• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

KK2.1.5 board, Gyro problems???

#1
Hey guys!

So my dad and I just built our first quadcopter (Electrohub quad kit with the KK2.1.5 board, generic HK 30a esc's NTM Prop Drive motors) and we took it for a test flight yesterday afternoon. When we tried to put a bit of power and started increasing throttle to the quad, it suddenly flipped really fast to one side and broke on of the booms (so much for the wood being basically as good as carbon fibre, Alex :p). Anyways so we tried it again and it flipped again. So we factory reset the board and went through the settings and redialed them in and recalibrated the gyros. When we went in the "Sensor Test" after we callibrated, we noticed the numbers on most axis of both the gyro and the accelerometer when jumping around in about a 10 digit "radius" (i.e. the number jumped around within about five numbers up and five numbers down). So we decided to do some testing. I held the quad in my hands, arms outstretched and my dad increased the throttle very slowly. Quickly I realised that the quad was pulling different ways. So it would start trying to flip left and then right then suddenly a huge burst of power from the right side forced it to twist in my hands. Luckily my father was quick to turn off the motors. After some extensive testing back at home, I have noticed that whichever motor I have plugged into the no. 3 motor slot on the board, after I have increased throttle fractionally, starts slowly spinning up faster than the other motors. While this is happening, all other motors are decreasing and increasing in throttle slightly. The when I picked up the quad and tipped it to the left. The left motors decreased throttle, the opposite of what it was supposed to do. Also as I dropped the right side, the right motors lost throttle, again the opposite of what they are supposed to do.

After researching extensively I can't find anything on this problem. So please if anyone holds any special knowledge on this please advise me of what is going wrong.

Thank you to anyone who read the entire thing and considered helping out :D
 
Last edited:

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#2
Hey Flyer,

Welcome to the forum!

The gyros on the board are super sensitive. I'm not surprised the numbers bounce a bit. from your description (not positive I understood it, but . . . ) it sounds like you're fighting with the control loops, not the gyros themselves.

More often than not, when a quad wants to flip on it's own the second it leaves the ground (aka "insta-flip"), there's something wrong with the layout of the motor connections to the flight controller, and it's typically order. Effectively, it's trying to keep the frame from rotating, but somehow it's giving inputs backwards, so the correction just makes it worse . . . so it corrects some more . . . and it's on it's back before you can cut your throttle.

If you take off the props, throttle up to 1/2 and hold the airframe, when you dip a boom the motors should resist the motion -- the dropping motor should speed up and the rising motor will slow down. Until you've got this right, there's no reason to put the props back on. It will flip every time.

Make sure the screen is on the top and buttons are closest to the rear, and check your ESC connections and motor directions against the motor layout screens. If it all still looks right, but it's not responding correctly, a few pictures of your setup or a video will do wonders for helping us troubleshooting it.
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#5
Yeah, it appears your plugs aren't in the right order.

yours:

M1->Front Right
M2->L (not sure which end)
M3->L (not sure which end)
M4->Rear Right


It should be:

M1->Front Left
M2->Front Right
M3->Rear Right
M4->Rear Left

In this case it's trying to lift the left side when it drops by speeding up the right side's motors (because it thinks they're on the left). Swap those 4 connectors around to get the motors where the board thinks they are, and it should fly :)
 
#6
Thank you for the reply again!

I will do this as soon as possible, however I do not understand one thing. How would the motor layout cause one motor to steadily increase throttle even when it is stationary?

Thank you!
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#7
Three things might be going on, I'm leaning toward the first:

- Because it's not responding to the correction, the "error" is accumulating over time and so the correction to reduce the error grows over time. There's also a fair chance you've not de-vibed your motors/props and you're getting vibration noise amplifying this effect.

- There is also the possibility the ESC is warming up, and as it does, the "resistance" in the ESCs drop a touch . . . but so will pack voltage under load . . . not a strong effect.

- You've got a soft-start programmed into the ESC and it's taking it's sweet time to go to the commanded throttle. if you haven't dinked with the ESC programming, this probably isn't an issue.

BTW, if this is board hasn't been reflashed, the stock firmware also has some bugs . . . and some odd behaviors. It's fine for testing the hardware and will give you a flyable craft, but I would *highly* recommend re-flashing your board with Steveis or RC911's latest ROMs. It will make a difference.
 
#8
Thank you so much

Thank you so much!

