Help Support RocketryForum by donating using the link above or becoming a Supporting Member.


Page 9 of 9 FirstFirst ... 5 6 7 8 9
Results 241 to 254 of 254
  1. #241
    Join Date
    26th November 2009
    Posts
    4,363
    Oh, one other thing. One can use the UBlox utility, UCenter, before and after radio installation. It didn't occur to me until after I checked my only available Neo7 GPS clone (the $9.00 special) and found it was serviceable. Got a nice lock indoors and could have fired a very small GPS guided missile or bomb to "take me out" without the neighbors hearing or knowing about it! (sic) That was feeding it to the computer with the FTDI board.

    After installation into the 3DR module there is definitely a difference in the quality of the lock. If O.T.T.'s supposition above is correct in #239, this still could be a dead end for a tight small, cheap tracker. Sure one could buy a "brand GPS chipset" from the csgshop above but there goes the $25.00 appellation.

    One thing I don't understand is if these rigs are so sloppy, how can anyone reliably use 'em and expect them to work with quad copters? I presumed it would be a piece o' cake to get them modified to simply transmit a 9600 bps GPS datastream with decent reliable range? Maybe not! Still it's fun to play with.

    Perhaps there are some different radio settings that would improve the reliability and increase the expected range. I see the quad people struggling to get
    4000 meters with these things. That's only 2.5 miles. I'll try to get some more pictures posted. Will consider trying to double stick a piece of sheet steel to the plastic case of the air module, tie it to the ground and put the GPS on top of that. If the radio design stinks that might not help at all either but I have the stuff
    around to try it. Plus, I can fire up UCenter before and after and see the results. I find it works better with the satellites directly overhead. I've sometimes
    seen locks of 4 to 8 satellites but the number of satellites held jumps around a lot. It might be 4, 6 , 8 every second so still somewhat unsteady. Kurt

    Last edited by ksaves2; 10th September 2017 at 04:42 PM. Reason: sp.

  2. #242
    Join Date
    10th July 2007
    Location
    Melbourne, Australia
    Posts
    963
    I think Rocket Locator can show the receive signal strength or s/n ratio of the satellites in view. A good indication of your shielding performance would be to have the unit in the open and look at a signal strength for a satellite directly above you. We would be working on the assumption that all satellites are transmitting the same power to your location. Probably a reasonable guess. The way to really guarantee good comparison results would be to have two units, one with and one without shielding, working simultaneously and look at the signal statistics.

    Number of satellites jumping around really sounds to me like some of the correlators are having a hard time in the GPS. Modern GPS units are normally quite solid once signal is acquired. Modern units often have more than one correlator per satellite as there is generally only between 8 and 11 visible in the sky. If you look at a modern GPS Rx (say SiRF) it has 48 channel (48 correlators) in the processor. May as well use the extras for something, so they use more than one correlator per satellite and reduce the position error. The early GPS units had one correlator and had to hop from sat to sat. They were horrendously slow. As the electronics became more powerful for the same amount of input power, the number of correlators in the Rx was increased (but battery life expectancy was still similar).

    TRA 13430, Level 3

    "Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

  3. #243
    Join Date
    26th November 2009
    Posts
    4,363
    One thing is when I connected up the GPS through the FTDI serial board directly to the computer there was a very good lock on 8 or more satellites and there was minimal jitter from indoors. I do have a 3DR 100Mw, 900 Mhz set in the basement I'm going to pull out and try. Now a better quality GPS can be tried but
    there goes the economy. Kur t
    Last edited by ksaves2; 12th September 2017 at 07:46 PM.

  4. #244
    Join Date
    26th November 2009
    Posts
    4,363
    I have a Neo-M8N clone coming. If that doesn't show improved performance with the 500mW 433Mhz 3DR radio I will call this a dead end.
    Perhaps the 100mW radios perform better but why waste your time dinking around when there are 100mW and 250mW commercial solutions that
    though they cost a bit more are affordable and reliable on 900Mhz that don't require a Ham license?

