VTOL -- freeman2100 vertical take-off and landing fixed wing +pixhawk

ArduPlane VTOL Plane
276

The only tilt-servos mentioned are the robot servos RDS3225 and RDS3115. These are 360 degrees-servos. Is that needed? Why robotservos? What manufacturer is it? Are they reliable?

Why dont use any standardservos with the similar power, double ballbearings and metalgears? Like Hitec HS-5585MH 20kg. Should it be metalcase and holes for the screws?

I see now that those RDS come with the motor-brackets, that is a plus…
RDS 3225: https://pt.aliexpress.com/item/32907625266.html
RDS 3115: https://www.amazon.co.uk/RDS3115-Degree-Mount-Digital-Servo/dp/B07V82RJ3H

What do you use?

277

No, I’ve been using the 180 degrees servos. I have found the labeling on the RDS3115 servo to be inaccurate so as long as they go about 110 degrees then it is fine.

They come with the arms to attach to the motors.

I use the RDS3115MG 15KG Large Torque 180 Degree Servo because they are inexpensive.

Cheers!

278

Thanks Greg!

Are there any other manufacturers of robotservos? I have used hundreds of servos in planes and helicopters (mostly Hitec, Futaba, MKS), but I am not used to robot servos, guess it is a jungle there too. :slight_smile: (just trying to learn)

You all have big batteries, would one 6S 10000 LiPo be enough? Just under 1,4 kg. But I guess an Amass XT90 with anti spark are too weak for the hover -> transition?

I am trying to find rear motors for the Freeman 2300. The total weight of our Freeman is around 6,1 kgs wit 6S 10000 LiPo (as is used in the Believer that flies very well) and an Sony A6000 camera. I have found two motors I think should work? Havent balanced the setup yet.

  1. T-Motor MN4014 KV400 4-8S 30A 171g with cables 900W
    50% 1530g (17x5,8")
    100% 3360g (17x5,8")
    https://www.3dxr.co.uk/product/t-motor-mn4014-kv400/

  2. KDE Direct KDE3510XF-475 KV475 4-6S 30A 120g 665W
    50% 1580g (15,5x5.3")
    100% 3520 g (15,5x5.3")
    https://www.kdedirect.com/collections/uas-multi-rotor-brushless-motors/products/kde3510xf-475|

279

@GregCovey

I am at the final stages of completing my VTOL build.

I was thinking why not use two batteries, and separate them, where one is used for the push motor and the other one for the quad motors rather use them in serial or parallel and have a common ground.

the Quad dedicated battery can also power things like long distance video feed, RFD9000 modems etc.

As the quad motors will be used much less, after take off, it will become sort of like a backup battery safety feature.

Just thinking out loud here.

What do you think?

I can design the electronics to switch/connect in parallel the quad batteries for push motors if needed for during a long flight via Tx switch. With FrySky Lipo monitor, we can monitor the batteries remotely.

280

Asim, in two of Vulcan multirotors I have built we use a Mauch Hub to monitor the two LiPo’s that are in in parallel (two Mauch sensors). If one LiPo fails, the other still works. Alarm is sent to the RC-transmitter. This setup dont carry around a deadweight (second LiPo) but have double the capacity. Splitting the consumption (lower current draw per battery) and you dont have to use big connectors or fat cables.

If you have bad signal on the videofeed if you use the same battery as any/all of the motors, use filters or a different battery (2-3S small one will do).

281

MickeM Hi,

I am using Mauch for my Hexacopter.

Which of these configs did you use?

The only thing I don’t like about Mauch is no individual Lipo cell monitoring. Over all Lipo Volt/Current monitoring, including fail safe is just not a viable solution during long flights, unless backed by Yaapu.

I suggested to “Christian” last year to consider adding such support.

I love how Yaapu telemetry works along with FrSky Lipo Cell monitor(s) connected to S-PORT. Its a kick ass combo solution where I can keep an eye on every single pack as well as cells within each Lipo.

Twice, one of the cells in a 6S solution dropped below safe thresholds. Besides your UAV may crash, you end up loosing money on expensive 6S + batteries…no good to me.

p.s. I use like DJI 7km (Occusync, new FPV video) feeds and we are developing our own HD video system. Long way to go.

1 Like
282

Yes, many different brands, quality, and strength. Try a search on Banggood, FoxtechFPV, or Amazon and you will find some.

That is what I am using. There is plenty of room for more battery but I already had them from the Nimbus VTOL. Remember my posting above that my Freeman only draws 40amps for hover. As more people test their new builds, we will see if this is common. Certainly, an XT90 connector will work fine.

