So glad you're covering more iNav related things! It's where I'm trying to live right now so it's very nice to have you experimenting and ahead of me as I come up to problems as well :-)
I'd like to suggest the Bardwell Shirt Change Complexity Index: the amount to learn on a given topic is proportional to the number of shirt/wardrobe changes Joshua goes through in his video on the topic. This particular video seems to have one of the highest BSCCI numbers I can recall, 7 by my count.
ok, one thing I could add to this tuto, the GPS quality makes a big difference. I was having a ton of problems getting the GPS stable until I saw another youtuber saying that he replaced his GPS and got all his problems solved. He had the exact same GPS as me: FOXEER M10-Q120. It is days and night different. I always assumed there would be very small differences between one another but it happens that the difference is much larger than expected. Thanks for that video!
A couple months ago I literally tried to convert my Chimera 7 to iNav because I wanted to try out a different looking OSD, position hold and iNavs return to home... I gave up after trying to get everything to work for days... return to home just kept trying to kill my quad...so I switched back to Betaflight. Ill have to try again now after watching your video!
Nice to see an update on this. I have a few tips for you for the compass-less navigation stuff. A few requirements are necessary for this to work the best: - Well balanced props with as little vibrations as possible and/or properly tuned filters. I saw a lot of jello in your feed even at idle so a mechanically noisy setup will struggle to keep the fused sensor data consistent - Clean pid tune. Even the wobble from a bad tune can cause erratic data that make it hard for INAV to determine wind speed and direction. - Enable the wind estimator OSD element as well to judge how well that works as wind speed and direction needs to be kind of accurate to correctly determine the ground track vs heading. - raise your navigation speed. The default navigation speed is just 3m/s for safety reasons. for a big 7" this is very slow. The faster your navigation speed is, the easier it will be for inav to recalculate heading during RTH and you will have less risk to get it off course. Its in Advanced tuning. If you start RTH with a slight misalignment and its toilet bowling, as long as it reaches altitude and starts moving towards home, you are fine as it will quickly regain the correct heading.
This is why I've seen some put the compass up in a mast, but that's a cumbersome and not very crash-friendly solution. After seeing this, I think I'll just continue to stick to M10 GPS without compass on my builds.
The magnetometer goes on a fiberglass mast 6 to 8 inches high. If you don't do that you have to go though what you just did to get it to work and it's just not worth it. Betaflight pilots are going to be offended by the crash resistance of that configuration but thing is autonomous multirotors don't really crash, that's a human problem. Ardupilot has a tool to characterize the throttle dependent aberration of the magnetometer and subtract it but I don't imagine that's in Inav.
@@DFEUERMAN Yes and no. Those times it gets knocked off (it is kind of stuck out there) you need it to go back exactly there it was or you need to recalibrate. You don't want to be tempted to just bend it up and go, that causes fly-aways. In principal you could make a mag mount base that pops off easy and drops into the same position every time but I never felt comfortable doing that without testing that it's really is far enough away from the magnet and I haven't gotten around to testing that yet.
@@brocktechnology Interesting. Recalibration seems to be unavoidable. Mag mount might introduce more unwanted field noise, although maybe the noise can be calibrated-out since magnetic field from a magnet is static. The magnet would have to be a very powerful and heavy? Better to just not crash :-) Easier said than done. I've been learning how to TIG weld. I'll recomment here or post a youtube video if my aluminum mast works.
I did this 2 years ago it's distance to motors 3" needed few cm out the back, and 5" Inch was about and inch was enough. Worked perfectly. Magnets in motors plus power was my biggest realisation
This takes me back to the early days - TBS Discos and HobbyKing MultiWii boards, Arducopter, and GPS masts. Thanks for making me feel ancient, Josh ;-)
brilliant thanks, changed my M8 to M10 gps, wound copper tape around my wires & moved my power leads to the middle, got 6sats in no time & indoors (speedybee 5)
not sure if anyone else commented this but the LI-ion batteries have a metal shell and word from everywhere i've looked up the same issue you had is that they therefore create a stronger magnetic field. Josh please keep developing this series. A series on inav 7+ is desperately needed. so many videos are so quickly out dated. I struggle so hard with it all because I unfortunately have bought a whole bunch of parts that are all very good, they just dont like each other. like DJI 03 with the v2 googles. ( I think - the old ones with the 4 antenna ) OSD is just not working for me atm, thats actually why i came to your videos, because everything else is set in inav 6 - 4. its killing my life. But yeah steel shells on the cylindrical li-ion batteries. thanks mate keep up the good work.
Also if you travel places, calibrate compass before flight. It's always a good habit to at least check your heading readings before flight make sure they are still correct. Facing north or with your phone and a compass app. p.s. You can use stick command throttle/yaw top-right and pitch down to enter compass calibration
Another solution that worked for me, not just the compass situation but also in bigger (heavy lift) drones when they arm, the current is strong enough to disable some extra cameras I had, the solution was pretty simple, wrap an aluminium foil around the wire in question (here is the battery wire) and shield it properly, and you are good to go. Tested and worked well, not pretty but I will live with that.
that video was much needed for me. After watching this video I will prefer to increase the distance between GPS and battery wires in my quad.thank you so much for posting this valuable information with demo.
