Really questioning why I chose a Gremlin....

silvrrubi592

New member
I've gotten past some of the OpenTX problems caused by a lack of instructions, but I can't rule this out as a possible problem still. There's plenty of info on YouTube on how to work BetaFlight. Having a hard time finding build/setup videos of the Gremlin though.

Spin the motors up in BetaFlight….smooth as silk, and they emit a beautiful in unison hum/whine. Flip the radio on though (when it finally chooses to link up.....) and it's a ratchety, clankety grind as 3 motors spin at different speeds and one motor just sits there doing nothing. Okay, let's re-zero the accelerometer/magnetometer and try again.....same thing. Maybe the table isn't level, lets try the floor. Same thing. What if I pick it up? Well, that may have gotten worse... It's like the balance the ball game. It's always trying to go off in a different direction and it sounds like a coffee grinder. Tried moving trims on the X-Lite, and that only made it go faster wherever it felt like going. Could it be that since I can't make the "mode" switch be recognized/function in BetaFlight, that the FC is in crazy freestyle race mode?

Do other people have this kind of trouble on initial setup, or am I some kind of special?
Anybody want to share their settings for a Gremlin build from BetaFlight (Bixler CF on the 2s Hyperion batteries, all stock kit parts, XM+ rx, X-Lite tx) ?
 

ElectriSean

Eternal Student
Mentor
Spin the motors up in BetaFlight….smooth as silk, and they emit a beautiful in unison hum/whine. Flip the radio on though (when it finally chooses to link up.....) and it's a ratchety, clankety grind as 3 motors spin at different speeds and one motor just sits there doing nothing.

This is caused by noise/vibrations getting to the FC. Can you post some pics of the build?

Also, you should never use transmitter trims on a multirotor. Acro mode doesn't need it at all, and self-leveling modes (angle/horizon) should have the accelerometer trimmed in betaflight.
 

PsyBorg

Wake up! Time to fly!
Seems to me you are locked into setting up a quad like you do fixed wing craft after seeing your other threads.

When you spin the motors in the gui it ignores pids and fliters and all of that stuff which is why they spin smoothly.

I will take a guess and say you messed around with filter settings and all the bells n whistles beta flight does.

Start by defaulting or even reflashing the fc. I then suggest you reflash the esc's to run multishot instead of dshot as it is faster then dshot 300 those esc's do.

Then in your radio delete the model you are working with and make a new one using the airplane script. I have my radio defaulted to a TAER channel map but use what you are used to.

Once those two things are done plug the quad into the computer and set all channel centers to 1500 + - 2 points as it will bounce a bit. Use channel subtrim in the radio to do that.

While you are in each channel disable the trims.

After that use end points to set lows to 1000 and highs to 2000.
DO NOT think the lows are fine at defaults they dial in around 97.8%.
This will allow for maximum resolution as well as help with arming issues sometimes.

The next thing is to calibrate the esc's. No radio needed. Plug the quad into betaflight. Go to motors tab click the box and make sure NO PROPS are on the quad.

Set the main slider to full. Plug the battery into the quad and when the tones start to sound drop the sluder to 0 throttle. At that point the esc's will reboot.

Once you done all that come back and update us and we will move to switch mapping and all that.
 

silvrrubi592

New member
I couldn't get the arm switch set up in OpenTX on the X-Lite. Took a couple hours to find a video that showed setting switches up in 2 places (inputs AND mixers).
I now have an arm switch, and in theory 2/3rd's for a mode switch. Still not sure why it won't recognize third position - problem to be solved later. But it does arm, finally!!
I have no intentions of using the trim switches. I was trying to figure out why it was flying off in a 45 degree angle when I tried to take off. Not pitched at 45 degrees, but using points of a compass. If it's pointing North and South, with North being away from you, sometimes it wanted to go North-East, sometimes North-West, etc., all with zero stick inputs. Never touched filters. Actually, never touched anything beyond the basic radio setup (which was also messed up somehow). I did delete the model I saved and tried again - that did straighten out many issues.
I calibrated the ESC's, and that didn't help. But I have found the problem.

Soft mounting!!!

The initial video I saw was about soft mounting the motors to reduce vibrations to the FC, and that lead to others who soft mounted their stacks. The basic gremlin build is just the SP Racing board, which I mounted just like TJ did in the FT build video....maybe I used the wrong tape? There's no mention of vibration even being a concern though.
I folded 2 small pieces of the double sided tape into tubes (not much room between the plates on the Bixler), and put one in front and one in back, and buttoned it all back up and tried again.