Thanks to you my quad is now hovering! I switched the motors that you told me to switch and now everything is good! Amazing! You are my hero! Thank you!

Off to go fly now! (and probably crash too xD)
 
#9
Thank you for the suggestions!

I have not balanced my props yet as I was so eager to fly however I will do that now. I have also been looking into getting one of those USB devices that I can use to flash my board with better software.

I will balance my props now and hopefully take the quad out to the "field" (soccer club) this afternoon and see how it does.

Thank you for all the help! :D
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#10
Flyer,

Congrats on getting it in the air -- hard to tune a craft until that's done ;)

Keep working on it, and don't hesitate to show off your new build in flight with a video or two :)
 

Xonos

Junior Member
#11
I know this is an old thread but I found it and figured I should tell you a story. Last year, I bought the KK2.1.5 - flashed it with the Steveis firmware, bought Multistar escs (biggest mistake ever) and multistar motors (also another mistake). If you find out after some time that you just get this wobble (or instability) with the deadcat / spider configuration using the electrohub & kk2.1.5 - trust me... You are not alone. The solution is not a simple one. My brother bought an identical setup as me but with different motors/esc's with a standard X configuration which was flawless. Meanwhile, my deadcat constantly wobbled and would never fly as elegantly as his did. I crashed a few times which I hadn't realized, damaged the motor bearings on a very small scale.. enough to cause additional wear over time. Determined to correct the issue, I did everything I could think of. Months of tinkering, researching & replacing lead me to the best thing I could've ever done. I bought high poll count ESC's with the notorious firmware + heatsyncs and T-Motors (anti-gravity ones to be exact - my bro got a great deal for me on ebay). As soon as I installed them, my quad floated like a cloud and it resolved all of my issues. Now, I'm not saying you bought intolerant equipment like I did but what I am saying is, if you think your configuration 'should' be working - don't rule out the possibility of hardware issues. What I thought was a configuration-based wobble in my flight was actually a motor grinding marbles in my attempt to fly aggressively. (I took them apart to confirm bearing damage - looked like WW2 after popping off the bearing). Since then, I've learned quite a bit and I'm going to retire my kk2 to a mini h quad racer for fun - ill be graduating on up to a multiwii for my arial footage stuff.

TLDR; don't waste money on cheap esc's and motors. The greedy pay twice.
 
#12
Same Issue

Hey FliteTest guys,
I have the same issue, since I fly with the KK2.1.5 HC instead of the naze. My frame is the sk450 deadcat, i also calculated new angels and the quad fließ better than my friends dji 450 with a naza! But I still have a very close Problem: the quad flips by giving hard and agressive throttle. And one more thing; sometimes (not very often, ca every 3rd flight) the quad flips whether just hovering! Are there any solutions? I'd be very thanks to if there were any comments or answers soon.
Jacob
 

acsrandd

Junior Member
#13
Issues with KK 2.1.5 FW V1.19 uncommanded change in flight

My Quad. X config., Firefly standard 450mm frame with generic ESCs (blt. in BECs only one BEC powers Quad). I have approx. 13 flights on this quad. (HW Ver. KK 2.1.5 v, FW Ver. 1.1981 Pro Steveis).

This is my second Quad w/KK 2.x.x controller. First quad crashed while in flight (one motor just stopped!). ESCs and motors were not defective, nor was the RX, no apparent reason for the in flight failure (KK 2.0 was damaged).

Shortly after liftoff while hovering at approx. 30 ft. the two forward motors (motors #1 and #2) stopped!!! Post-crash these motor run with no issue, as do the ESCs, and KK2.1.5 survived no issues, RX survived. It is like the KK2.1.5 FC interpreted a command to fly forward at max. speed, but my Tx sticks were commanding to move forward at a slow rate. Anyone else experiencing anomalous behavior i.e. sudden uncommanded change in flight.

Post Crash Readings from KK 2.1.5 FC:
Sensor Test:all OK
ACC Calibration: Performs Cal. OK

Debug:
GyroRollZero: 505
GyroPitchZero: 509
GyroYawZero: 513
accXzero: 506
accYzero: 507
accZzero: 642
ESClowlim: 448
BatAlarmvolt: 428
ServoFilter: 39
VectorX: 00000000
VectorY: 00000000
VectorZ: 00000000
Accangleroll: 1
Accanglepitch: 1
BatteryVoltage: 485
Vectorlen: 00FFFFFF

Anyone have any idea of what may be going on or any suggestions?