I use my drone as a home inspector to view inaccessible roofs. I was about two inches from landing when my typhoon H stopped responding to controls. It banked hard forward and I was able to kill the rotors before it crashed into anything however the front two rotors ate a potted plant. I cannot figure out why it stopped responding to controls and now I’m getting a rotor 5 error. I am reluctant to fly it again. Any suggestions?
Near flyaway?
- Thread starter Rhinosaurus
- Start date
If the 5 rotor error was caused by the contact with the potted plant, it should clear when shutting the motors down and restarting. That would be the more common occurrence. It is also normal to get another 5 rotor error if you try to spin the motors above idle with no props attached.
Another occurrence is when there is actual damage to a motor or the ESC board. That one could occur as soon as you powered up the drone without even starting the motors, or as soon as you start motors, even if you do not raise them above idle.
Do either of these scenarios fit what you see? If not, please provide more detail of what you are seeing.
You can also upload your flight logs for review. That could help identify the cause of the original unexpected movement. If you need help uploading the flight logs, refer to Attachment 57 (Typhoon H Flight Log Retrieval) of the PDF document attached to "Way To Fix Drones project".
Another occurrence is when there is actual damage to a motor or the ESC board. That one could occur as soon as you powered up the drone without even starting the motors, or as soon as you start motors, even if you do not raise them above idle.
Do either of these scenarios fit what you see? If not, please provide more detail of what you are seeing.
You can also upload your flight logs for review. That could help identify the cause of the original unexpected movement. If you need help uploading the flight logs, refer to Attachment 57 (Typhoon H Flight Log Retrieval) of the PDF document attached to "Way To Fix Drones project".
Log number 39 was the one where it acted up on me. Log 40 was when I was at home troubleshooting. I looked over the flight logs but I am not knowledgeable enough with it to be able to decipher what I see.
@Rhinosaurus,
I can see part of the issue. When you pushed the Red Button, you were holding down J1 to land. It appears you inadvertently pushed the Yaw stick when you pushed the button. That much is relatively normal. What happened next is NOT normal. You did not get a "Yaw" response. You got lots of "Pitch" and "Roll" response.
I'm still looking for clues. But please check your controller. Verify it is set for Flight Mode 2, and check the "Final Output Display" (fast double tap the Volts icon on your ST16) to ensure the correct channels are responding, and none are duplicated. There is a chart in Attachment 36 (ST16 Signal Processing Overview) of "Way To Fix Drones project" to help read the final output display.
I can see part of the issue. When you pushed the Red Button, you were holding down J1 to land. It appears you inadvertently pushed the Yaw stick when you pushed the button. That much is relatively normal. What happened next is NOT normal. You did not get a "Yaw" response. You got lots of "Pitch" and "Roll" response.
I'm still looking for clues. But please check your controller. Verify it is set for Flight Mode 2, and check the "Final Output Display" (fast double tap the Volts icon on your ST16) to ensure the correct channels are responding, and none are duplicated. There is a chart in Attachment 36 (ST16 Signal Processing Overview) of "Way To Fix Drones project" to help read the final output display.
That chart makes sense for the actual incident, looks exactly like what happened now that its all mapped out. My ST16 is in mode 2 so that's all set, and when I look at the final outputs chart i can see only one channel moving at a time per joystick push. I get a slight wiggle from the readout on all of them but its a brief wiggle, that's it. The default of my CH1 throttle is at 50%. Full back only drops to 20%, and full up rises to 80%. Looking at the chart for typical values it would indicate 0-100 for a range. Could that be part of my problem? Thank you for your time by the way, I really appreciate it. I was contemplating heading to a Best buy tomorrow to buy another drone until i could figure this one out. I can't really afford to do that but i can't afford to not have one either. I have a three story Gambrel house to inspect on thursday and will not be able to inspect the roof without a drone.
It would be a good idea to check Hardware Monitor before we go too much further. If the sticks don't look right there, it might be easy to begin the fix.
Have you done anything in either Channel Calibrations or Channel Settings recently?
Have you done anything in either Channel Calibrations or Channel Settings recently?
By the way, 80% up and 20% down is normal for throttle if the rate slider is in "rabbit" "turtle". I assume that's where the slider is?
Note: corrected bad comment.
Note: corrected bad comment.
Last edited:
I have done anything in channel calibrations or channel settings, wouldn't know where to begin. And my throttle slider is in "turtle". I have never moved it out of turtle because the flying I do requires me to be hovering around chimneys and rooflines and I would rather a turtle pace than a rabbit pace.
This is a small side issue, but landing in turtle mode causes many TH drones to hop, bounce and even tip over. Probably not related to the current issue, but might be something to consider. Most folks put the slider full up when they are at the landing point.
The main problem is that I can see something of what happened, but do not understand WHY it happened. And I can't explain that some of the stuff started moving a little BEFORE the yaw stick apparently moved.
