Thread Tools
Apr 06, 2019, 08:55 PM
Registered User
Thread OP
Discussion

Tarot 680 setup/preflight


I have a second hand Tarot 680 that has pretty much all the electronic hardware other than a camera gimbal. I have flown a handful of different RTF multicopters, as well as built and flown CP helis and several fixed wing aircraft. I'm not worried about the flying part.

But with all the different electronics this has, I am looking for suggestions on where to start on the setup, and preflight checks I need to do. Electronics are installed, albeit a bit messy IMO. I am in the process of documenting all the wiring, and will likely remove all of it to go over the chassis. Debating whether I should power it up first though and see what works and what doesn't. That is where I would like some advice.


Here is a list of the components it has:

FrSKy X8R
Pixhawk PX4 FC
DJI 3508 415 kV motors/DJI E600/E800 20A escs/DJI 1242 props
3DR GPS and compass? on folding mast
Minim OSD v1.1
FPV camera and VTx (unknown make)
switch and buzzer connected to FC
small board inline with main power lead that connects to "power" on FC



My plan is to gather whatever docs I can online for each of the components, and start as if building it myself. But any advice is welcome.


Brian
Sign up now
to remove ads between posts
Apr 06, 2019, 11:20 PM
Registered User
Thread OP
After making notes on the components and staring at it for a couple hours, I also have some basic mechanical issues I'm curios about.

On winged craft, I tend toward the obsessive on straight and true surface alignment. When the arms are folded, the motor ends are not all at the same height when viewed as if sitting on a table top, not so concerned about that. More concerning there is a minor tilt to the motor axis around the arm on some of them. The flight controller I'm sure can compensate for some amount of misalignment, but how much is too much? I haven't set up to measure it accurately, but lets say eyeballing less than 5 degrees.

Brian
Apr 07, 2019, 05:43 PM
FA3PR9WM79
Oldgazer's Avatar
Any amount it too much. Making the FC compensate for a crooked build wastes power, and every watt counts. Square the motors up with each other so the individual thrust lines are parallel.

To learn about Pixhawk and the ArduCopter firmware, start here: http://ardupilot.org/ardupilot/index.html
Apr 08, 2019, 07:06 AM
Registered User
don't worry about motor misalignment . you won't notice it .
Apr 08, 2019, 05:48 PM
FA3PR9WM79
Oldgazer's Avatar
Quote:
Originally Posted by swervo View Post
don't worry about motor misalignment . you won't notice it .
That statement is totally wrong.

The misalignment of an individual motor thrust vector induces a torque in the airframe. This torque is "felt" in the direction the motor shaft is pointing relative to the nose (forward end) of the aircraft.

For example, if the right rear motor is tilted inboard, the torque will be felt as a right yaw. If the right front motor is tilted inboard, the induced torque is felt as a left yaw.

Have you ever wondered why all of the motors don't spin in the same direction? Have you ever wondered why motors that are diagonally opposite of each other spin in opposite directions? This is done so the individual rotor torques will cancel each other out. Have you ever wondered how the flight controller makes an aircraft yaw? By changing motor speeds relative to motor spin direction and desired yaw direction..

Motor misalignment can also cause the aircraft to present unwanted roll and/or pitch behaviors.

Depending on the flight controller/firmware combination and log taking functionality motor misalignments can also be diagnosed by comparing motor PWM input values, Desired roll vs actual roll, Desired pitch vs actual pitch, and desired yaw vs actual yaw.

The bottom line is all of the motors must be aligned such that their thrust vectors are perpendicular to the aircraft frame.
Apr 08, 2019, 11:33 PM
Registered User
Thread OP
Quote:
Originally Posted by Oldgazer View Post
Any amount it too much. Making the FC compensate for a crooked build wastes power, and every watt counts. Square the motors up with each other so the individual thrust lines are parallel.