Hahaha🎉. Another series I’m going to have to watch “in-depth”! Just finished with my AOS7, decided to dedicate it to long range and return to home. Inav being the best choice for the job. It has been a challenge just to tune it. Got that sorted out now to play with all the things Inav brings to the table. Mostly GPS related stuff. RTH, POS hold etc. I love the way you go about “splainen” things. 😅. Inav, here we go!!!!!!😊
I have not used inav for many years...but i still have an ancient Q450 quad that used to run and up until today is runing on inav 1.1 with a flip32+ board 😆 yes 1.1 wich at the time i built it was the most recent inav😅😅...and i remenber i had to put the external compass and the gps up on a mast to avoid interference as it would make it drift away specially when throttling up...and guess what...up until the last time i used it... it worked amazingly fine...it would do gps position hold, RTH, altitude hold...all those gps assisted modes perfectly fine...i actually think i still have a video on my channel where i was testing it...even turning my radio off it would come back home and land smoothly in the exact same place where it took off.
Tab connecting the lipo pack cells are all up where the main wire comes out so makes sense basically current flow more isolated to one side of the battery overall.
The battery actually works as a microstrip antenna! Even though its dc current its still pulsing to an extent before it hits any filtering at the esc. I put a 40v capacitor on the nickle strips of my battery in my long range plane and it managed ~5db of extra signal strength!
@Meleonlight as far as what? Micro strip antennas are formed with small lines of wire next to each other separated by a non conductive layer (similar to a battery). If you have a an ac brushless motor turning, the esc is taking bursts of power at a very high frequency. At different throttle levels those frequencies can match up and interfere with video and telemetry signals. On my plane I have a 3s-2p LiOn battery consisting of 6, Samsung 21700 cells, join together with spot welded nickel strips. I have a 40v 480uF capacitor soldered directly to output terminals of the battery. This helps negate the high frequencies within the battery and the cables. As an added measure I ran the battery cables in the shortest manner possible with some aluminum tape covering them from the direction all of my communications electronics which are in the wing. Just a little bit of EMI sheliding and countermeasures can make the biggest difference. But quite honestly if you just space them out far enough, it will more than likely be enough for the average person. I frequently fly 20km+ way point missions with ardupilot and need every bit of signal I can get. (Edit) if you ever get the chance to tear apart a high end consumer drone or plane like something from dji, take notes on how the engineers use small amounts of sheilding in very specific places in order to get the maximum amount of performance with what they're allowed to sell to the general public. Antennas and just radio frequency related stuff in general is a deeeeeep rabbit hole you can go down with trying to optimize and fix problems within your systems. I really find it quite interesting and love to talk about it so sorry for the essay lol.
@@pickeljuice9325 thank you so much for the explanation, you very much said everything i wanted to know, didn't kown about that kind of antennas. And like you, but not at the same level of knowledge lol, want to squeeze every bit of performance/signal out of my 7 inch lt quad that i just build. I Just did the shielding with an usb cable that came shielded and replaced the original wires that came with the gprc m10dq. Before thata, the model on the bf configuratr never stopped to drift. Did the change to inav 7.1 and now seems to be solved, tomorrow its gonna be the day ill be testing that thing jaj. Thanks man
@Meleonlight honestly I've never even though of using shielded usb cables as a means of connecting my components, that might just be the best idea I've ever heard. Going to replace all of my servo leads im using for signals with that now!
I got my inav 7" L.R DRone working great after messing with it for months. I finally got it working great on inav 6.1 and then 7.1 Before RTH would go wrong direction or just sit in place, pos. Hold didn't work either, but now I'm Golden.. Took me months to get it going. Videos on my ch. Of what it did and how it works now
@Joshua One good solution to minimise the influence of magnetic disturbance to the compass and so on is to twist the main cables! And olso the signal lines of serial connections, that should work out nice. Still waiting for my Gps to come and find it out... thank you so much for your great way of explaining complex things in a simple way and for your great channel in general, great work!! Kind regatds Tixi
Well done, 7.1 has some really great features. It does take some time to become proficient like anything else, and the devs seem to be hard at work. 8.0 looks very cool!
Electronic connections give of rf. Because the 18650 battery packs are constructed with numerous spot welds or solder joints, there is more rf interference.
PID tune is also a very importante thing hier. Tunning your PID correctly will contribute a lot with autonomous inav flight. And not only the Giro PIDs but specialy for althold and gpshold PIDs. 😉❤
Copper foil tape that is used for stained glass windows works perfectly. And it's easy to solder a ground wire to it. It fixes all the problems I had with gps connections.
I think the problem is that the compass is affected by the magnetic field caused by current loops. I would suggest keeping the battery cables close to each other and if that does not help shield the cables and battery with aluminium tape. Thanks for your great videos.