ALMOST BLISS!!

It's not perfect. It still wants to fall off in random directions on take off a little, which I know is kind of normal until it's flying in a direction, but it DOES takes off now. I'm not expecting a rock solid, GPS/Gyro based locked hover. But I do know that it's not supposed to flip itself over on the ground when throttle is applied either. There is definitely some control over it now. Needs to go outside with some room for mistakes for a real test.

Mapping the Mode switch, and maybe fine tuning the settings is about all that is left now. So any numbers shared or direction on that would be very helpful.

I was starting to wonder if the board I got in the kit was damaged somehow during the build, because I followed the build video pretty religiously, and the end result was not what anybody showed.
 

PsyBorg

Wake up! Time to fly!
This is what my Gremlin does. Over weight with real cameras and vtx on 3s. No fancy tricks, soft mounting, filters... all stock on the bixler frame.

Your choice to go with generic videos and "i have no clue how to set up a quad" fad if the moment ideas or you can go step by step and set it up with what actually works...

Your choice but lemme know when you want it to fly for real.

 

silvrrubi592

New member
This is what my Gremlin does. Over weight with real cameras and vtx on 3s. No fancy tricks, soft mounting, filters... all stock on the bixler frame.

Your choice to go with generic videos and "i have no clue how to set up a quad" fad if the moment ideas or you can go step by step and set it up with what actually works...

Your choice but lemme know when you want it to fly for real.

You said "generic videos and "i have no clue how to set up a quad" fad if the moment". I specifically mentioned the FliteTest TJ Gremlin build video. It is their baby I'm building. It's their instructions I'm following. So I'm not sure how that qualifies as a "i have no clue how to set up a quad" video, but then ...I definitely have no clue how to set up a quad.
 
  • Like
Reactions: FDS

silvrrubi592

New member
This is what my Gremlin does. Over weight with real cameras and vtx on 3s. No fancy tricks, soft mounting, filters... all stock on the bixler frame.

Your choice to go with generic videos and "i have no clue how to set up a quad" fad if the moment ideas or you can go step by step and set it up with what actually works...

Your choice but lemme know when you want it to fly for real.

I gave your advice a try, changing the ESC settings from ONESHOT125. Tried DShot 300, not smooth. Tried Dshot 125....looks pretty good on the bench, lets put on props. Okay, it's now almost steady off the ground, will it move in a direction under control? Yes!
CRAP!!!
It seems to be in angle mode, because once I pushed the stick in a direction, it stayed tilted and moving that way. Remember, my mode switch doesn't seem to want to program, so I can't select a mode. So, in the extremely brief time it was out of contact with the ground, almost flying, it hit something that didn't move out of its way. Now, the motor that made contact appears to be dead. CRAP!!!

So, using my limited troubleshooting knowledge: lets swap things out and see what combination still works. The motor spins up fine on another ESC. Again, my limited understanding of all things drone, if motor 1 (contact motor) spins up when connected to ESC 2 (non contact side), the odds indicate that motor is probably A-OKAY. Re-soldered all the connections on the contact ESC, re-attached contact motor, NOTHING. Other then swapping the ESC to another motor (and/or another corner of the FC) how does one check the ESC for function with a multimeter?

I'll add a side note: in BetaFlight, after the motor made contact and stopped spinning, the motor tab showed the throttle advancing on that motor - but that was probably just the FC showing a signal to that corner, right? So I probably killed the ESC?
 

silvrrubi592

New member
I've ordered new ESCs and a replacement FC board too. I see this thing breaking every time out for a while. Replacements are not the Emax 6A. Pretty sure this is a dumb question, but all 4 ESCs should be identical, right? Will still be 6 amp, just a different brand.
 

ElectriSean

Eternal Student
Mentor
I've ordered new ESCs and a replacement FC board too. I see this thing breaking every time out for a while. Replacements are not the Emax 6A. Pretty sure this is a dumb question, but all 4 ESCs should be identical, right? Will still be 6 amp, just a different brand.

As long as they have enough current capacity, brands, sizes etc don't matter. The only essential thing is that they are all able to run the same protocol. For example if you were running DSHOT600 but your replacement is only able to run DSHOT300, you will have to run them all on DSHOT300.