To learn about Pixhawk and the ArduCopter firmware, start here: http://ardupilot.org/ardupilot/index.html
OG,

I'm inclined to go with you're advice on motor alignment. Same reason I always set subtrim on servos and make linkages with correct geometry, on fixed wings or CPhelis, before setting the control surface trims. That will come when I get to reassembly.

I completely disassembled this thing to go over everything, and glad I did. The wiring was a bit messy, but in going over that, I found a few solder joints and/or connections that where poor and/ or failing, and likely would have led to bad things.

Biggest wiring issue right now is that the DJI escs have a coax lead for power supply(something I had not seen before). I've studied how to terminate those and have no issue with that. However, one of them was already cut to the absolute shortest length to get through the arm, and had intermediate connectors as well, before it terminated at the PDB of the Tarot. I can't bring myself to put it back together this way. There was 2 other ESC leads that the ground lead from the coax was very weak. One broke just in handling on disassembly, and the other is about to. So, at minimum all those will be clipped and re-terminated. But the one esc, I almost need a new esc as the case is not removeable to replace the power lead.

I have downloaded Mission Planner, and was able to connect to the Pixhawk and at least verify the version(PX4 v.2). I was hoping to be able to go through the configuration process but at one point it was obvious it needs to be installed in the vehicle and connected to all the peripherals before I can complete that.


I have some work ahead to get this flying.


Brian
Apr 08, 2019, 11:55 PM
Registered User
Thread OP
I've searched but only found a few threads on the coax power leads, mostly about how to terminate them. The terminating aspect I can handle. What I haven't found is a definitive answer whether it is needed in place of just standard leads. I understand the long DC leads on the copter arms can introduce problems with RF devices. Relative to my experience in automotive wiring, would twisted +/- leads for the ESC power accomplish the same thing if I decide to go with all new escs?

Brian
Apr 09, 2019, 09:21 AM
FA3PR9WM79
Oldgazer's Avatar
Quote:
Originally Posted by jpbldr View Post
I've searched but only found a few threads on the coax power leads, mostly about how to terminate them. The terminating aspect I can handle. What I haven't found is a definitive answer whether it is needed in place of just standard leads. I understand the long DC leads on the copter arms can introduce problems with RF devices. Relative to my experience in automotive wiring, would twisted +/- leads for the ESC power accomplish the same thing if I decide to go with all new escs?

Brian
Yep. That's why the US Navy uses twisted pairs for audio. Still, since you are running a Pixhawk, running CompasMot (Compass/Motor Calibration) will take care of magnetic interference. The only RFI I have encountered was from a STorM32 gimbal control. These are real "dirty" on the 900mHz band and they play hell with telemetry radios. Putting a ferrite ring on the input power lead and a 10uf 35v electrolytic cap across the power input pads took care of that.

Unless you are dead set on using the DJI ESCs, I highly recommend Hobbywing X-Rotor 40A OPTOs. I used them on my aerial photography rigs. They don't need any special attention. Just do the All At Once ESC calibration in Mission Planner and you're good to go.
Apr 10, 2019, 01:00 AM
Registered User
Thread OP
For now, I'm going to try to get one new matching DJI esc to deal with the too short power lead. Then work on terminating them

I'm not set on the DJI parts at all, or any of it for that matter. This just happened to be a complete package I got for a good price all in. So I guess what I mean is, I want to make this fly with the existing setup. Then I will keep what I like and replace what I don't as needed.

I just got a new FrSky X10 delivered today. Successfully bound to the Rx that was included with the Tarot, and another Rx for one of my sailplanes.

Brian
Apr 10, 2019, 01:06 AM
Registered User
Thread OP
Quote:
Originally Posted by Oldgazer View Post
I highly recommend Hobbywing X-Rotor 40A OPTOs. I used them on my aerial photography rigs. They don't need any special attention. Just do the All At Once ESC calibration in Mission Planner and you're good to go.
OG,

So, to clarify, do those ESCs just use standard +/- parallel leads alongside each other? And there is no interference with photo gear? How about interference with FPV gear?

