Home › Forums › Everything about the Tricopter V4 › F3FC & iNav experiences
Tagged: GPS, iNav, Loiter, position hold SPF3, tricopter
- This topic has 163 replies, 22 voices, and was last updated 6 years, 2 months ago by Marlon.
-
AuthorPosts
-
27 October, 2016 at 04:00 #34380DaveLParticipant
I just ignored the motor-direction arrows in the configurator and wired up the motors to get the directions recomended by Triflight. Works the same with Cleanflight/Triflight/iNav, no rewiring required to switch firmware.
27 October, 2016 at 11:44 #34382oxivanisherParticipant@DaveL Thanks for the information, great news. 🙂
27 October, 2016 at 18:10 #34385billydParticipantOr you can just use the cleanflight connection in BLHELI suite and reverse the motor direction in the esc fw. Seems easier to me than resoldering wires.
28 October, 2016 at 00:01 #34391DaveLParticipantI took some pictures of my GPS-Mount, it is a 1/4 inch solid plastic rod with some rubber tubing at each end, attached to the tail boom with zip-ties. It is a very nice solid attachment, with a bit of elastic flex, and the zip-ties will break in a crash without damaging the boom. The top has an automotive barbed plastic fitting jammed into another piece of rubber tube, and some double-sided-tape (henceforth called dst) to hold the gps puck in place, around 18cm above the tailboom.
I’ll replace the plastic rod with a lightweight carbon tube some day.Attachments:
28 October, 2016 at 01:16 #34395billydParticipantIf you add a sheet of carbon fiber underneath the GPS antenna, about 50×50 square, it will lock sats faster.
31 October, 2016 at 16:00 #34461oxivanisherParticipantI have a strange problem, and i don’t know if my servo is bad or if I have to configure something. I just installed iNav and used the config for 1.3rc2 posted earliert (and changed rx and modes and so on).
But my tail servo is behaving strangely IMHO. It looks like it takes “too big steps” and if i try to hover, it sometimes stutters on the z axis back and forth, as it would tell me “no, no, no”. 😀Do i have to set the servo ways or anything like that? I did not find any documentation (remember, it is iNav, not triflight).
Edit: I forgot to mention. No MAG/GPS at the moment (waiting for a Ublox8 with no pull up resistor like the Ublox7 one has). ACC calibration seems to have worked, accgain_x, accgain_y and accgain_z are not 4096.
31 October, 2016 at 18:43 #34466wilco1967ParticipantHi oxivanisher
Isn’t your problem not simply a too aggressive Yaw – P setting ? Try lowering yaw P (reduce to half the previous value or so) and see if it goes away…. halving the value might be too much, but you will quickly find out if you’re on the right track. it should still fly, but may become less ‘crisp’
For I-Nav, you don’t have to set the servo limits or anything. It also doesn’t make use of the feedback wire like Triflight… As far as I know, you may leave it connected, but it will simply ignore it. I have an ordinary tail servo, without the feedback wire… works just fine.
Did it work with cleanflight / triflight / whatever ?
if you’re unsure if the servo is ok, just hook it up to a normal receiver (bypassing the flight controller alltogether), and test it. I had a few that were missing some teeth in the gears after a crash… that may seem to work, but gives very erradic flight… if it doesn’t move smooth when testing, replace the faulty gears, (or the entire servo).
Another thing you may want to check: Is your servo update rate set at 50 hz (set servo_pwm_rate = 50) ? I see no reason to increase it, and some older (analog) servo’s don’t tolerate any higher as 50 Hz (apparently, never tried it…).
Let us know if this helps…
1 November, 2016 at 08:40 #34475oxivanisherParticipant@wilco1967: Thanks for your succession! I tried lowering the yaw P value (from 85 to 40) and it looked better (holding it with my hands at about 5cm off the floor) meaning the servo did not first wait and then jump to 30° and thus overshooting. The yaw input seemed to react better, although I still feel like the reaction takes too long overall from yaw stick movement to the servo reaction. Since we now have the superfluous daylight saving time here, I will probably not be able to test it until Saturday “in the wild”. I will keep you posted.
1 November, 2016 at 18:33 #34485oxivanisherParticipantI still think something is wrong with my tail servo. It still feels wrong.
The Taranis is configured in the iNav receiver tab (1000/1500/2000) as it should. And as far as I remember from the BatBone TriCopter, the movement was much more extreme.I made a video of the servo movement:
Do your servos also move only this little distance?
7 November, 2016 at 12:11 #34633krkljanParticipantAny new videos or experiences with INav?
7 November, 2016 at 12:34 #34634laukaParticipantI got my V4 doing 3D position hold pretty well.
Shortly what I changed:
– Added Beitian M7N GPS on i2c
– Added separate HMC5883L compass on the tail boom, 50 mm apart from all electronics.
– Sandwiched the front motors between 10 mm soft foam pads.Then I burned something on my F3FC while connecting the battery. 5V bus is now short circuited. So that was the end of my iNav testing :).
7 November, 2016 at 12:49 #34636krkljanParticipantSorry to hear that!
Any possibility to merge Inav and Triflight?
7 November, 2016 at 23:21 #34661oxivanisherParticipantSince Cleanflight is the common code base, it should probably go Triflight >< Cleanflight >< iNav. This is not a hard requirement, but probably the best way to do it.
This is the issue where Mr. Cleanflight himself asks for this, but sadly there was nothing going further in that regard: https://github.com/lkaino/Triflight/issues/3027 December, 2016 at 15:45 #35638krkljanParticipantHas anybody tried INAV 1.5 RC?
27 December, 2016 at 16:20 #35640DaveLParticipantYes I flew it first time yesterday! I did a series of short tests with alt & pos hold (cruise option) to try to figure out how to tune out wobbles. It holds position well but when I move the tricopter in pitch, roll, or yaw, it wobbles like a canoe. It is very steady in angle mode, just wobbles in pos-hold.
I want it to be steady like a flying tripod when yawing, and make smooth transitions when I begin and end movements in cruise mode. Currently the wobbles make it unsuitable for video (and inelegant to watch it flying).For those of you who are new to this (like me 🙂 ), cruising during position-hold means being able to aim and move the multicopter at a slow speed with self-leveling (angle, not acro/rate) and have the flight controller compensate for wind to prevent drifting. When you center the pitch/roll/yaw it stops moving and holds at the new position.
I need this because I am a new & unskilled ‘pilot’ and wish to have assistance while I learn. And I want to make smoother video. And give friends and family a few minutes of supervised fpv hovering/sightseeing. And not have the thing drift off in the wind and find a tree 😀
Btw, it is very steady in angle mode, just wobbles in pos-hold.
INAV works great for this, I just need to find out which pid parameters to adjust to tune out the wobbles.
Cheers!
Dave L -
AuthorPosts
- The forum ‘Everything about the Tricopter V4’ is closed to new topics and replies.