    Perhaps an $80.00 commercial GPS will deal with the front end noise of the 3DR radio but there goes the economy. Better with a 100mW Beeline GPS on 70cm.
    Or if one has the room and just needs to find the rocket an AP510: https://www.sainsonic.com/ap510-aprs...aprsdroid.html

    I'll ride this pony till the end, dead or no. Stay tuned.

    One question, if these 3DR radios are so dirty how in the world can anyone use them in a quadcopter that requires absolute positive control? Kurt
    Last edited by ksaves2; 14th September 2017 at 03:25 PM. Reason: added question

  5. #245
    Join Date
    26th November 2009
    Posts
    4,363

    NOW WE'RE TALKING!

    Ok, got an 8N clone here: https://www.aliexpress.com/item/UAV-...305.4.7.RHk2gz We're talking cheap. U-Center allows programming and the son-of-a-gun allows concurrent Glonass use.

    I have it running outside now with Glonass active along with GPS and it's working great! I can have it plugged into the 3DR radio and once the positions are fixed, they
    come right across! Not like the chipset in the first post.

    Bad news is my APRS tracking program APRSISCE/32 can't get a fix with GPS/Glonass both turned on. Shut off Glonass and use GPS only and it works fine with my particular program.
    About 6 satellites fixed on GPS. 12 with GPS and Glonass active! UCenter shows a really darned good fix with minimal jitter.

    Now............... Have to see if I can program a callsign in there. Will post pictures screen saves tomorrow. I be happy.

    Nice thing is the clone GPS chip is programmable. I turned on the jammer protection and that seems to help. Kurt

  6. #246
    Join Date
    28th June 2017
    Location
    Northern VA
    Posts
    34
    Quote Originally Posted by ksaves2 View Post
    Ok, got an 8N clone here: https://www.aliexpress.com/item/UAV-...305.4.7.RHk2gz We're talking cheap. U-Center allows programming and the son-of-a-gun allows concurrent Glonass use.

    I have it running outside now with Glonass active along with GPS and it's working great! I can have it plugged into the 3DR radio and once the positions are fixed, they
    come right across! Not like the chipset in the first post.

    Bad news is my APRS tracking program APRSISCE/32 can't get a fix with GPS/Glonass both turned on. Shut off Glonass and use GPS only and it works fine with my particular program.
    About 6 satellites fixed on GPS. 12 with GPS and Glonass active! UCenter shows a really darned good fix with minimal jitter.

    Now............... Have to see if I can program a callsign in there. Will post pictures screen saves tomorrow. I be happy.

    Nice thing is the clone GPS chip is programmable. I turned on the jammer protection and that seems to help. Kurt
    Sounds like you might be on to something

  7. #247
    Join Date
    26th November 2009
    Posts
    4,363
    Here is the Rig:
    Click image for larger version. 

Name:	rig.jpg 
Views:	15 
Size:	161.0 KB 
ID:	328131

    The little board off to the side is the 5V regulator Pololu http://www.ebay.com/itm/Pololu-5V-1A...8AAOSwacdZbWdl It's 1amp and I plug a 2S pack to it. 5V out in a very stable fashion.

    The GPS is the 8N clone. The radio settings are in this screen:
    Click image for larger version. 

Name:	setup.jpg 
Views:	13 
Size:	155.2 KB 
ID:	328132

    My setup is ideal in that the GPS can be disconnected from the radio (unplugged actually) so the radio can be reprogrammed. Can't be attached while you are doing it.
    Click image for larger version. 

Name:	Great.jpg 
Views:	12 
Size:	176.6 KB 
ID:	328134Click image for larger version. 

Name:	Great1.jpg 
Views:	12 
Size:	124.3 KB 
ID:	328135

    Picture on the left is the NMEA feed coming in with the radio/GPS out in the yard. The screen on the right is the UCenter feed with the GPS attached via the FTDI Basic 5V USB serial
    breakout board:
    Click image for larger version. 

Name:	FTDI.jpg 
Views:	9 
Size:	69.0 KB 
ID:	328137

    If one is bold, it's essential you have one of these boards because if you lose lock on your radio pairs you're screwed without it. I just so happened to have this around with a variety
    of JST cables to fit the radio and GPS.