I also found 18650 batteries to really mess with the compass more for sure! And my analogy was exactly as you explained here 😎 I have position held a drone for 12 mins without compass and it didn’t drift over that time but I’m aware that some setups drift more than others and I was just lucky
i use a thin ethernet cable and i don't solder anything just cut the ends off and use the wires inside for GPS/Compass. For GPS alone a thin USB cable works flawlessly
your compass was off because you yawed too much. as you said it needs some straight flight to align at first, this also applies during flight. just needs some smooth cruising in between to get everything right again.
8:00min joshua i am visiting iflight regular and i am pretty sure nobody there is interested or keen to jump on you for stating something so obvious. all around super nice people there. saying the chimera is not made for INAV is the same as saying 99% of all quads we can buy now are not made for that. most important is that you fixed it and show others how to do the same. nice work!
Watching this video really made me appreciate learning some basic CAD skills and having a 3D-printer. With those things, moving the gps and compass a few inches becomes trivial.
Glad watched this. Making me reevaluate building a INav Flywoo Explorer... everything is with in an inch of everything.. Oh, don't forget the Pixhawk! Just ordered a kit:)
Just shield the gps wires and ground the shielding (copper tape). And you need to get it far away from the machine especially the vrx so if your vtx is in the back you'd need to extend the unit out the back Father put or put in middle.of the front arms away from the motors
Oh, I do hope you just didnt care about the tune, and were only focused at making compass work. Because, I dont really need any of inav features if the quad flies like this. Following. Thanks for making our FPV lives easier. One of only couple channels I recommend to people.
I use iNav a LOT on my planes. But use BF for my FPV freestyle quads. Even on my 5" GepRC with a M10 GPS/Mag. I have never gotten a good RtH with the mag enabled, probably because I have the battery connector right back there by the compass, just like you did. I'm curious what your impressions are with BF 4.5 now that you've "fixed" the EMF issues on the compass???
Brilliant video! As far as foil tape for the gps wires , for years I have used copper foil tape and you can solder a ground wire to it easily and ground anywhere .
What if you shield everything? I would use the mesh shielding from an old rg59 tv cable? Or just mount the compass on top of the battery strap? What if you twist all your positive and negative wires as a "twisted pair"? Twisted pairs offer some shielding with zero cost and zero weight gain.
just use a stock usb cable cable for conect the gps, i use a geprc m10 DQ with a usb cable from a p30 lite huawei smartphone, 13 cms long, enough long from motors and no interference, rth work superb, also almost always above 30 sats fix
Would like to see more ardupilot video, it really is excellent at automatic flying and good enough for acro flying. iNav simply just losses some reliability features but simpler to tune
am have apex hd5 frame and I created 3d model , now gps is mounted to gopro holder with 0 degree crouch and I turn on galileo ... in past I had max 9 sats and now I have 23-35 sats ...I am using hglrc m100-5883 gps and it's working nice with compass ...
HELLO joshua, thanks for the video, we learn each time; by the way i see on your o3 video the same OSD display faults: question marks are displayed instead of units (mW, arrow, etc.). Did you not manage to resolve this display problem? If yes, how? best regards from France
I also recently bought three of the Flywoo Goku M10 mini GPS with compass. But neither INav or Betaflight wants to recognise the compass. Pretty sure my wiring is correct.
lol , I have a Martian II as you know a knock off from banggood from 7 years ago, with a banggood, ómnibus F4, with Inav from 5 years ago, 2.6.0 and does the same as my DJI Malik, lol just perfect flight modes 5 years ago
Man, been out of the hobby for years and it's crazy to see the advancements but also crazy to see the same struggles. lol Side note, have you not tuned the PID's? or is this the best Inav can do?
18650 Vs lipo is probably because of the steel cases of the batteries cells having higher magnetic permeability. That is the steel cases of the 18650s are acting more like an electro magnet Vs the aluminium cases of the lipo cells.
I have dehydrated my TPU and tried everything I can. It's so stringy and globby. Your mount is so impressive. I am forced to use PLA or PETG because I just can't get anything useful out of TPU :(
Some cables will have a foil shield built into them to attenuate electronic and magnetic interference, would wrapping the battery lead in Mylar from a computer motherboard bag help?
I use 3m copper adhesive tape, its ridiculous expensive but 1 roll should do like 200 qauds soo (worth it?)..... and i always solder it to a ground and my m8Q's will "insta" lock 25+ sats on my chimera. :)
Would it be necessary to ground the shielding to the quad for better results, like twisting a piece of aluminum, isolating it with heatshrink and soldering it from the shielding to a ground spot on the quad?
Ok question. If I wrap battery leads from esc in copper and maybe gps wires in copper tape would that help. Right now just have kaplon tape to hold gps wires down to not touch battery leads
Have you (or anyone :)) tried shielding the battery cables with braided metal cable shielding. I'm guessing that the braid would need to be grounded to the frame. What about a ferrite bead?
In the iNav Alignment Tool, should the heading that moves as you move your quad be the correct heading (90 for east)? I had to mount my flight controller pointing backwards (180 degrees) for wiring reasons. My GPS points forward. No matter what I do, the compass is not correct. I tried setting the Alignment and re-calibrating, but it still tells me the wrong heading in the alignment tool.
@JoshuaBardwell Why does it look like the spacing between plates at the rear of the top plate is twice as high as the front? It looks like the two printed pieces are stacked on each other from what I can see at the 24:49 mark? It looks like the side plates have to come completely off, at least on one side for the battery to be routed to the side. Did you fly it that way?
Hi I searched for the stl files for that mount for the gps and antenna and I couldn't fine it in thingiverse can you share the link or the file if you may
I think a lot of people don’t want to try iNav because of these problems. How did Dji do it so right over 10 years ago with the Naza v2/lite. Just install and go. Maybe they can copy what dji did right so long ago. This is not to downplay the work the devs do on iNav. It’s an amazing firmware that can do considerably more.
A lot of factory quads at that time positioned the GPS antenna and compass in a place that goes and doesn't have electromagnetic interference with the quad or on newer quad they have copper/aluminum shielding between the GPS and electronic components of the quad. This info gets overlooked when buying ready to fly quads or DIY builds because most people use Betaflight without the GPS capability which got added relatively recently.
Omg great video. I have inav on my LR 7" & 8". Wanted after done testing put inav on my 10" but I do have my lead in the back next to compass. Wonder if that why on my betaflight ( I have compass on I notice it's off when on by like 20° when on not just on usb) On my 8" I had to push crossfire antenna wire further away. Also wrapped in copper trying to help. At first compass wouldn't work.
My chimera7 compass is located exactly where yours was, and is working perfect. I think you got the alignment wrong. A drift when moved, shows a conflict between the accelerometer and the compass. I can also see that in the first video, you didn’t align it like in this video.
I know, but on the first video you didn’t set it to flipped, and you had drift. When moved, the heading reacts to the accelerometer first, and then drifts to what the compass says.
Hi JB, what are those two numbers in the OSD just above the compass heading bar? and what is the significance of the differenece between them? Thanks 🙂
The left one is the direction that inav believes the drone is facing, based on all available sources of information (gyro, compass, GPS, etc). The right one is the GPS heading.
My heading doesn't move, and I don't see any drift on the compass but when I use poshold or RTH I get serious toilet bowling so no idea what's up with it.. inav says hardware health is fine
Am I understanding correctly that iNav uses a Kalman filter (the "sensor fusion" part)? The more I get into FPV, the more stuff I've learned about in University makes sense :)
So glad you're covering more iNav related things! It's where I'm trying to live right now so it's very nice to have you experimenting and ahead of me as I come up to problems as well :-)
I'd like to suggest the Bardwell Shirt Change Complexity Index: the amount to learn on a given topic is proportional to the number of shirt/wardrobe changes Joshua goes through in his video on the topic. This particular video seems to have one of the highest BSCCI numbers I can recall, 7 by my count.
This is incredible! 😂
ok, one thing I could add to this tuto, the GPS quality makes a big difference. I was having a ton of problems getting the GPS stable until I saw another youtuber saying that he replaced his GPS and got all his problems solved. He had the exact same GPS as me: FOXEER M10-Q120. It is days and night different. I always assumed there would be very small differences between one another but it happens that the difference is much larger than expected. Thanks for that video!
A couple months ago I literally tried to convert my Chimera 7 to iNav because I wanted to try out a different looking OSD, position hold and iNavs return to home... I gave up after trying to get everything to work for days... return to home just kept trying to kill my quad...so I switched back to Betaflight. Ill have to try again now after watching your video!
Nice to see an update on this. I have a few tips for you for the compass-less navigation stuff.
A few requirements are necessary for this to work the best:
- Well balanced props with as little vibrations as possible and/or properly tuned filters. I saw a lot of jello in your feed even at idle so a mechanically noisy setup will struggle to keep the fused sensor data consistent
- Clean pid tune. Even the wobble from a bad tune can cause erratic data that make it hard for INAV to determine wind speed and direction.
- Enable the wind estimator OSD element as well to judge how well that works as wind speed and direction needs to be kind of accurate to correctly determine the ground track vs heading.
- raise your navigation speed. The default navigation speed is just 3m/s for safety reasons. for a big 7" this is very slow. The faster your navigation speed is, the easier it will be for inav to recalculate heading during RTH and you will have less risk to get it off course. Its in Advanced tuning. If you start RTH with a slight misalignment and its toilet bowling, as long as it reaches altitude and starts moving towards home, you are fine as it will quickly regain the correct heading.
This is why I've seen some put the compass up in a mast, but that's a cumbersome and not very crash-friendly solution. After seeing this, I think I'll just continue to stick to M10 GPS without compass on my builds.
