Eggfinder Problem

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Some GPS trackers will still transmit Rf if they lose a lock from the GPS satellites. Some won't. If they stop transmitting an Rf carrier when they lose a GPS lock, one can't resort to using RDF techniques that could save the recovery. I actually experimented with a 900 Mhz GPS tracker (Eggfinder actually) and found a 900Mhz Yagi did provide a directional fix with garbled packets right at the limit of the Eggfinders ground footprint. Was an easy peasy thing to test as I had a visual direction where the rocket went down and lost the Rf link. (Can do this with a ground test too with a large enough field.) The Yagi did point in the right direction when I got closer before viable positions came in. If the GPS antenna was in the dirt, as long as Rf energy was being transmitted to swing an Rf meter, I could have done an RDF recovery from there on out.
Most say that with 900Mhz RDF isn't possible but in the terminal phases of recovery of a rocket "on the ground" where you know the general whereabouts of the rocket, it is highly DOABLE.
I have to be honest I have not ever had a totally sight unseen rocket with a 900Mhz tracker go up/down. Yeah, lost 'em on the up side but someone would see 'em coming down. If sight was lost during the whole flight, I'd use my best guess and go out using my 900Mhz Yagi as it does indeed improve the ground footprint. No attenuator is required on this band. On 400Mhz or 144Mhz an attenuator is helpful for RDF but if flying a GPS tracker is not needed generally. My 400Mhz and 144Mhz handheld Yagis have attenuators on them as they are essential for pure RDF tracking. The ground footprint is larger on 70cm and 2 meter bands. On 900Mhz not so great a footprint but good enough.
If serious about 900Mhz tracking/recovery, one should invest in a 900Mhz Yagi to have onsite just in case. If sport flying where one "sees" where the rocket is landing be it in corn or tall grass with a GPS tracker a Yagi is likely not needed though as I said it does increase the ground footprint and helps if one knows the general direction but doesn't see where the rocket came down.
Sure, if flying on the playa out west in the wide open spaces, a Yagi might be more needed as the salts on the salt flats suck up Rf energy like a sponge and folks tell me the tracker ground footprints are ridiculously small on the ground. That last known position is the key along with knowing the drift trend to get a rocket back. When one knows the way, they can point the handheld Yagi in the right direction to maximize the prospect of getting a ground signal and the rocket back.
With the "dirt" in the cornfields of the Midwest, we have better ground radio footprints so recovery is easier. Plus we don't fly as much totally sight unseen flights as we want to stay within the constraints of the launch sites we fly as lest we piss off the "neighbors".
Kurt
 
Last edited:
At Airfest, I used the eggfinder tracker with my android phone using the rocket locator app, and as with my OP above - the tracker showed me the location on the phone and I walked to the location and there was no rocket at that location. So then I started walking around the location all the while checking the tracker location and after 15 -20 minutes of searching, the tracker suddenly showed a different location for the rocket - in one case about 150 feet from the initial location and in the other case about 500 or more from the original location. The entire time I was receiving a signal from the GPS transmitter. I turned off the phone network so that the phone had to use GPS and could not triangulate cell towers.

If the GPS receiver has lost lock, does the transmitter continue to transmit the last lock location? When outside, it normally doesn't take 15 minutes to establish lock - so why would it take so long after a launch?

Plus for each launch I could clearly see the descent pattern for a rocket under parachute - so it had a position lock during the descent - could it lose lock on impact and then why would it show a position that is 150 feet or more from the last lock position? and the at the wrong location that wasn't even along the descent path!

I have a log file from the rocket locator software, but I cannot see how to decode the contents - the log file is not ascii text and so I cannot determine if it contains NMEA data. - it is a .APK file - does anyone know how to look at the log file for this program?
Did you try putting the coordinates into a navigation app to see where it takes you? If you got coordinates, they're good and in real-time... it won't transmit "bad" coordinates.
 
Did you try putting the coordinates into a navigation app to see where it takes you? If you got coordinates, they're good and in real-time... it won't transmit "bad" coordinates.

I did not try this at the time. I should have - I'll have to try it next time.