This is how Foxtech does it. For my hobby use, I don’t require it.
Foxtech PG20 V1.1 Dual Battery Swapping Board

Cheers!

283

Hi,

I’m having an uncommon behavior with the motors on my freeman2100. Before upgrading ( I was using 3.9 and was working fine), after upgrading to 4.1 in manual mode just one motor are working/spinning as in this video: https://youtu.be/P9zjcS-c47Q

Any help ?
best regards

284

Hi Vinicius,

I am using version 3.9.11, not 4.0 or 4.1. Just a guess here…perhaps your parameters did not remain static. Can you load an earlier version of your .param file from when the 3.9 firmware was installed?

Lastly, try posting your issue in the Plane 4.0 forum to see if anyone else is using it or create a new thread here on the VTOL plane forum so the appropriate title shows up on the main Ardupilot page.

Good luck!

285

The problem was a mistake in the Q_TILT_MASK, I was using 4 instead 5.

286

Did this change during the firmware upgrade?

I believe the parameter is called Q_TILT_MASK, not Q_TILT_BITMASK. My concern with the question above is if the firmware upgrade caused the issue.

Cheers!

287

Hi @GregCovey,

I update the parameter name.
The configuration wasn’t changed in the update, the value of 4 was made by myself.

288

I have an another question.

I was reading about qautotune, were the the pitch, roll and yaw axis will be tuned.
This means the auto tune for plane unnecessary or we need to first to qautotune and in flight do the plane autotune ? Or the qautotune will adjust the pitch and roll for motors and ailerons/elevator/elevons ?

289

QAUTOTUNE is for the quad motors only. AUTOTUNE is for forward flight servos for pitch, roll, and yaw.

I would stress caution on using any AUTOTUNE feature unless you have previously flown the QuadPlane and feel that the default parameters are not sufficient. Even then, I do a small tweak to the PID values manually and check for improvement. Typically, you only need to adjust the P value slightly for each axis.

I will ask Mr. Chang for the Freeman PIDs and post them here. I’m curious to compare them against the Nimbus 1800 PIDs.

Cheers!

290

Anyone? Have I got it right when selecting one of these for the rear motors for a Freeman 2300 at 6,1 kg take off weight?

  1. T-Motor MN4014 KV400 4-8S 30A 171g with cables 900W
    50% 1530g (17x5,8")
    100% 3360g (17x5,8")
    https://www.3dxr.co.uk/product/t-motor-mn4014-kv400/

or

  1. KDE Direct KDE3510XF-475 KV475 4-6S 30A 120g 665W
    50% 1580g (15,5x5.3")
    100% 3520 g (15,5x5.3")
    https://www.kdedirect.com/collections/uas-multi-rotor-brushless-motors/products/kde3510xf-475

And what is a suitable KV for the front motors in a 6,1 kg Freeman 2300 in a 6S-setup? Around 500? What safe thrust is enough for throwing from hands? 1,0?

Asim, I have used Pixhawk 2.1 + Mauch Sensor Hub X2 V2 , Mauch PL BEC + backup-BEC, 2x PL-200A Sensors. Measuring from BEC from PDB where both batteries are connected. But I dont get how they have thought about which one to use between the two sensor-values that came on a little piece of paper for each sensor.

291

Not sure if this is off topic.

But has anyone looked into this plane.

image

Baby Shark VTOL RTF.pdf (440.3 KB)

this company has some really interesting VTOL’s small to fairly large

292

Not clear on your question. Are you asking how the system decides between the two sensors?

Which setup are you using?

this is my setup using V3 not V2.

image

293

Mr. Chang sent me the parameter files to be used as a guideline or reference. I posted them below.

I have gone through the parameter file differences with my own and did not find many changes. I believe the AHRS_EKF_TYPE change comes from using different Pixhawk hardware. For forward flight, they are using higher P values so I may change to those. Their transition speeds are slower but aided by custom firmware that I prefer not to post until I test it, if needed. The modified firmware improves transition stability.

Parameter File Differences

Greg’s Parameter then (MFE’s Parameter)

AHRS_EKF_TYPE,3 (2)

ALT_HOLD_RTL,10000 (-1)

BRD_TYPE,2 (No Entry for MFE)

PTCH2SRV_D,0.04 (0.1)
PTCH2SRV_FF,0.4 (0)
PTCH2SRV_I,0.15 (0.2)
PTCH2SRV_IMAX,3000 (3000)
PTCH2SRV_P,0.7 (1.5)
PTCH2SRV_RLL,1.15 (1.3)
PTCH2SRV_RMAX_DN,0 (60)
PTCH2SRV_RMAX_UP,0 (60)
PTCH2SRV_TCONST,0.5 (0.5)

Q_A_ACCEL_P_MAX,110000 (30000)
Q_A_ACCEL_R_MAX,110000 (30000)
Q_A_ACCEL_Y_MAX,20000 (27000)

Q_A_ANG_PIT_P,6 (6)
Q_A_ANG_RLL_P,4.5 (8)
Q_A_ANG_YAW_P,3 (3)

Q_A_RAT_PIT_I,0.35 (0.2)
Q_A_RAT_RLL_D,0.004 (0.012
Q_A_RAT_RLL_I,0.4 (0.25)
Q_A_RAT_RLL_P,0.4 (0.3)
Q_A_RAT_YAW_D,0.07 (0)
Q_A_RAT_YAW_I,0.08 (0.18)
Q_A_RAT_YAW_P,0.7 (0.3)
Q_A_SLEW_YAW,3000 (6000)

Q_ANGLE_MAX,4000 (1500)
Q_ASSIST_ANGLE,30 (45)
Q_ASSIST_SPEED,10 (14)
Q_ASSIST_ALT, n/a (20)
Q_ASSIST_ALT_MOD, n/a (18)

Q_LAND_SPEED,50 (40)

Q_TILT_MAX,65 (32)
Q_TILT_RATE_DN,50 (12)
Q_TILT_RATE_UP,200 (45)
Q_TILT_YAW_ANGLE,13 (10)

RLL2SRV_P,1 (3)

YAW2SRV_INT,0.3 (0.15)
YAW2SRV_RLL,1 (0.8)

MFE Freeman 2100.param (19.5 KB)
MFE Freeman 2300 .param (19.5 KB)

2 Likes
294

You need to enable the BATT2_MONITOR parameter and a second set of battery parameters will become available.

Yes, that sounds correct. Use the 2300 Configuration List to reference the motor Kv needed. Max take-off weight is 8kg.
Freeman 2300 Configuration List.pdf (74.9 KB)

Yes, those designs are very nice but they are expensive and require assembly/disassemby at the field. Do a search for Mugin VTOL. The “Shark” label is for FoxTechFPV.

295

Asim: Im using the same setup as in your attached picture, but not the PowerCube. I am using a Mauch PL 2-6S / 1x BEC to power the Cube and the Hub. What I meant and what I forgot to mention was that when setting up the power monitoring in Mission Planner (BEC on Power1 and BackUp-BEC on Power2 and third backup on rails) you can use only one of the small papers with values that came with the PL-200A-sensors, which one should I use? They dont have the same values. Both sensors are connected to the Hub and this one is powered by the BEC that is getting the current/voltage-sensorsignals from the Hub and bridgeing them/one of them (?!) through to the Power1 on the Cube. Are you going to use Mauch Dual battery redundancy with the Sensor Hub in the Freeman? I thought of this idea but may skip it and run on only one battery.

GregCovey: BATT2_MONITOR monitors the Cube’s Power2-port, right? There I got the Mauch BackUp-BEC which only have ground and power, no sensor-signals for current or voltage. But I guess the sensor-signals from the other PL-200A-sensor can be connected to the Power2-port sensor-signals inputs instead, and use BATT2_MONITOR using the other piece of paper with values (whatever this values are called, Amperes per volt and Voltage Divider if I remembered it right).

Max takeoff-weight for the Freeman is 8 kg but it will never get that heavy for us, thats why I am trying to find another motor-setup that works well with the aprroximately 6,1kg weight which is a lot less than 8 kg. My goal is to have rear motors that hovers at just under 50% PWM with 17-18" propellers and front motors that are suitable for the 6,1kg weight and still power effective and can transition easily and safely. Thats why I was wondering what power-to-weight-ratio is needed. 1:1-motors in the front would make the Freeman hover nose up, but should it be more powerful or less? I dont like handlaunching weakly powered planes, thats why I dont want the Freeman to transition with a huge dip. :slight_smile: But I dont like putting on over-powered setups either, they are heavier and/or draws more current. The Believer we use with KDE Direct-motors have a power-to-weight-ratio at litte above 1:1 (more power than weight). But I dont know how much, the propellers we use are not in any performance datasheet ( https://cdn.shopify.com/s/files/1/0496/8205/files/KDE_Direct_XF_CF_Brushless_Performance_Testing_-_KDE2814XF-515.pdf?7734511287488513374 ) and we use APC 11x5.5". But they should be almost the same in thrust as the mentioned 12.5x4.3", maybe a little more powerful. Believer weight is 4488 and motors with 12.5x4.3" reaches max 4720 gram. And I dont find any performance datasheet for the MFE-setup for the Freeman either.