It's not as bad as that it just can't be near the motors why the extra inch at back needed
The magnetometer goes on a fiberglass mast 6 to 8 inches high. If you don't do that you have to go though what you just did to get it to work and it's just not worth it. Betaflight pilots are going to be offended by the crash resistance of that configuration but thing is autonomous multirotors don't really crash, that's a human problem.
Ardupilot has a tool to characterize the throttle dependent aberration of the magnetometer and subtract it but I don't imagine that's in Inav.
Is Aluminum ok instead of fiberglass?
@@DFEUERMAN Yes and no. Those times it gets knocked off (it is kind of stuck out there) you need it to go back exactly there it was or you need to recalibrate. You don't want to be tempted to just bend it up and go, that causes fly-aways. In principal you could make a mag mount base that pops off easy and drops into the same position every time but I never felt comfortable doing that without testing that it's really is far enough away from the magnet and I haven't gotten around to testing that yet.
@@brocktechnology Interesting. Recalibration seems to be unavoidable. Mag mount might introduce more unwanted field noise, although maybe the noise can be calibrated-out since magnetic field from a magnet is static. The magnet would have to be a very powerful and heavy? Better to just not crash :-) Easier said than done. I've been learning how to TIG weld. I'll recomment here or post a youtube video if my aluminum mast works.
I did this 2 years ago it's distance to motors 3" needed few cm out the back, and 5" Inch was about and inch was enough. Worked perfectly. Magnets in motors plus power was my biggest realisation
This takes me back to the early days - TBS Discos and HobbyKing MultiWii boards, Arducopter, and GPS masts. Thanks for making me feel ancient, Josh ;-)
brilliant thanks, changed my M8 to M10 gps, wound copper tape around my wires & moved my power leads to the middle, got 6sats in no time & indoors (speedybee 5)
Awesome to see you covering iNav.
not sure if anyone else commented this but the LI-ion batteries have a metal shell and word from everywhere i've looked up the same issue you had is that they therefore create a stronger magnetic field. Josh please keep developing this series. A series on inav 7+ is desperately needed. so many videos are so quickly out dated. I struggle so hard with it all because I unfortunately have bought a whole bunch of parts that are all very good, they just dont like each other. like DJI 03 with the v2 googles. ( I think - the old ones with the 4 antenna ) OSD is just not working for me atm, thats actually why i came to your videos, because everything else is set in inav 6 - 4. its killing my life. But yeah steel shells on the cylindrical li-ion batteries.
thanks mate keep up the good work.
I love that feeling of finally getting something to work after try after try after try.
You should use copper tape for the shielding. You shouldn't have problems with soldering by using it.
Also if you travel places, calibrate compass before flight. It's always a good habit to at least check your heading readings before flight make sure they are still correct. Facing north or with your phone and a compass app.
p.s. You can use stick command throttle/yaw top-right and pitch down to enter compass calibration
Another solution that worked for me, not just the compass situation but also in bigger (heavy lift) drones when they arm, the current is strong enough to disable some extra cameras I had, the solution was pretty simple, wrap an aluminium foil around the wire in question (here is the battery wire) and shield it properly, and you are good to go. Tested and worked well, not pretty but I will live with that.
Oh I just reached the part where I add the foil :) still, I would suggest you do the same around the batt one
that video was much needed for me. After watching this video I will prefer to increase the distance between GPS and battery wires in my quad.thank you so much for posting this valuable information with demo.
So happy your diving into Inav
Hahaha🎉. Another series I’m going to have to watch “in-depth”! Just finished with my AOS7, decided to dedicate it to long range and return to home. Inav being the best choice for the job. It has been a challenge just to tune it. Got that sorted out now to play with all the things Inav brings to the table. Mostly GPS related stuff. RTH, POS hold etc. I love the way you go about “splainen” things. 😅. Inav, here we go!!!!!!😊
What vtx you using? Digital? What kind?
I have not used inav for many years...but i still have an ancient Q450 quad that used to run and up until today is runing on inav 1.1 with a flip32+ board 😆 yes 1.1 wich at the time i built it was the most recent inav😅😅...and i remenber i had to put the external compass and the gps up on a mast to avoid interference as it would make it drift away specially when throttling up...and guess what...up until the last time i used it... it worked amazingly fine...it would do gps position hold, RTH, altitude hold...all those gps assisted modes perfectly fine...i actually think i still have a video on my channel where i was testing it...even turning my radio off it would come back home and land smoothly in the exact same place where it took off.
Thanks. This explains why my battery-mounted GPS/compass was 30-45 degrees off just sitting idle.
Tab connecting the lipo pack cells are all up where the main wire comes out so makes sense basically current flow more isolated to one side of the battery overall.
The battery actually works as a microstrip antenna! Even though its dc current its still pulsing to an extent before it hits any filtering at the esc. I put a 40v capacitor on the nickle strips of my battery in my long range plane and it managed ~5db of extra signal strength!
Can you explain a little furter or show some pic?
@Meleonlight as far as what? Micro strip antennas are formed with small lines of wire next to each other separated by a non conductive layer (similar to a battery). If you have a an ac brushless motor turning, the esc is taking bursts of power at a very high frequency. At different throttle levels those frequencies can match up and interfere with video and telemetry signals. On my plane I have a 3s-2p LiOn battery consisting of 6, Samsung 21700 cells, join together with spot welded nickel strips. I have a 40v 480uF capacitor soldered directly to output terminals of the battery. This helps negate the high frequencies within the battery and the cables. As an added measure I ran the battery cables in the shortest manner possible with some aluminum tape covering them from the direction all of my communications electronics which are in the wing. Just a little bit of EMI sheliding and countermeasures can make the biggest difference. But quite honestly if you just space them out far enough, it will more than likely be enough for the average person. I frequently fly 20km+ way point missions with ardupilot and need every bit of signal I can get. (Edit) if you ever get the chance to tear apart a high end consumer drone or plane like something from dji, take notes on how the engineers use small amounts of sheilding in very specific places in order to get the maximum amount of performance with what they're allowed to sell to the general public. Antennas and just radio frequency related stuff in general is a deeeeeep rabbit hole you can go down with trying to optimize and fix problems within your systems. I really find it quite interesting and love to talk about it so sorry for the essay lol.
@@pickeljuice9325 thank you so much for the explanation, you very much said everything i wanted to know, didn't kown about that kind of antennas.
And like you, but not at the same level of knowledge lol, want to squeeze every bit of performance/signal out of my 7 inch lt quad that i just build. I Just did the shielding with an usb cable that came shielded and replaced the original wires that came with the gprc m10dq. Before thata, the model on the bf configuratr never stopped to drift. Did the change to inav 7.1 and now seems to be solved, tomorrow its gonna be the day ill be testing that thing jaj. Thanks man
@Meleonlight honestly I've never even though of using shielded usb cables as a means of connecting my components, that might just be the best idea I've ever heard. Going to replace all of my servo leads im using for signals with that now!
We need more INAV content!
I got my inav 7" L.R DRone working great after messing with it for months. I finally got it working great on inav 6.1 and then 7.1
Before RTH would go wrong direction or just sit in place, pos. Hold didn't work either, but now I'm Golden.. Took me months to get it going. Videos on my ch. Of what it did and how it works now
@Joshua
One good solution to minimise the influence of magnetic disturbance to the compass and so on is to twist the main cables! And olso the signal lines of serial connections, that should work out nice. Still waiting for my Gps to come and find it out... thank you so much for your great way of explaining complex things in a simple way and for your great channel in general, great work!!
Kind regatds Tixi
Well done, 7.1 has some really great features. It does take some time to become proficient like anything else, and the devs seem to be hard at work. 8.0 looks very cool!
Electronic connections give of rf. Because the 18650 battery packs are constructed with numerous spot welds or solder joints, there is more rf interference.
PID tune is also a very importante thing hier. Tunning your PID correctly will contribute a lot with autonomous inav flight. And not only the Giro PIDs but specialy for althold and gpshold PIDs. 😉❤
Try this, add the foil tape to the battery leads as well.
Copper foil tape that is used for stained glass windows works perfectly. And it's easy to solder a ground wire to it. It fixes all the problems I had with gps connections.
Same here. Makes a world of difference
I've found the super thin ethernet cables or USB cable works like a charm too
The accomplishment Glory call was awesome. Like 3 WEEEEKS. Omg 😱 this is in my future
I think the problem is that the compass is affected by the magnetic field caused by current loops. I would suggest keeping the battery cables close to each other and if that does not help shield the cables and battery with aluminium tape. Thanks for your great videos.
I also found 18650 batteries to really mess with the compass more for sure! And my analogy was exactly as you explained here 😎 I have position held a drone for 12 mins without compass and it didn’t drift over that time but I’m aware that some setups drift more than others and I was just lucky
Copper tape and ground it. Works perfect
brilliant video series. I do a compass calibration everytime i change flying sites by more than 50 miles, or if changing to canyon flying.
I've used copper tape before, you can solder a wire to it ez and it seems to work amazingly.
i use a thin ethernet cable and i don't solder anything just cut the ends off and use the wires inside for GPS/Compass. For GPS alone a thin USB cable works flawlessly
your compass was off because you yawed too much. as you said it needs some straight flight to align at first, this also applies during flight. just needs some smooth cruising in between to get everything right again.
8:00min joshua i am visiting iflight regular and i am pretty sure nobody there is interested or keen to jump on you for stating something so obvious. all around super nice people there. saying the chimera is not made for INAV is the same as saying 99% of all quads we can buy now are not made for that.
most important is that you fixed it and show others how to do the same. nice work!
Now with optical flow!! That’s been a real pain in my ass
I am agree I can´t make it work
Watching this video really made me appreciate learning some basic CAD skills and having a 3D-printer. With those things, moving the gps and compass a few inches becomes trivial.
This is the content I've been looking for! Thank you!
Glad watched this. Making me reevaluate building a INav Flywoo Explorer... everything is with in an inch of everything.. Oh, don't forget the Pixhawk! Just ordered a kit:)
Keep the inav content coming!
Do you think it would be worth it to try shielding the battery wires?
Just shield the gps wires and ground the shielding (copper tape). And you need to get it far away from the machine especially the vrx so if your vtx is in the back you'd need to extend the unit out the back Father put or put in middle.of the front arms away from the motors
who else just wants to see Bardwell make an edit if him ripping his yard like the good old days?
I sure do!!!
18650 cells use a steel casing (feromagnetic material) that causes more interference with the compass magnetic field. 9:48
18650 or other cased batteries are in steel casing which is magnetic,
Oh, I do hope you just didnt care about the tune, and were only focused at making compass work. Because, I dont really need any of inav features if the quad flies like this. Following.
Thanks for making our FPV lives easier. One of only couple channels I recommend to people.
Fantastic work, Joshua! Thanks a bunch for the tutorial! 😃
Stay safe there with your family! 🖖😊
Thanks for the inav video. Would love to see a video how to configure HDZero with Inav.
Thanks man I was struggling with the same issue for a while
What about putting a ferrite core on the cable as close as possible towards the connector?
Thanks!!! This helped so much with my AOS UL7. Can you please do a basic inav tuning video??
I use iNav a LOT on my planes. But use BF for my FPV freestyle quads. Even on my 5" GepRC with a M10 GPS/Mag. I have never gotten a good RtH with the mag enabled, probably because I have the battery connector right back there by the compass, just like you did. I'm curious what your impressions are with BF 4.5 now that you've "fixed" the EMF issues on the compass???
Brilliant video! As far as foil tape for the gps wires , for years I have used copper foil tape and you can solder a ground wire to it easily and ground anywhere .
What if you shield everything? I would use the mesh shielding from an old rg59 tv cable?
Or just mount the compass on top of the battery strap?
What if you twist all your positive and negative wires as a "twisted pair"? Twisted pairs offer some shielding with zero cost and zero weight gain.
just use a stock usb cable cable for conect the gps, i use a geprc m10 DQ with a usb cable from a p30 lite huawei smartphone, 13 cms long, enough long from motors and no interference, rth work superb, also almost always above 30 sats fix
used in a mark 4 7 inch
core thing about this, it comes shielded...
Curious how DJI dealt with compass placement issues on their small drones like the AVATA or mini
It's more like the wonders of programming.
Battery casing acting as a waveguide, maybe? Copper tape
?
Would like to see more ardupilot video, it really is excellent at automatic flying and good enough for acro flying. iNav simply just losses some reliability features but simpler to tune
I wounder if you can make a cinewoop hold in the air like an avata
I started using Inav a few months ago and have used BF for years. So many more features on Inav but BF does fly better imo...
am have apex hd5 frame and I created 3d model , now gps is mounted to gopro holder with 0 degree crouch and I turn on galileo ... in past I had max 9 sats and now I have 23-35 sats ...I am using hglrc m100-5883 gps and it's working nice with compass ...
image : drive.google.com/drive/u/1/folders/1NWxARE8Q61q5HfAsJ-9iXVGmBJKer2Oi
If it took Bardwell 2 weeks to figure it out, I'm screwed!!! Lmao
HELLO joshua, thanks for the video, we learn each time; by the way i see on your o3 video the same OSD display faults: question marks are displayed instead of units (mW, arrow, etc.). Did you not manage to resolve this display problem? If yes, how? best regards from France
You should defy do a PIDS and filter setup “how they compare and or transfer if your coming over from betaflight”.
I also recently bought three of the Flywoo Goku M10 mini GPS with compass. But neither INav or Betaflight wants to recognise the compass. Pretty sure my wiring is correct.
lol , I have a Martian II as you know a knock off from banggood from 7 years ago, with a banggood, ómnibus F4, with Inav from 5 years ago, 2.6.0 and does the same as my DJI Malik, lol just perfect flight modes 5 years ago
if you buy the compass separately and place it anywhere without touching the GPS navigator. I just don't know if there are special compasses on sale.
Nice ! Love to see you install ardupilot on this quad :)
I only have 2 questions:
-why the quad is shaking so much and if you managed to fix it.
-what are the 2 osd elements showing the degrees?
Just one question: did you watch the video?
@@mixmasteru 1:?;2:OSD_GROUND_COURSE/OSD_HEADING and btw I used less chars to respond than you...;)
Man, been out of the hobby for years and it's crazy to see the advancements but also crazy to see the same struggles. lol Side note, have you not tuned the PID's? or is this the best Inav can do?
Inav is easy to use, i use pos hold and cruise for fishing drone.
18650 Vs lipo is probably because of the steel cases of the batteries cells having higher magnetic permeability.
That is the steel cases of the 18650s are acting more like an electro magnet Vs the aluminium cases of the lipo cells.
I have dehydrated my TPU and tried everything I can. It's so stringy and globby. Your mount is so impressive. I am forced to use PLA or PETG because I just can't get anything useful out of TPU :(
Some cables will have a foil shield built into them to attenuate electronic and magnetic interference, would wrapping the battery lead in Mylar from a computer motherboard bag help?
So a GPS Waypoint mission is next? then onto Fixed Wing, then you’ll be good for another 50 thousand subs?
I use 3m copper adhesive tape, its ridiculous expensive but 1 roll should do like 200 qauds soo (worth it?)..... and i always solder it to a ground and my m8Q's will "insta" lock 25+ sats on my chimera. :)
I never even thought about this too deeply... I flashed my 7" with inav and it just kinda worked... I suppose my battery lead is far enough away
Great video Bardwell as always.
I name that drone, "The Mayfly."
Is shielding worth it. E.g. aluminium foil or copper tape?
Ive been wondering about better shielding a little lately.
Adding foil tape doubled the number of GPS sats I locked.
Would it be necessary to ground the shielding to the quad for better results, like twisting a piece of aluminum, isolating it with heatshrink and soldering it from the shielding to a ground spot on the quad?
I think that would improve its effectiveness, but my tape was not soldered, and still made a big difference.
Ok question. If I wrap battery leads from esc in copper and maybe gps wires in copper tape would that help.
Right now just have kaplon tape to hold gps wires down to not touch battery leads
use a thin USB cable or Ethernet cable instead.
Have you (or anyone :)) tried shielding the battery cables with braided metal cable shielding. I'm guessing that the braid would need to be grounded to the frame. What about a ferrite bead?
In the iNav Alignment Tool, should the heading that moves as you move your quad be the correct heading (90 for east)? I had to mount my flight controller pointing backwards (180 degrees) for wiring reasons. My GPS points forward. No matter what I do, the compass is not correct. I tried setting the Alignment and re-calibrating, but it still tells me the wrong heading in the alignment tool.
Btw , how you change arrow (direction to home) in inav ?
I can't thank you enough!
@JoshuaBardwell Why does it look like the spacing between plates at the rear of the top plate is twice as high as the front? It looks like the two printed pieces are stacked on each other from what I can see at the 24:49 mark? It looks like the side plates have to come completely off, at least on one side for the battery to be routed to the side. Did you fly it that way?
Hi I searched for the stl files for that mount for the gps and antenna and I couldn't fine it in thingiverse can you share the link or the file if you may
What about using 2 gps modules. One in the front and one in the back and then calculate the heading from their offset. 🤔
standart precision gnss receivers has to be at least 1.5m away from each other for that to work
I think a lot of people don’t want to try iNav because of these problems. How did Dji do it so right over 10 years ago with the Naza v2/lite. Just install and go. Maybe they can copy what dji did right so long ago. This is not to downplay the work the devs do on iNav. It’s an amazing firmware that can do considerably more.
Quads were bigger then so greater clearance distances also most builders put the GPS?Compass unit on a tall mask. not many NAZA FC's went on 5" quads
A lot of factory quads at that time positioned the GPS antenna and compass in a place that goes and doesn't have electromagnetic interference with the quad or on newer quad they have copper/aluminum shielding between the GPS and electronic components of the quad. This info gets overlooked when buying ready to fly quads or DIY builds because most people use Betaflight without the GPS capability which got added relatively recently.
And in this episode... Bardwell learns middle school physics and electronics.
Is the tune really bad on that quad or is that just how iNAV flies? Sure looked really shaky during your manual parts of the flight.
Omg great video. I have inav on my LR 7" & 8". Wanted after done testing put inav on my 10" but I do have my lead in the back next to compass. Wonder if that why on my betaflight ( I have compass on I notice it's off when on by like 20° when on not just on usb)
On my 8" I had to push crossfire antenna wire further away. Also wrapped in copper trying to help. At first compass wouldn't work.
My chimera7 compass is located exactly where yours was, and is working perfect. I think you got the alignment wrong. A drift when moved, shows a conflict between the accelerometer and the compass.
I can also see that in the first video, you didn’t align it like in this video.
The first video was the factory alignment exactly as iFlight delivers.
I know, but on the first video you didn’t set it to flipped, and you had drift. When moved, the heading reacts to the accelerometer first, and then drifts to what the compass says.
Hi JB, what are those two numbers in the OSD just above the compass heading bar? and what is the significance of the differenece between them? Thanks 🙂
The left one is the direction that inav believes the drone is facing, based on all available sources of information (gyro, compass, GPS, etc). The right one is the GPS heading.
Hi JB , can you point me how it is called in osd tab in inav ? Looks i found one but another not thank you
@@JoshuaBardwellcan ypu point me whats the name of those 2 element in osd tab in inav ?
My heading doesn't move, and I don't see any drift on the compass but when I use poshold or RTH I get serious toilet bowling so no idea what's up with it.. inav says hardware health is fine
Am I understanding correctly that iNav uses a Kalman filter (the "sensor fusion" part)? The more I get into FPV, the more stuff I've learned about in University makes sense :)
Good video Joshua, but you need to tune that bird, it was NOT flying good. I have a bird or two on older Inav and it works just fine.