I do have the log file so if there are coordinates in the log file I could pull them out and put them in a navigation app. Unfortunately I still haven't yet been able to decode the log file.
 
I wonder if the issue is precision. I notice if I only am seeing 3-4 satellites the resulting
GPS coordinates are off by a lot. Sometimes 50 yards. If I’m seeing 7-8 or even more the result is very close.

Before I go to the pad I always look at the broadcast coordinates on my phone to see if they are accurate. If not, I wait until more satellites are acquired. Where we launch in Ohio, I routinely see over 10 (* on LCD).

Do you remember how many satellites you were seeing when you took the rocket to the pad? Assuming you were using a LCD receiver of course? I fire my mini up a good 15 minutes before I head to the pad. I notice that the mini takes longer to acquire satellites than the TX.

I agree the coordinates that are transmitted are what the GPS says they are. How precise that reading is depends on number of satellites. I think that’s an important piece of the puzzle.
 
I did not try this at the time. I should have - I'll have to try it next time.

I do have the log file so if there are coordinates in the log file I could pull them out and put them in a navigation app. Unfortunately I still haven't yet been able to decode the log file.
I had hacked two instances of GPS Rocket Locator so I could display my local position on a map along with the rockets position. Not that hard to do but even if you just show the rocket, make sure you zoom in close on the final position. That helps a lot. Kurt
 
I do have the log file so if there are coordinates in the log file I could pull them out and put them in a navigation app. Unfortunately I still haven't yet been able to decode the log file.
I think you are looking at the wrong files. On my phone the "Rocket Locator" log files are stored in the path shown here. Note each file is named "log_datetime.log
Screenshot_20220912-211920_My Files.jpg

Opening that file which just uses a log file viewer or text viewer. Gives the data streams from the received RF info. These are standardized strings which give aside for type of data, coordinates, #of satellites, quality of fix etc...
Here is header and first few lines of data...
Screenshot_20220912-212119_Log Viewer.jpg
 
I wonder if the issue is precision. I notice if I only am seeing 3-4 satellites the resulting
GPS coordinates are off by a lot. Sometimes 50 yards. If I’m seeing 7-8 or even more the result is very close.

Before I go to the pad I always look at the broadcast coordinates on my phone to see if they are accurate. If not, I wait until more satellites are acquired. Where we launch in Ohio, I routinely see over 10 (* on LCD).

Do you remember how many satellites you were seeing when you took the rocket to the pad? Assuming you were using a LCD receiver of course? I fire my mini up a good 15 minutes before I head to the pad. I notice that the mini takes longer to acquire satellites than the TX.

I agree the coordinates that are transmitted are what the GPS says they are. How precise that reading is depends on number of satellites. I think that’s an important piece of the puzzle.
You need to put a noisemaker on your apogee or main harness that puts out a pretty load sound. You get within 50 feet and your ears make an excellent terminal tracker. Until I did that, I would wander around awhile while tracking. Works perfectly. Put a noisemaker in there along with a GPS tracker and you'll be in great shape.

I mastered porting the incoming positions to either GPS Rocket Tracker or other Ham radio APRS tracking software so I can plot a rocket track on a live map in real time without manually inputting data. Makes life so much easier and I did it over 10 years ago. Posted here about it and I don't think anyone knew what the heck I was talking about.

Since I have a General Ham Radio license, I invested in a Beeline GPS tracker early on and a Garmin D72A. I could port the output from the D72A to a Garmin 60Cs or 60 CsX mapping GPS with a single cable and plot the rocket on the mapping screen of the Garmin directly with the interface. Had a map in hand and had no problem recovering rockets. Get to the last known position and likely the rocket is nearby and will give you a new coordinate if you don't see it by then. Sometimes one has to really "Zoom-in" on the map and a noisemaker on the harness really helps in "fine tuning" the recovery.

Alternatively, port it to a laptop (been there, done that) or a large screen Android device.

Guys this stuff has been out there for I believe 20 years or so. Ever since APRS on the Ham side which was first and now the NMEA no license required 900Mhz trackers. Granted it cost quite a bit of money for ARPS, GPS tracking in the early days but shoot it was better than the "Rocket Tracker" and "Walston" RDF systems even back then. Shoot they charged an arm and a leg for the RDF stuff and just by getting a Ham Technician license one could have done APRS/GPS tracking for just about the same amount of money. Even back then.