    Good news/bad news:

    1. Guess what? The 8N comes with GPS/Glonass systems activated. Guess what? The 3DR radios will duly report all the information coming across in a serial terminal. Very nice indeed
    and with the UCenter GPS program monitoring the stream (the right screensave above), there is very little jitter. That is the position jumps around when the tracker is stationary.
    (It was late and I didn't get a picture of that.)

    2. Good news is one can have their Android device use Bluetooth GPS and select "Allow mock GPS locations": https://play.google.com/store/apps/d....android.btgps
    and one can track with the "Blue" GPS as their rocket. Can't navigate but you can see the blue dot where your rocket is. This is using the combined GPS/Glonass stream.
    Really helps with the accuracy once locked.

    3. Bad news. None of my Ham apps including the Android app "GPS Rocket Locator" can track with the combined Glonass/GPS strings coming in. That trick in #2 can allow you to see the
    rocket with "GPS Rocket Locator" as the Blue GPS. That program B/T GPS once running and minimized ignores your internal GPS and pipes the strings coming in over the receiver
    as "your" GPS. For some reason, "GPS Rocket Locator" can display the position as the "blue dot" but not as the red push pin. None of my Hams apps could read it either...........But...

    4. With that FTDI USB/serial interface, one can go into the GPS units setup with UCenter and shut off using the Glonass system. WAAS and SBAS is fine. Do that and all the programs will work perfectly fine. The picture on the left is with the U.S. GPS positions coming in on APRSISCE/32 using an NMEA port. The constellation was in a very good position last night.

    5. One more bad news item. The frontend of the radio intereferes with the GPS even with jamming protection enabled. I couldn't have the GPS lying on top of the radio a' la shrink tubing
    it to the case. That's a no go. Spread out from the radio on a cable (and perhaps a ferrite filter might help) and it improved on the GPS reception situation immensely.

    Bottom line: Go buy a commercial unit if looking for a 100mW 900Mhz tracker or even a 250mW tracker. The Ham stuff has the potential for better propagation but the unlicensed GPS trackers are doing a darned good job judging by what folks are experiencing.

    I'm not so certain that spread spectrum is reasonable for rocket tracking. We only need a one way link from the tracker to the receive station and a two way link is superfluous.
    The drone guys seem to gripe about limited range and they are depending on these things for a satisfactory two way link. Besides, the radios are so dirty you can't shrink tube
    them together on the GPS.

    500mW on 70cm seemed like a decent power point but if it won't (and I don't know this) provide long range and reliable high rate position reporting I'd place 3DR on the fringe of
    rocket tracking. Again, if you don't have the stuff lying around to experiment nor have an inkling of what's going on (that hasn't stopped me any!) buy commercial.

    I'll keep hammering at it. If a callsign can be stuffed in one of the strings, it would at least be worth test flying in a modest sized rocket that could handle the limitations.

    $25.00? Naaaahhhh. Not even close to it. Probably more like $75.00 and you won't know if it will work at range. Kurt

  8. #248
    Join Date
    26th November 2009
    Posts
    4,363
    Hmmmmmmm, I just had a halfed-a**ed idea. With #2 above, I said that GPS Rocket Locator could track the combined GPS/Glonass strings as long as it's coming through the minimized Android
    app "Bluetooth GPS". Well, well, stupidhead here had an epiphany when I hit send. I've got four B/T GPS sources around here that are old and can do 9600bps, If I can pair them to my
    Android device, velcro it to the top of my ball cap and if I can select the local B/T dongle as the "rocket" might be able to track with the "red pushpin" being "me" and the blue dot
    being the rocket. Positions reversed. Well that could be a cool work around!

    Shoot, now I got to go reprogram my GPS back to GPS/Glonass and give it a whirl! Still, the cobbled arrangement would only work with GPS Rocket Locator and no other Ham apps I got. Kurt

  9. #249
    Join Date
    26th November 2009
    Posts
    4,363
    Went for a ride with "GPS Rocket Locator". The local GPS in the phone was the blue dot and the red pushpin was an ancient Royaltek RBT-1000. It works.

