Solved Can't get tail servo working properly on tricopter...

[Aerobot]

After installing Triflight 0.7 Beta 3 onto a Naze32 rev.6 FC, the tail servo is jutting left and right uncontrollably when initiating tail tune mode. The FC was flashed using Baseflight Configurator: 0.68.3 and configured using Cleanflight Configurator: 2.6.0. The FW line from the dump is:

• Triflight 0.7 Beta 3 / NAZE 3.1.7 Oct 21 2017 / 13:51:32 (17fa321)

Motors spin up fine. Receiver is an X8R and I'm using the Taranis X9D + (2019) radio. I'm using the hardware from David W with the Bluebird 210 DMH with the feedback wire preinstalled. If I could land this on my FC somewhere and get the added accuracy of the servo's position with the FW, that would be ideal.

Thanks to @LitterBug and perhaps others, I hope to be in the air soon. Thanks in advance for any and all help.
@CarolineTyler : I'm glad you were able to point me in the right direction, and you gave me the energy to not give up on my tri.

 

[LitterBug]

Digging out my tricopter hat.... I have quite a few tricopters and most are in some sort of reconfiguration. I really need to get them all in the air again. 2 Baby Tris, 2 mini tris, a 550 monster HK fiberglass tri, and a 3D printed nano tri. My only currently flying tri is a baby tri with a F3FC that I didn't load up newer firmware on. I need to look at the firmware I have stashed away on my laptop to see what I have capable of running on the naze, or at least get a list of other compatible boards. If I remember correctly, the 0.7 triflight was not my favorite due to low speed oscillations.

 

[LitterBug]

Looking through what I have available, the dRonin GCS (configurator) only runs on linux. The FCs supported look to be:
RCExplorer F3FC
BrainFPV RE1
Luminer LUX
SPR F3 EVO
DTFc
Matek F405 CTR
Revolution
Naze32 Pro
Omnibus F3

 

[Aerobot]

I may have a working tri now. I just reflashed that same FW and the tail tune feature seems to be well behaved. Stay tuned, I may have a working model soon. No need to do any digging at the moment. However, if I find a roadblock I'll be in contact with you again. Thanks @LitterBug

 

[LitterBug]

Good to see there are still a few of us insane tricopter peeps out there!

FYI, Pictures would be greatly appreciated!

 

[Aerobot]

FYI, Pictures would be greatly appreciated!

LOL. What you'll see is David's F3FC tricopter with a hole drilled through the PCB and wires going from the NAZE 32 through the hole to the arms of the tricopter. But I'll share them anyway. I hope to build a couple homemade versions for show and tell.

Also I'll try to load the dRonin onto my Ubuntu at some point to take advantage of some of the other features that are mentioned for the tricopter setup.

 

[LitterBug]

I have a few threads on here of my builds....

 

[Aerobot]

@LitterBug I was able to get the modified F3FC tri in the air yesterday. After some configuration woes, the tail servo began to work as expected, though some unpredictable moves at times. In the end, I really didn't use the tail tune (servo tune) configuration on the bench. Since I don't have the feedback wire from the servo landed anywhere on the NAZE32 rev. 6 board running Triflight 0.7 Beta 3, I was not able to perform the servo tune on the bench. Somehow, there were two servo channels that were dependent upon one another (I think it was #6 & #7). I had to set high, low & mid at 1500 to get the mysterious second servo in the firmware to be inoperable. When they were connected together, I could not get the physical range of the servo's swing full left, but had no problems getting the full range to the right. When I armed the copter and sent it up (stayed aloft for about 2 min - I have a 45 sec video if interested / was not sure the best way to post a 50MB video clip of this flight), it began to bounce, rolling about 15 degrees or so. The pitch seemed to be quite stable. Occasionally, the bird hovered without any rolling. I did not attempt a tail tune mid-flight since I was just trying to get it to hover. I was able to land and disarm on two flights. The maiden flight was using Angle mode with the Headfree (since I didn't know if I would have rotation about the vertical axis or not . As I flew for about 2 min, the yaw did not change drastically, so I was happy about that

Here are the concerns I have after my two short flights:

1. PIDs are definitely not where they need to be for a stable flight.
2. Could there be prop wash into the barometer since it is completely exposed on the top of the FC? Does the barometer have any effects in Angle / Headless flight mode?
3. Is there a landing pin for the feedback wire from the tail servo?
4. I'm still not able to see any telemetry back to the Taranis X9D plus. I've placed the sbus wire on pin #4 and bridged pins #7 & #8 to attach the Smartport wire - both from the X8R. I'd like to have VBatt on the Taranis display. I do have a wire going from the battery to the VBatt+ pin on the NAZE32 rev. 6 and followed the instructions to activate telemetry here.

All this being said, I think the hardware is great. And I'll consider this maiden a success... since the tri did not immediately flip upside down and drill into the back lawn, or look at me and simply smoke

Thanks in advance for any feedback.

 

[Aerobot]

@LitterBug [and any others who might be able to provide insight] I now have a working tricopter. I was able to find the right PID settings yesterday. So, now I have a stable flying tri. However, it ascends at a very slow rate. I don't have extra weight on the frame. Is it due to a Cleanflight setting that I can change? Or, is it the result of have a 3S 11.1V LiPo battery? Any responses would help me out. Thank you in advance.