Man if you can get your output automatically on a map, you'll be much better off rather than "inputting by HAND the last known position" into a hand held GPS with all the issues with "units". I've seen guys use the wrong units go on wild goose chases. Get a system that automatically takes your positions and puts them on a map for you. I can try to look up my past posts that outlined that and repost them.

I don't know if the Android program GPS Rocket Locator is still available. With that program, I was able to batch download photomaps of the flying sites I went to in various zoom modes so I actually didn't have to have an internet connection when out in the field.

With that program, I actually had a color aerial photo of the ground with the rocket track and my position. I could easily see impediments to a recovery and sometimes a straight line is not the best way to go to recover a rocket.

Even using other Ham Radio tracking programs like Linux Xastir or there are some other Windows programs out there, one can have a map in hand on a laptop or small tablet.

Two instances of GPS Rocket Locator can give your local position and your rocket's last known position and make it so much easier to walk or drive to your downed rocket.

Mapping GPS programs have been out there for years but indeed the learning curve was quite steep. Hope more take the time to use it because once over the hump, it makes it so much easier to recover sight unseen flights and go on to fly your next rocket. Kurt KC9LDH
 
Thanks Tractionengines!

Ok I found the log files and they indicate the exact issue I am talking about. The red mark is from the NMEA data, the white line ends at where I actually found the rocket. The NMEA data indicates that I had lock with 8 satellites, and HDOP was 1.5. The rocket was launched from the west so the rocket was never over the red mark so it could not be the last valid lock position.

You can see that the rocket was actually 500 feet further to the west than the GPS location from the tracker. What I still don't know is how to interpret the HDOP data - what does HDOP of 1.5 actually mean in regards to location uncertainty? - the information I found says HDOP of 1 to 2 is excellent so why the 500 foot discrepancy in position?

Screenshot (1).png
 
Some GPS trackers will still transmit Rf if they lose a lock from the GPS satellites. Some won't. If they stop transmitting an Rf carrier when they lose a GPS lock, one can't resort to using RDF techniques that could save the recovery. I actually experimented with a 900 Mhz GPS tracker (Eggfinder actually) and found a 900Mhz Yagi did provide a directional fix with garbled packets right at the limit of the Eggfinders ground footprint. Was an easy peasy thing to test as I had a visual direction where the rocket went down and lost the Rf link. (Can do this with a ground test too with a large enough field.) The Yagi did point in the right direction when I got closer before viable positions came in. If the GPS antenna was in the dirt, as long as Rf energy was being transmitted to swing an Rf meter, I could have done an RDF recovery from there on out.
Most say that with 900Mhz RDF isn't possible but in the terminal phases of recovery of a rocket "on the ground" where you know the general whereabouts of the rocket, it is highly DOABLE.
I have to be honest I have not ever had a totally sight unseen rocket with a 900Mhz tracker go up/down. Yeah, lost 'em on the up side but someone would see 'em coming down. If sight was lost during the whole flight, I'd use my best guess and go out using my 900Mhz Yagi as it does indeed improve the ground footprint. No attenuator is required on this band. On 400Mhz or 144Mhz an attenuator is helpful for RDF but if flying a GPS tracker is not needed generally. My 400Mhz and 144Mhz handheld Yagis have attenuators on them as they are essential for pure RDF tracking. The ground footprint is larger on 70cm and 2 meter bands. On 900Mhz not so great a footprint but good enough.
If serious about 900Mhz tracking/recovery, one should invest in a 900Mhz Yagi to have onsite just in case. If sport flying where one "sees" where the rocket is landing be it in corn or tall grass with a GPS tracker a Yagi is likely not needed though as I said it does increase the ground footprint and helps if one knows the general direction but doesn't see where the rocket came down.
Sure, if flying on the playa out west in the wide open spaces, a Yagi might be more needed as the salts on the salt flats suck up Rf energy like a sponge and folks tell me the tracker ground footprints are ridiculously small on the ground. That last known position is the key along with knowing the drift trend to get a rocket back. When one knows the way, they can point the handheld Yagi in the right direction to maximize the prospect of getting a ground signal and the rocket back.
With the "dirt" in the cornfields of the Midwest, we have better ground radio footprints so recovery is easier. Plus we don't fly as much totally sight unseen flights as we want to stay within the constraints of the launch sites we fly as lest we piss off the "neighbors".
Kurt
I think to effectively use the 900 MHZ signal as a radio beacon you would need a way to adjust receiver sensitivity.
 