    Click image for larger version. 

Name:	Screenshot_2017-09-17-10-36-45[1].png 
Views:	10 
Size:	540.6 KB 
ID:	328183Click image for larger version. 

Name:	Screenshot_2017-09-17-10-46-54[1].png 
Views:	13 
Size:	514.6 KB 
ID:	328184

    Thing is, is the blue dot or your local position is seemingly sampled once every 5 seconds. The red rocket pushpin is once a second. In this case I was using the B/T Royaltek RBT-1000 as a "local" GPS source.

    As I mentioned above, with the Android app B/T GPS I can attach the stream coming in over the 3DR radio that is using both the GPS/Glonass strings to determine position. If one has an internet connection on site, they can
    run B/T GPS and "see" the rocket position on the map. B/T GPS is of no use without maps. If B/T GPS is active on one's device and is minimized along with "allow mock GPS locations" the HC-05 or 06 that is feeding
    data from the 3DR link can be patched to the blue dot, local position. GPS Rocket Locator cannot decode the native GPS/Glonass strings but if presented to GPS Rocket Locator by B/T GPS the positions will be plotted.

    Problem is, is I think the rocket "blue dot" position will now be once every 5 seconds and the red pushpin "local" position coming in over the Royaltek will be the desired once per second.

    Oh well, I'll give it a run later tonight after I reprogram the GPS back to GPS/Glonass from just the plain US GPS.

    Be aware, if you get an 8N clone like I show, it will come with GPS/Glonass active and none of my local Ham apps or GPS Rocket Locator could decode the strings. Bluetooth GPS was the only thing that plotted the position
    but you can't navigate directly with it.

    Basically, I'm switching the positions. The rocket is the blue dot and the red pushpin is local. Kurt

  10. #250
    Join Date
    26th November 2009
    Posts
    4,363
    Ok, Got home, reprogrammed the 3DR radios and reprogrammed the Ublox-8N clone to use both GPS/Glonass. Got the Royaltek RBT NMEA Bluetooth source out.

    Click image for larger version. 

Name:	IMG_20170917_173051[1].jpg 
Views:	12 
Size:	283.9 KB 
ID:	328201

    This is the setup. The 3DR radio is obvious along with the little Pololu 5V 1amp stepdown voltage regulator that feeds 5V to the radio and the -8N clone. The Royaltek RBT1000 bluetooth
    GPS is down in front. The 3DR ground module is in the back. I used an HC-05 with the angled pin connectors. I can use the Bluetooth or disconnect the + wire on the module and
    plug it into a laptop for programming proposes. With the HC-05 module connected the USB plug is just providing power from that big mother black battery

    Soooo. I set the air module outside, plugged the ground module into the battery and wait until there is a lock on the GPS. I pair the HC-05 B/T module to the Android phone and then
    call up the B/T GPS app: https://play.google.com/store/apps/d...id.btgps&hl=en I select allow mock GPS location and minimize the thing. There will be a small indicator in the upper left of the screen.

    Click image for larger version. 

Name:	Screenshot_2017-09-17-17-05-02[1].png 
Views:	7 
Size:	579.4 KB 
ID:	328202

    You can see the little blue ball in the upper left. The blue dot is the "rocket location".

    Next step is to pair the outside base station GPS source. We can't use the internal GPS on the phone or tablet using GPS/Glonass sentences combined. They have to run through
    the B/T GPS program so GPS rocket locator can "see them". I'm using the Royaltek RBT 1000 and you can get these things used for pretty cheap.
    I then pair that up and now fire up "GPS Rocket Locator". Now right off the bat you'll see the "blue dot" which is now the tracker position. Next you have to go to the setup pulldown
    and select the B/T source which in my case was the Royaltek. Took a few minutes but the Red pushpin with my local position came up.

    You gotta remember the positions are reversed. Blue dot rocket, red push pin your local position.

    Click image for larger version. 

Name:	Screenshot_2017-09-17-17-23-37[1].png 
Views:	8 
Size:	464.9 KB 
ID:	328208Click image for larger version. 