The most experienced log reader, @h-elsner lives in Bavaria. It's a few minutes after 3 AM there right now. @DoomMeister and @Steve Carr are both better at this stuff than I am. I hope they will see the thread and be able to comment.
The main problem is that I can see something of what happened, but do not understand WHY it happened. And I can't explain that some of the stuff started moving a little BEFORE the yaw stick apparently moved.
The most experienced log reader, @h-elsner lives in Bavaria. It's a few minutes after 3 AM there right now. @DoomMeister and @Steve Carr are both better at this stuff than I am. I hope they will see the thread and be able to comment.
My bad. I just realized why you responded the way you did. I said "rabbit", but meant "turtle".
Thank you so much for your help. And it definitely started to have a mind of its own. I felt like i was trying to catch up to it. Thank you again, i really appreciate it.
I’ll try to take a look tomorrow. I’ve been busy with golf all day. It actually paid to play today!
Just got home a couple hrs ago myself. But I don't play for money. ? ?
I suspect there is actually nothing wrong with the H or the controller. The "bouncing" roll and pitch is indicative of the H doing a bouncy landing. This happens very fast when the H touches one side of the gear and tips the aircraft. The flight controller immediately tries to correct the tilt which causes the H to bounce across the ground, creating that same pattern in the telemetry. When this happens the flight controller ignores all stick input except for the throttle. So when a landing goes wrong you can only recover by going max throttle to get the H back in the air.
@WTFDproject is exactly correct. Takeoffs and landing should always be done in Rabbit mode. If you prefer to use Turtle you should use the Red Button landing. When you are 6-8" above the ground, press and hold the red button until the H lands and shuts off the motors.
Do another test flight after you remove the camera and do a compass and accelerometer calibration. If there is a problem with one of the motors you will get a 5 rotor error as soon as you start the motors.
If you get past the 5 rotor issue and can do the flight test as recommended by @Steve Carr, you can also check to see if Yaw control is affecting tilt and roll. Give it some Yaw. A little at first, more if it looks good.
It should stay pretty much in the same place and just rotate (center of rotation might not be the center of the drone, but it should not be far off.) If it starts actually spiraling outwards, you've got a problem.
h-elsner
Premium Pilot
- Joined
- Mar 23, 2016
- Messages
- 1,949
- Reaction score
- 1,643
- Location
- Bavaria / Germany
- Website
- h-elsner.mooo.com
Looking at the area and landing position in GoogleMaps, too much trees and close to the building. I think the GPS coordinated are not precice due to multipath receive. This theory supported by distance computing from given coordinates (1). There are a lot of small hops that the flight controller tries to level out. Also gps_accH is getting poorer (2).
This may result in unexpected movements of the drone, just that what we see in pitch/roll (3). In combination with turtle mode the pilot is not able to go against that. This effect is more rigorous than lower the drone is. Also the landing point seems to be to close to the building. Buildings are a common source of multipath receiving.
There are some indications that the position of the RC is not good. The Controller GPS comes and goes frequently during the whole flight.
At the end of the flight the frequency of received telemetry data sets is lower than the rest of the flight. Question: Do you have a ST16 with 3 external antennas or with two? How is the direction of the antennas? Both in paralell or 90° to each other?
- In difficult area with many obstacles, bad GPS receive or during landing use rabbit mode. You have more power to correct things.
- Also RC should have godd GPS receive.
- If there is no free area around or near walls, rocks and so on fly with switched off GPS. At the end its easier.
- Have an eye to antenna directions. Do not point rod antennas to the drone.
br HE
This may result in unexpected movements of the drone, just that what we see in pitch/roll (3). In combination with turtle mode the pilot is not able to go against that. This effect is more rigorous than lower the drone is. Also the landing point seems to be to close to the building. Buildings are a common source of multipath receiving.
There are some indications that the position of the RC is not good. The Controller GPS comes and goes frequently during the whole flight.
At the end of the flight the frequency of received telemetry data sets is lower than the rest of the flight. Question: Do you have a ST16 with 3 external antennas or with two? How is the direction of the antennas? Both in paralell or 90° to each other?
- In difficult area with many obstacles, bad GPS receive or during landing use rabbit mode. You have more power to correct things.
- Also RC should have godd GPS receive.
- If there is no free area around or near walls, rocks and so on fly with switched off GPS. At the end its easier.
- Have an eye to antenna directions. Do not point rod antennas to the drone.
br HE
The bouncing is exactly as you described it, and the controller completely ignored all input. Its as though you were there. I always fly in "turtle" because I am always near buildings but i will put my H into "rabbit" when landing from now on. The five rotor error has not repeated itself.
You guys are all amazing! Its pouring down rain here for the first time in almost three months with lots of gusty wind, so an attempt at re-calibration and test flights are out for today. As soon as the weather cooperates I will do the compass and accelerometer calibration and give it a go. I will be sure to keep you all posted. Thank you @Steve Carr, @WTFDproject, @h-elsner and @DoomMeister for your time and input, you have all saved me a bunch of time and money. I cannot thank you all enough.
Similar threads