I wonder if the issue is precision. I notice if I only am seeing 3-4 satellites the resulting
GPS coordinates are off by a lot. Sometimes 50 yards. If I’m seeing 7-8 or even more the result is very close.

Before I go to the pad I always look at the broadcast coordinates on my phone to see if they are accurate. If not, I wait until more satellites are acquired. Where we launch in Ohio, I routinely see over 10 (* on LCD).

Do you remember how many satellites you were seeing when you took the rocket to the pad? Assuming you were using a LCD receiver of course? I fire my mini up a good 15 minutes before I head to the pad. I notice that the mini takes longer to acquire satellites than the TX.

I agree the coordinates that are transmitted are what the GPS says they are. How precise that reading is depends on number of satellites. I think that’s an important piece of the puzzle.
see my post above - the data shows I had 8 satellites with HDOP of 1.5

Randy
 
@rjenkins133 If you want to meet up at one of the NE Ohio MetroParks, send me a PM with your contact info. I can bring my Eggfinder TX, Mini, LCD Receiver, and BT feed to cell phone.

I am still still learning the system too, but I have been around GPS systems for many years... Earlier in the summer, having my kids play "hide-and-seek" with the TX in a nosecone, left me get a feel for the range, software, system, etc. While not a match for a real flight profile. I did learn how close is "close", what is ground range, how fast/far to move to get LCD update, etc. Then took the log file from the OPENLOG module on the TX and compared it to the log file in Rocket Locator App with Google earth.

(I used to do a lot of hiking in the Allegany National Forest in PA. There I needed to know every detail of my GPS. Plus how to read a map and compass if GPS died )
 
see my post above - the data shows I had 8 satellites with HDOP of 1.5

Randy
One simple question is where were you located on the map when you lost the signal when the rocket was coming in under main chute?
If the rocket is so far away when it transmits the last known position or it is out of range of the receiver, it might drift farther to touchdown than what your position on the map shows.
Another thing is a lighter rocket can be dragged across the ground while you are out of range.
Did you get an updated position after you started your ground search that showed the rocket was 500 feet west?
If the rocket wasn't there at the last known position, I would have proceeded in the direction of the current prevailing ground winds. It has been pointed out if the GPS receiver antenna is facing the dirt, positions may be lost or garbled. The prospect of that is pretty remote though. You should to start to receive an updated position as you get closer and into the trackers ground footprint. Kurt
 
I think to effectively use the 900 MHZ signal as a radio beacon you would need a way to adjust receiver sensitivity.
A Yagi is a roundabout way to do that as it increases the incoming signal strength. Problem with a 900Mhz Yagi so I'm told is the beamwidth of the receive or for that matter transmit lobe (antenna pattern) is supposed to be very narrow and in the direction you point the antenna. (Of course am only using the antenna to receive here.)
The narrow beamwidth 900mhz Yagi is not that difficult to point at a stationary transmitter on the ground.
I've proven this to myself several times. ie. a downed rocket.
On a downed rocket where I knew the general location by last known position, I'd stick the Yagi on the Eggfinder LCD and point it in the general direction of the rocket holding the elements horizontally. That has to do with antenna/signal polarity. Just as soon as I started to get a signal from the downed rocket, I screwed the little stock vertical back on and the signal disappeared.
It's not a big deal since I had a general idea where the rocket is on the map. The stock antenna is going to start picking up the final positions as I walk closer anyways. Don't get jerked about a Yagi as most sport fliers don't need one on 900Mhz. I found a 7 element one that was very economical and just wanted to experiment. I was aware that a Yagi wasn't recommended for in flight tracking during high speed flight before I got one.
I don't use the Yagi for in-flight tracking on 900Mhz for the reason stated above about beam width. I use the stock antenna.
On the 70cm and 2 meter Ham bands, the beamwidth of those Yagis are wide enough so you will see people using them and pointing at the rocket for in flight tracking. For ground use on the Ham bands (70cm and 2 meters) an electronic attenuator needs to be used for RDF work as the receiver gets saturated when one gets close.
The 500 foot discrepancy is still pretty odd and sounds like the rocket landed quite a bit out of sight from the launching area?
I've always mentioned for folks trying out a GPS tracker for the first time, don't punch the rocket up into thin air but use a motor that will keep it in sight for most of the time. You can get a good test of your tracker systems, transmit and receive and if a glitch occurs, since you've kept the flight visual, you can make a visual recovery and get your rocket back and all the "stuff" back to work out the bugs.
When walking to a recovery site of a 900Mhz tracker just at the fringe of the radio footprint, I can get a null without an attenuator on the Yagi as to the low signal strength due to distance. Get a few steps closer and the receiver will pick up the positions and the LCD will show me where it's at as I have it connected via Bluetooth with a device with a live mapping program on Android or Window programs. I've used the Yagi right up to the recovery site though in an open field I can see the rocket from quite a distance away. In tall grass or corn, that's where a harness noisemaker comes into play. The louder the better. Kurt Savegnago KC9LDH
 
@rjenkins133 & @Zbench I saw the email that the next NOTRA Launch is tentatively this Sunday, pending weather cooperation.

I am hoping to come out and fly the Der Red Max HEAVY again. If you will be there we can "compare" position data from various GPS sources. ( I will need to leave early to get to a family gathering. )

Mike
 
@rjenkins133 & @Zbench I saw the email that the next NOTRA Launch is tentatively this Sunday, pending weather cooperation.

I am hoping to come out and fly the Der Red Max HEAVY again. If you will be there we can "compare" position data from various GPS sources. ( I will need to leave early to get to a family gathering. )

Mike
I’ll bring mine! We can put them all together and see what’s what.
 
I tested mine today for this weekend. I gave the nosecone with TX to one of the shop guys and said go put it near one of the cars on the parking lot, while I sat at picnic table out back.

This is what I got from LCD and Rocket Locator after starting up the app. The blue dot is where I waited, the Red marker is that BT module told App the rocket was at.
20220915_101850.jpgScreenshot_20220915-101803_Rocket Locator.jpg

Then this is the LCD screen and APP screen when I walked to the red locator in the first image. I didn't have a lot of time or space to spread out, but is was very close to actual location. (Within 1 parking space distance.)
20220915_102200.jpgScreenshot_20220915-102141_Rocket Locator.jpg

I think I am good to go... just didn't test the mini transmitter yet.
 
NOTRA launch for Sunday canceled. (Moderate Wind, but blowing toward the highway.)

Guess we'll have to wait to compare location info. Or get together outside of a launch.
 
I have done some additional research on GPS accuracy. It may supply some answers on the situation I described above. I had assumed that the error in tracking location would be roughly circular around the GPS tracking device. It turns out this is not the case. The shape of the error is dependent on the location of the identified satellites the GPS chip is using. so if the satellites being used are grouped together toward the horizon in the eastern or western sky then the north-south error band will likely be larger than the east-west error band.

In my case, the tracker eventually shifted the rocket location position to a location much closer to the actual rocket location on the ground, I think this happened because the receiver eventually added a satellite that allowed the error to be reduced.

I still have not found a formula to convert the HDOP value into error measurement value, perhaps other than HDOP of 1 would likely be close to the 2.5 meter error specified in the GPS chip data sheet. the log for my flight showed an HDOP of 1.5 to 2 - which translated to an tracking error of about 500 feet to the east.


So - the key take-aways for me:

1. This level of error will likely continue as it is inherent in the design of this GPS tracking system. It is controlled by the number and location of satellites.

2. Accuracy could be improved by selecting chips that support concurrent reception of GPS, GLONASS, Beiduo and Galileo signals as this increases the number of available satellites and the more satellites the receiver can find, the better chance for tracking accuracy to improve.

3. Accuracy could be improved by selecting chips that support dual channel receivers as this reduces errors caused by signal degradation from the atmosphere and reflections.

4. Both 2 & 3 above require the use of more expensive (and probably harder to get) receiver modules so I think this is unlikely to happen. Therefore - I will likely be moving to a RF beacon based tracking system.

Randy
 
I have done some additional research on GPS accuracy. It may supply some answers on the situation I described above. I had assumed that the error in tracking location would be roughly circular around the GPS tracking device. It turns out this is not the case. The shape of the error is dependent on the location of the identified satellites the GPS chip is using. so if the satellites being used are grouped together toward the horizon in the eastern or western sky then the north-south error band will likely be larger than the east-west error band.

In my case, the tracker eventually shifted the rocket location position to a location much closer to the actual rocket location on the ground, I think this happened because the receiver eventually added a satellite that allowed the error to be reduced.

I still have not found a formula to convert the HDOP value into error measurement value, perhaps other than HDOP of 1 would likely be close to the 2.5 meter error specified in the GPS chip data sheet. the log for my flight showed an HDOP of 1.5 to 2 - which translated to an tracking error of about 500 feet to the east.


So - the key take-aways for me:

1. This level of error will likely continue as it is inherent in the design of this GPS tracking system. It is controlled by the number and location of satellites.

2. Accuracy could be improved by selecting chips that support concurrent reception of GPS, GLONASS, Beiduo and Galileo signals as this increases the number of available satellites and the more satellites the receiver can find, the better chance for tracking accuracy to improve.

3. Accuracy could be improved by selecting chips that support dual channel receivers as this reduces errors caused by signal degradation from the atmosphere and reflections.

4. Both 2 & 3 above require the use of more expensive (and probably harder to get) receiver modules so I think this is unlikely to happen. Therefore - I will likely be moving to a RF beacon based tracking system.

Randy
Randy,

I guess those are all possibilities, but seems unlikely to me. I’ve been using mine for years, also close to you ( in Ohio) and see the same satellites you are seeing and have never had an error of more than 10 feet ever. I don’t know anyone else that has either. Have you ruled out that you aren’t on the same frequency as Someone else and are actually seeing their tracker not yours? That seems like the only likely reason to me.
 
The other issue is time..

1) the GPS chip starts "cold" every time. (It doesn't remember last location or anything at power up. So make sure to turn it on early. The more time it has to find it's location the better. As it then finds more satellites to bring any location errors down.

2) if the chip loses lock on the flight, it takes some time ( hopefully less than 1 above) to get back to high accuracy position. (It knows where it was when it lost lock.)

In my case, the tracker eventually shifted the rocket location position to a location much closer to the actual rocket location on the ground, I think this happened because the receiver eventually added a satellite that allowed the error to be reduced.

I think your recovery was "close enough" (the NOTRA field is small after all ) that you were walking toward recovery and getting position updates, but they just hadn't been "refined" yet.
 
The other issue is time..

1) the GPS chip starts "cold" every time. (It doesn't remember last location or anything at power up. So make sure to turn it on early. The more time it has to find it's location the better. As it then finds more satellites to bring any location errors down.

2) if the chip loses lock on the flight, it takes some time ( hopefully less than 1 above) to get back to high accuracy position. (It knows where it was when it lost lock.)



I think your recovery was "close enough" (the NOTRA field is small after all ) that you were walking toward recovery and getting position updates, but they just hadn't been "refined" yet.
Mike,

I agree with all that. A further data point. On my level two certification flight earlier this summer, I equipped a TX with a data logger just for fun. The data I got back was very confusing. I finally realized that the gps lock was immediately lost on launch, but, I got it back when the rocket was under descent about 2000 feet from the ground. I was expecting to see one or two points straight up and then a slow arc coming down. What I got was the tail end of the decent arc.

Point is, it reacquires the location pretty quickly when in the air.

Finally, when I build a new mini or TX, I test them in my backyard. Even when it only has a 3 satellite lock it’s within 75 feet, 4 without 30, 5 or more within 10. This has been consistently my experience no matter if it’s the old or new GPS chip. I can’t wrap my mind around a 500 foot difference.
 
Randy,

I guess those are all possibilities, but seems unlikely to me. I’ve been using mine for years, also close to you ( in Ohio) and see the same satellites you are seeing and have never had an error of more than 10 feet ever. I don’t know anyone else that has either. Have you ruled out that you aren’t on the same frequency as Someone else and are actually seeing their tracker not yours? That seems like the only likely reason to me.

I have ruled out others on the same frequency.

1. I investigated this possibility while in Argonia - I used the method Cris recommends and I never saw another transmitter the entire time I was there. It seems pretty unlikely that another transmitter was on only when mine was on and turned off only when mine was off - on three different days.

2. More importantly - the data I posted previously was the actual NMEA GPS data that the rocket was at a location that it clearly was not close to. I was 3000 feet from the flight line when this data was recorded. There was no other rockets close to me so it seems very unlikely that it was another Eggfinder GPS transmitter near me - also if there was and the data was accurate - I would have found the other rocket!

The location accuracy is completely dependent on the number and position of the satellites the GPS receiver sees. Maybe the GPS module you have is more sensitive or mine is less sensitive. The suppliers often source GPS modules from different vendors. In the last year, I have gotten two GPS modules from eggtimer that are different manufacturers, probably with slightly different performance characteristics.

My receiver locator TX came with a Lantronix A2235H GPS module - which one do you have?

Based on the limited log files I have stored, I typically get 8-9 satellites and an HDOP of 1.5 - can you tell me what your receiver typically sees? If you consistently get more satellites and lower HDOP you will have better positional accuracy - but you probably had no reason to look at this data since your tracker always takes you to the rocket.

I am certainly open to other explanations but so far I have not gotten an alternative one that adequately explains the behavior of my tracker.


Randy
 
Last edited:
All,

I just did a little experimentation in the backyard. The test consisted of 3 devices, a Eggfinder TX with the older chip you had to solder on yourself, an Eggdinder Mini with the newer presoldered gps chip, and an Eggfinder LCD. I turned on the LCD and let it warm up and acquire a GPS fix (I have a GPS card on board). I then switched on the TX and closely watched the receiver and took pics of the screen each time the satellite number changed. The log file for this event is here. I annotated it with a XXXXXXXXXXXXXXXXXXX444444444444 for each line where the number of satellites first changed. The receiver picked up 4 satellites in 50 seconds which was pretty quick. Every time the view changed on the receiver, I took another picture with my camera so I could analyze them later. I did this because I prefer the digital format that the receiver uses as opposed to the raw data that can be found in the log file. Since the log file records the time stamps for each event, total elapsed time was 2 minutes and 51 seconds from 4 to 7 satellites.

I then turned on the receiver, switched it back on, let it acquire GPS lock, and then repeated the same experiment with the mini. It was fast! I had 8 satellites in 1:15 and after 5 minutes didn't see it incrementally add any more, so I switched it off. I only saw 3 entries on the receiver, it came out of the gate seeing 5 satellites.

I took all that data and created a google earth project. The yellow locations in the picture below is the TX locations in the order in which the satellites were received. So, 4 SAT shows the location when the receiver was picking up 4 satellites. The TX saw a max of 7. The green locations are for the Mini, only saw 3 entries on the screen and they were very quick as noted above.

The black X shows the true location of the GPS receiver, which was laying on a granite surface near my outdoor grill. I placed both the TX and Mini in the same spot. For distance scale, the black line represents the distance between two stone pillars in the back yard. The distance between them was measured to be 8 foot 9 inches.

What I take away from this is for my equipment, the most inaccurate reading for the TX was still within 25 feet of the true location. the HDOP as can be seen in the log file started at 2.6 with 4 satellites and was reduced to 1.3 when seeing 7.

Randy, you might consider recreating this same experiment in your own back yard and reporting what you get. This kind of accuracy is routine in my experience for the kits I've constructed. Interested in your thoughts and results should you chose to do it,

Eggfinder_Analysis.jpg
 
I will recreate your test and see that I get this week. Both my TX and Mini are older chips I soldered on. My LCD also has the GPS module, on a BOARD I soldered together.

I expect similar based on the test I did last week in the parking lot, but both had solid fixes, and I was looking to see how far the LCD appeared to be away.

I like the idea of seeing how position accuracy changes with time and number of satellites.
 
I'll try it as well - I am still building my GPS LCD receiver - just powered it on this morning and it seems to be working. I need to mount it in a box and will then run the test.
 
OK... Interesting test, and comparison to @Zbench I will be looking forward to your testing @rjenkins133 (NO RUSH.) LOL

I DIDN'T LOG the data, but I did take photos of the LCD Screen each time it updated with new # of satellites. My Process was Start the LCD. Wit for it to get IT'S Location, and screen change to "Waiting for FIX". Then I Connected Battery to MINI (or TX), and took photos each time # of satellites updated. Once it stopped adding in satellites, I checked position at 5min marks to see if it continued to refine the position. (It did for the first 5min check, but didn't after that. BUT NOTE: The Mini did pickup a satellite [from 6 to 7] at about 15minutes, and that improved accuracy. It then stayed the same for next 20 minutes.

Here is the MINI shown in the 2 images below. Blue Pin = Actual, Yellow Pins GPS

First Fix, Pin1, 3 Satellites, Error 309ft, less than 1 min after start
Pin2, 5 Satellites, Error 28ft, less than 1 min after start
Pin3, 6 Satellites, Error 17ft, 1 min after start
Pin4, 6 Satellites, Error 6ft, 5 min after pin 3
Pin5, 7 Satellites, Error 3.5ft, 12min after pin 3 (13min after start)
(No change for next 20 minutes.)

Mini_LCD_Pins_1-5.JPGMini_LCD_Pins_2-5.JPG

Here is the TX shown in the 2 images below. Blue Pin = Actual, Red Pins GPS

First Fix, Pin1, 3 Satellites, Error 754ft, less than 1 min after start
Pin2, 4 Satellites, Error 61ft, less than 1 min after start
Pin3, 8 Satellites, Error 28 ft, less than 1 min after start
Pin4, 8 Satellites, Error 12 ft, 5 min after Pin3
Pin5, 8 Satellites, Error 11 ft, 10 min after Pin3
(No change for next 20min)

TX_LCD_Pins_1-5.JPGTX_LCD_Pins_2-5.JPG

My take away from this is get the GPS Receiver powered up for ~20 minutes prior to launch.

I want to do one more test. I will repeat above with the MINI. But at the 30 min. mark after powering it up. (With hopefully stable position, like this test.) I will bring it inside & shield it, so that it looses reception to satellites. Then after 5min take it back outside to reacquire data. Document the same information from time out of shielding. I want to see how it adds satellites, and refines position after loosing it, while still powered up.
 
Mike,

Seems very consistent with my results. It is interesting to note that in both cases, in less than 1 minute you are within 30 feet of the actual location. I also find it interesting that the mini appears to acquire and resolve locations much faster than the TX which was also my experience.

Pete
 
Also keep in mind that the accuracy is dependent on the number of satellites -and the position of the satellites relative to your ground position. It would be interesting to see if you both had the same set of satellites through the test process.

Is the GPS receiver for the mini different than the standard sized GPS receiver?
 
Also keep in mind that the accuracy is dependent on the number of satellites -and the position of the satellites relative to your ground position. It would be interesting to see if you both had the same set of satellites through the test process.

Is the GPS receiver for the mini different than the standard sized GPS receiver?
The GPS satellites have an orbital period of about 12 hours. We tested a couple of days apart, and at different times. While our location is less than 100 miles apart, I am sure the satellite array was different between us.

Nope, MINI, TX and LCD GPS card all use the same receiver.
 
The GPS satellites have an orbital period of about 12 hours. We tested a couple of days apart, and at different times. While our location is less than 100 miles apart, I am sure the satellite array was different between us.

Even so - it might be interesting to see if the distribution of satellite positions are roughly similar or radically different.
 
Back
Top