Brian
Feb 24, 2020, 12:04 AM
Registered User
Thread OP
Finally getting back to this after getting sidetracked on other projects....

Resolved the esc/ and power distribution with one new esc to match the others, and an MRo power module. I've done the basic calibrations on the pixhawk, verified all the motors spin correctly, setup flight modes, etc.

I'm now working on an isolated deck for the pixhawk and radio components. I'm using some generic silicone dampers between the top chassis plate and my flight deck plate. This is a prototype part to finalize component positions. Final part will be cut from .060 G10 plate.

Brian
Feb 24, 2020, 08:46 AM
FA3PR9WM79
Oldgazer's Avatar
I use pre-fab Vibration Damper Mounts.

Smaller, lighter weight, and they will fit under a canopy...
Feb 24, 2020, 10:05 PM
Registered User
Thread OP
OldG,

The actual silicone isolators I'm using are similar to what is in your link. I wanted as many of the components attached to the isolated plate as possible; to increase its mass, and reduce the wiring that is subject to the movement between the main chassis and the isolated plate.

If this approach is outright wrong, I'm open to critique...

I'm fairly close at this point to doing some restrained bench testing with the props on, and then maiden flight. I will have the vibe analyzer running in pixhawk when I do so, and hopefully the results are acceptable.

Brian

To your specific points... while my isolated plate may be larger than what is in your link , the lower portion of what you linked is unnecessary, as the isolators directly engage the existing upper chassis plate of the 680Pro. I'm not concerned about a canopy at this point, though I think it could still be mounted directly, possibly requiring slightly longer standoffs.
Last edited by jpbldr; Feb 24, 2020 at 10:39 PM.
Feb 25, 2020, 12:26 AM
FA3PR9WM79
Oldgazer's Avatar
That larger isolation plate may just work.

I use canopies for 2 reasons. First, a canopy tends to shield the FC from air turbulence which helps mitigate vibration, and second, a canopy gives the aircraft a "finished" look.

To bench test the motors leave the props off and use the Motor Test function in Mission Planner > Setup > Optional Hardware. This test will let you verify motor order and motor direction of spin. NOTE: The motor sequence (A, B, C...) in Motor Test DOES NOT coincide with Motor Order (1, 2, 3....). Motor Test starts with front right (A), and moves clockwise as seen from above.

By default, Motor Test starts with a motor PWM of 5% which may or may not work. I have found that using 10% always works. The default run time is 2 seconds, but you can use any value.
Feb 29, 2020, 07:08 PM
Registered User
Thread OP
I had already done a motor test to verify they all worked and establish correct rotation. Did a bench test today with 2 of the 6 props installed, so that I could get a baseline calibration done for current sensing. That's done.

I briefly spooled up on my driveway with all props but it was getting wobbly near the point of liftoff. I'm guessing this will act like my FBL helis, and require a quick liftoff to go smoothly?? I don't want to do that in my driveway, so next test will have to be in a more suitable area.

As far as vibes... When monitoring in MP, I can see the graphs move when manually smacking it around, but when spooled up with just the 2 props on bench, nothing. Hopefully that is good. I need to get the logging setup, I suppose, to get actual data in real flight conditions.

Brian


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Help! My first multirotor setup, (Tarot 680 Pro) and silly questions! amdnikos Scratchbuilt Multirotor Drones 13 Jul 04, 2016 01:25 PM
Help! What battery setup for my Tarot 680 Pro? perlan44 Multirotor Drone Power Systems 3 Feb 02, 2016 01:05 PM
Help! What KV on a 4s Tarot 680 Pro setup? James7 Multirotor Drone Electronics 1 Sep 04, 2015 06:26 AM
Discussion Tarot 680 hybrid setup ( Need Advice) texasstormchaser Multirotor Drone Talk 1 Oct 08, 2014 06:04 PM
Discussion Setup help with a tarot 680 pro build quadcopterlover Multirotor Drone Talk 1 Jun 05, 2014 06:28 AM