Name:	Screenshot_2017-09-17-17-24-36[1].png 
Views:	9 
Size:	498.4 KB 
ID:	328209

    Left shot is overview of a drive the right shot is closeup in the driveway. The major surprise is once you get both spots displayed it's almost instantaneous when the positions come in.

    With the standard "blue dot" local position and "red pushpin" rocket, it seems the red pushpin is tracked once a second and the blue dot once every 5 seconds. That is perfectly fine for
    the ground position to lag.

    Now with the "blue dot" as the rocket and the "red pushpin" representing one's local position, the blue dot jumps once a second and the pushpin follows almost immediately.

    Now to try to get a callsign in there and see what happens with a real flight. Kurt
    Last edited by ksaves2; 18th September 2017 at 01:17 AM.

  11. #251
    Join Date
    10th July 2007
    Location
    Melbourne, Australia
    Posts
    963
    Looks like great progress there Kurt.

    If you have a piece of FR4 PCB material (unetched) you can try mounting the Tx module on one side and the GPS, battery, and regulator on the other. Connect the sheet of copper to ground. For extra brownie points you can add feedthrough caps, and/or a ferrite toroid or two, and see what improvement that makes for EMC. That would be a great test. I think a ferrite toroid around the wires to the Tx module is important. Even one or two turns will ensure the return currents are encouraged back along their correct return path, rather than increasing emissions that can worm their way in elsewhere.

    If you shape the PCB piece like a sled you can then mount it in a rocket and fly it if the tests come out much improved
    Last edited by OverTheTop; 18th September 2017 at 02:40 AM.
    TRA 13430, Level 3

    "Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

  12. #252
    Join Date
    26th November 2009
    Posts
    4,363
    Quote Originally Posted by OverTheTop View Post
    Looks like great progress there Kurt.

    If you have a piece of FR4 PCB material (unetched) you can try mounting the Tx module on one side and the GPS, battery, and regulator on the other. Connect the sheet of copper to ground. For extra brownie points you can add feedthrough caps, and/or a ferrite toroid or two, and see what improvement that makes for EMC. That would be a great test. I think a ferrite toroid around the wires to the Tx module is important. Even one or two turns will ensure the return currents are encouraged back along their correct return path, rather than increasing emissions that can worm their way in elsewhere.

    If you shape the PCB piece like a sled you can then mount it in a rocket and fly it if the tests come out much improved
    I've got some double sided PC board lying around here somewhere. Used to make ematches in the old days. Got good at it too. I have a large collection of toroids lying around. What size you think?

    Running GPS Rocket Locator Bass ackwards is weird but actually I like it better than running it in the standard mode with just the GPS satellites. Glonass adds to the accuracy if you ask me. I have no idea why I have to run the B/T GPS program to get the
    strings decoded properly and sent to GPS Rocket Locator. None of my Windows tracking apps can read the combined GPS/Glonass strings either.

    I have to give my Linux boxes a shot and see. Actually Xastir gives the prettiest screen for APRS or NMEA tracking. With the NMEA trackers, one fires up a python script to intercept the NMEA strings which converts it to a pseudo APRS packet that gets presented to Xastir. It gets displayed on the map very nicely.

    I know I'll be able to get the ground module to fire off the strings to an NMEA port (B/T or USB) so I'll see if I can get Xastir to deal with it. If does, that means me and one other guy (the fellow who wrote the python script), Jason KG4WSV can use it! Jason has the script posted publicly and it has his callsign in the rem notes of the program. Says he flies rockets too. He's out of Alabama.

    Also, I discovered last week an Android app ROOT/Gnu Debian and XSDL. It's a bit of a gyration but it fires up Linux and lxde X-windows on top of Android! I did an apt-get install xastir and it installed and I was able to get my maps in.
    It connects fine to an -IS server (ie. google aprs or aprs fi) and the only thing left is to see if I can get a B/T device to connect. Sheesh, I attached a B/T keyboard and mouse through Android and it's "just there" when I fire up
    X-windows. Same thing with WiFi or the cell data network. If it's connected via the Android side it's "just there" when I fire up X-windows. I've never had be that easy before. I need to see if I can get a user definable device transparent on the linux side so I can pipe my NMEA strings in or use a B/T TNC for APRS stuff. I got GNU/Root and XSDL running in a Nexus 7 2012 3G, 2013 LTE and a cheap Blu Studio G plus. I bought all this stuff used for
    way the heck less than a single Apple product. Root 'em and tweak them I say and one can really be surprised.