Setup
FW: Triflight 0.7 Beta 1 / NAZE 3.1.6 Apr 9 2017 / 12:59:47 (bf500bf)
HW: NAZE32 rev.6 FC, 550 mm V4 RCExplorer frame, LittleBee 30A ESCs, BE2217 - 1300KV motors, HQ 8x5 props.
Total weight of tri without battery: 610g
My Admiral 1800 mAh, 30C LiPo has a weight of 150g

 

[JasonK]

@LitterBug [and any others who might be able to provide insight] I now have a working tricopter. I was able to find the right PID settings yesterday. So, now I have a stable flying tri. However, it ascends at a very slow rate. I don't have extra weight on the frame. Is it due to a Cleanflight setting that I can change? Or, is it the result of have a 3S 11.1V LiPo battery? Any responses would help me out. Thank you in advance.

Setup
FW: Triflight 0.7 Beta 1 / NAZE 3.1.6 Apr 9 2017 / 12:59:47 (bf500bf)
HW: NAZE32 rev.6 FC, 550 mm V4 RCExplorer frame, LittleBee 30A ESCs, BE2217 - 1300KV motors, HQ 8x5 props.
Total weight of tri without battery: 610g
My Admiral 1800 mAh, 30C LiPo has a weight of 150g

where is your throttle to hover?
It is probably just be that you don't have enough total thrust to climb any faster. The only way it wouldn't be is if your in some sort of position hold mode and the stickets are directing changes to the position of the 'hold' spot. But given that iNav doesn't even have that working perfectly, I doubt cleanflight does.

 

[LitterBug]

1300kv motors would probably climb better with a 4S battery. I would throw a 4S 1550mah 75C battery at it. I tend to overpower my builds and don't run 3S on anything but my Nano Tricopter 2" build. My biggest RCExplorer build is a slightly enlarged Mini tri, 7x3.5x3 props, T-motor F80 1900kv motors, 1000mah 100C 6S battery. It is a rocketship! Same basic frame as yours with longer than normal Mini-tri arms, but much shorter arms than yours.

 

[Aerobot]

Throttle hover is about 50% throttle on the Taranis X9D plus. My only batteries are 3S. They are several years old. It may be time to upgrade my powerhouse I've only had 30C batteries. I bought them when I was running APM 2.5 & 2.8 on 550 flamewheel quad frames. Thanks for your replies.

 

[LitterBug]

I had to retire a lot of my older batteries recently. They used to be great, but all had at least one sketchy cell that would cause them to sag badly above hover. have a few F450 quads and F550 hexs myself. Pretty fun to pump them up on 4 or 6S with modern FPV FCs.

 

[Aerobot]

Could you suggest a battery for the V4 frame? I'm not sure how many C's to take it to. Obviously the higher numbers the bigger the dollar signs. I don't need a rocket ship, simply a good working tripopter with several solid batteries.

 

[LitterBug]

Could you suggest a battery for the V4 frame? I'm not sure how many C's to take it to. Obviously the higher numbers the bigger the dollar signs. I don't need a rocket ship, simply a good working tripopter with several solid batteries.

I would go 50 minimum. Actually, these sale batteries might be of interest to you.
https://www.buddyrc.com/collections...ice-75c-1550mah-4s-14-8v-lipo-battery_on-sale
https://www.buddyrc.com/collections...e-95c-1550mah-4s-14-8v-lipo-battery-on-sale-1

 

[Aerobot]

@LitterBug - I went ahead and bought one of each battery you had suggested. I'll see what happens when I put that into the frame.

On another note, I do have a loose matek f405 STD FC. If I were to load triflight 0.7 on it, which hex file should I use from the assets in the beta 3 firmware? As I don't see an exact match for that FC.

 

[Liam B]

Also, plug for CNHL packs. I've had nothing but good experiences with ChinaHobbyLine packs.

 

[LitterBug]

@LitterBug - I went ahead and bought one of each battery you had suggested. I'll see what happens when I put that into the frame.

On another note, I do have a loose matek f405 STD FC. If I were to load triflight 0.7 on it, which hex file should I use from the assets in the beta 3 firmware? As I don't see an exact match for that FC.

I need to look at the F405 stuff. Pretty sure I have one on at least one build that the firmware is much newer than the old .5 and .7 versions. Possibly as many as 3. 1 baby, 1 mini, and a nano. At least one had an f7 though. Need to dust them off and see what is on them, and look through old emails with the DEVs....

Edit:
My monster rocket tri has an F405 STD with triflight based on BetaFlight 4.1.0. Need to find the .hex

 

[Aerobot]

If you find that hex file, I'd like to know what it is, and what special settings you had to run motors & servo. Also, I'd like to know how to include that feedback wire from the servo. Right now, it's just dangling on my tricopter cause I don't know how to include it. Thanks.

 

[LitterBug]

Yeah, I can do a dump of the settings and figure out what port(s) everything is set up on. I probably was running F.Port since there is only one wire hooked up on the FrSky RX....
It's all starting to come back to me. I don't think it has been flown with it's current config and this FC. I originally had a dRonin/DTF Seppuku F4 FC and was in the process of beta testing Triflight based on BetaFlight. dRonin didn't have native support for the Matek boards. This thing has probably been in limbo for about two years....