.something or 9.something or something all together different?

Your interpretation of where the rocket should have "landed" matches where I searched. Colin first used a Garmin and I didn't see the values he entered but we were searching the same areas I looked at on Tuesday. I used my iPhone (man is that hard to see in the sunlight) and had it set to decimal degrees which seemed to match the file.

Just to double-check, voltage on main/drogue going from 4 to 0 is an indication that the match fired and has gone open, correct?

Thanks again. As you know, I love the Telemetrum and am very surprised that I can't find this rocket!

The HDOP value is the decimal number just before the 32 satellite C/N0 numbers, starting at 96.0. Those are multiplied by 100 in your version, so the actual value is 0.96. And, yes, the drop in igniter voltage indicates that both charges fired as programmed. Note that the voltage doesn't always drop to zero as many igniters don't go open when fired. They do usually change in resistance, so you can generally tell when they've fired. Mysterious that you can't find the rocket. Kevin does have some experience in this matters, so he may be right and you may be looking for only a small hole in the ground, and those can sometimes be surprisingly hard to find, even in clear terrain.

 

[JimJarvis50]

The challenge may be that you're looking for a hole, instead of something above ground.

-Kevin

I had one like that a few years ago. I looked under the last gps reading and couldn't find anything. Then, I went and projected the angle from the last several readings. That moved the impact point 70 feet or so. After a few minutes, we found the hole. I learned that day that it is easier to remove the dirt from a hole than to put it back.

Jim

 

[ksaves2]

Like others have said, make sure one has their units correct when punching in coordinates. Going from decimal degrees to degrees/minutes/decimal minutes or
degrees/minutes/seconds can make a bit of a difference.

Since good data was received, as was pointed out, wasn't a paint issue. If the rocket was dual deploy and the Telemetrum showed both charges fired, pretty weird
situation if they both failed to "do something" to the rocket. Only thing that could account for a pure ballistic flight would be dual ematch failure or expected separation(s) didn't occur?

If the apogee was nominal, was the rate of descent consistent with ballistic descent or under drogue? If the rocket was under drogue or drogueless, would be two pieces to find perhaps in close proximity?

Bad battery? Doesn't explain why you got good telemetry just about all the way to the ground. If there was a drogue chute and it was windy, could be the final resting place is a fair distance away from where you received the last packet. If you remember the wind direction, that would be where I'd start looking.

If the main fired successfully, I would have expected a possible broken shockcord if the apogee charged failed to separate the rocket and it was coming in ballistic until main deployment.

I had a modified DD Talon 2 with a Beeline GPS riding on top in the ebay and and a Parrot altimeter riding behind it in the sled. Stupidhead didn't do an adequate
range test and the nice metallic paint sucked up the Rf during the I200 ride up. I had monkeyed with the apogee ematch at the pad because I forgot the Parrot didn't do an adequate aural warning. It gives the same continuity indication whether one or both ematches are connected. I reconnected the apogee charge but missed one leg of the ematch wire! Rocket went up 3k', no apogee charge, ballistic descent with no APRS packets (metal paint, learned my lesson). Once the main altitude was reached, pop goes the main and rip goes the shockcord. Nosecone must've smacked a fin because there was an elliptical slice in it and surprisingly the main chute looked brand new. I found the nosecone and the chute errantly fluttering down at the launch. Couldn't find the rocket just then.

Landowner gave me the remains a year later. I blew the apogee charge with my launch electronics at home. It worked. It was still with the remains when found. Yeah, I missed the connection at the terminal when I was jiggering it at the pad! The BLGPS was nowhere to be seen. The Parrot survived as I gusseted the bulkhead that separated it and the BLGPS royally. The piezo was cracked and I replaced it and the battery. The Parrot dutifully downloaded the near ballistic flight. About 83G's longitudinally and ~45G's laterally. I don't know if that is accurate but that's what it reported. To say the least, it hit hard. I believe the long upper tube hit and the rocket broke and hit flat. I lengthened the T2 a bit to have room for a main chute.

Same day I launched a stretched Wildman on an L1400 except all the events occurred nominally. One APRS packet received at 6000' only
as it too was metallic painted!!! I lucked out as the rocket came down within eyeball distance and got it back. 9500'. Interestingly, the Beeline GPS irregardless of the metal paint, had a lock on 7 to 9 satellites and I downloaded a perfect .kml file. I didn't turn the BLGPS off until I got home and it showed where I stopped off for an after launch pizza at a nearby resturant before the memory filled up! The paint only stopped the Rf from the transmitter. Best of luck with finding your rocket.
Makes it harder to figure out what went wrong without it. Kurt

 

[matthew]

Telemetrium for level 1 cert? Keep it simple. I would recommend standard recovery and a low and slow approach. It you need a telemetrium for level 1, I would recommend lower flight.

If someone wants to fly a minimum diameter I to 8000' @ M1.5 and use DD and a tracker for L1, why shouldn't they? Once successful, they'll be allowed to do so anyway.

 

[Area66]

this rocket land dart for sure

 

[ksaves2]

If someone wants to fly a minimum diameter I to 8000' @ M1.5 and use DD and a tracker for L1, why shouldn't they? Once successful, they'll be allowed to do so anyway.

It will be impressive no doubt but leaves a lot to go wrong. I did a rapid L2 with a 3fnc I came up with in Rocksim.
J350 and motor ejection. Works perfect. I got 'er done so I could fly H and I motors at Ex launches. Rules issue at that time (2007)
I then did small DD stuff and apogee only MAD flights with H and I motors to get skill. To each his own. Kurt

 

[tedapke]

Thanks all.
It does sort of look alien out there. Wouldn't call in the moon as it is pretty green this year. Something I haven't seen, in August, since I moved out here more than 10 years ago.

After Colin and I looked the first time and verified that the rocket was't sticking up anywhere in the general area of the last GPS we concluded that it must have buried itself, hence the aforementioned "metal detector". Unfortunately I couldn't get the metal detector to stop reporting buried treasure. I didn't get an instruction manual so I played with sensitivity and something labeled "discriminate" with no apparent change in results. I ended up using it only to help rule out suspicious holes, of which there were only a handful. None of those holes had fin slots.

I did do some math before I headed out for the second search effort and concluded that the impact point was not more than 10 meters from the last GPS (it was coming down pretty straight).

Until Keith pointed out the HDOP was .95 my best guess was going to be the one are of tall grass in the area about 50m North of the last GPS, inside of some barb wire surrounding the windmill/water pump. Colin did search the area on Sunday with no luck but that's about all I got.

Probably try again next launch.




Ted Apke
TRA3726 L3