    I'm surprised there aren't a few accomplished 'copter guys on this board. I would'a thought someone would of sounded off to tell me how to optimize the radio settings for what we are trying to do based on their experience but
    nary a peep. Probably no one who's an advanced quad copter person flies rockets? Kurt

  13. #253
    Join Date
    10th July 2007
    Location
    Melbourne, Australia
    Posts
    963
    What size you think?
    Only need something small. Usually as small as you can and still get the connector through. You could go smaller if you like.

    The larger cores saturate (lose their inductance) typically at higher currents, as the magnetic flux increases. We are using it as a common-mode choke, with all the wires to the Tx module passing through. Basically the current flowing in the +Ve flows back out the -Ve (GND) so the magnetic fields cancel to zero. Likewise for the serial lines if they have any signals on them We want to encourage the return currents to go where they are supposed to go (on the wiring loom), and this CM choke helps that. Any signals that want to go "the long way around", via parasitic couplings or other PCBs, end up with the CM choke sensing the imbalance in the return paths and helping by pulling the voltage in the direction that encourages it back through the choke. Anything that is single-ended, does not go out and come back through the CM choke, sees a higher impedance and attenuation.

    So, using the cores as a CM choke above you usually don't have to worry about size too much, or saturation, as the normal running current sums to zero.

    If you put a toriod on a single wire to create a series inductance, it needs to be sized for the magnetic flux at peak current (DC+AC) through the core. Remember too that more turns increases the magnetic flux induced, as well as the inductance. If you saturate the core it loses its magnetic properties temporarily and the inductance goes down the toilet.
    TRA 13430, Level 3

    "Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

  14. #254
    Join Date
    26th November 2009
    Posts
    4,363
    Ok, Latest tests:

    Click image for larger version. 

Name:	shield2.jpg 
Views:	12 
Size:	107.7 KB 
ID:	328327Click image for larger version. 

Name:	shield4.jpg 
Views:	13 
Size:	154.2 KB 
ID:	328328

    Right picture is with the GPS lying on the end of its short cable with a filter.
    The shot on the left is the GPS lying on top of the shielded case. I put Rf shielding on
    both sides of the case of the 3DR radio. Well this kinda stinks.

    I guess for 433Mhz, 500mW, Rf noise is too bad to shrink wrap the GPS on the radio.
    Perhaps in a large nosecone bay of a 3 or 4 inch rocket it might be workable. The GPS
    is on a short cable and it gets pretty good reception as long as it's not lying on the
    radio. Using a combination of Russian and US satellites. With a long nosecone it could
    work.

    Think I'll get a 500mW 900Mhz rig and see if it's just as dirty. It might not be as bad as the 433Mhz rig.

    When the weather breaks, I'll put it on top of the roof and do a range test. I can use "GPS Rocket Locator" and see what I get with my Android Tablet and the utility Bluetooth GPS.

    Kurt


Similar Threads

  1. Looking for a solution..warped fin
    By AKVP in forum The Watering Hole
    Replies: 6
    Last Post: 13th February 2015, 01:33 PM
  2. AD's SOLUTION
    By Gary Byrum in forum The Watering Hole
    Replies: 6
    Last Post: 11th April 2014, 04:00 PM
  3. Here's the solution for Kim jong un
    By closet astronaut in forum The Watering Hole
    Replies: 10
    Last Post: 26th April 2013, 10:31 PM
  4. It is not a good solution
    By SAC of MMMSClub in forum The Watering Hole
    Replies: 11
    Last Post: 6th August 2012, 01:05 AM
  5. Masking Solution
    By Hospital_Rocket in forum Techniques
    Replies: 0
    Last Post: 30th May 2005, 02:21 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •