Sunday, October 2, 2016

QX90 Quad-Drone Enhancements

Updated October 10, 2016

I was a little hard on my initial QX90 review, REVIEW: Eachine QX90 Mini-Drone.  Before I dive into the details the QX90 will fly out of the box but it flies even better with a few enhancements.  If you fly racing quads like me you have a good understanding of electronics and how to use a soldering iron.  These improvements will be easy for you.  This post covers the parts I choose, where to purchase them, and my impressions of the flight characteristics.

After the review I began to think about what it would take to get the Eachine QX90 micro quad to fly better.  Here are some parts I purchased frm RC-Drones.com.  These guys ship really fast and even put a personalized, "Thank You!", note in the box.   Who does that?  Nice touch.  The following are the parts I purchased.

CL-0820-17-11T Gear CL-0820-17-11T Dark Edition Coreless Micro Motor Set(4)8.5x20mm  $34x1 = $34

PROP-HUB3BLADE Hubsan Style Black/White 3-Blade Micro Props For The DP03 Brushless Motors (8 Props Total) $6.00 x 4 = $24

3.7v 750mAh 25C Eachine LiPo  (purchased separately from Bangood)

I wanted to get new flight camera but I noticed after I ordered the micro camera was made for 5v systems, not 4.2v-3.3v (or 1S).  The camera specs are definitely close and if I try it I will post the results.  Taking off the old motors and soldering on the new motors was a snap.  I didn't want to take the drone apart and solder the new motors to the board.  Instead, I went for the easy approach, on the stock motors I cut the leads about 1cm from the base of the motor and tinned the wires.  Next, I cut the long wires on the dark edition motors about 2cm from the base.  I wanted to give myself some extra room if I happen to cut through the leads when I trim the insulation off.  The new dark edition motors are the same size as the stock with the exception that the shafts are shorter.  Still plenty of room to the 3-bladed props on.

When I was flying stock motors I switched over to 380mAh cells.  These are the same size cells flown on the similarly sized Hubsan X4.  After you mount the new motors and props the first thing you will notice is that the aircraft is sluggish and does not perform.  The 380mAh cell does not have enough power.  I switched over to 750mAh cell.  The video Pimped QX90 is a video I made flying the dark edition motors, 750mAh cell, and 3-bladed props.


I don't recommend flying this aircraft in too much wind but it's loads of fun.  Mine still needs some tuning but it's definitely running well enough to fly and have some fun.  I don't believe this receiver has telemetry.  I can set a timer on my transmitter but I would feel better with a proper low-battery sensor.  The crash at the end of the video was due to a blade breaking off on one of my 3-bladed props which was weakened from a previous crash.

Crazy cool!  Amazing aircraft and loads of fun.

October 7, 2016, quick update.  I'm noticing the new motors get fairly hot between batteries.  I tried going back to 2-bladed props.  Seems to fly well on the 750mAh battery.  A made another video of some flights on the QX90.  Flies much better.  This aircraft is not like flying your 180mm/250mm quad racing drone but it's still great nevertheless.  Still it takes a few modifications and a far amount of tuning.  Spoiler alert, I almost got my QX90 run over by a car, lol.



October 16, 2016, I have stabilized on the stock 2-bladed props with the dark edition upgraded motors and upgraded 750mAh battery.  This provides a flight time of around 3 minutes.  There is no telemetry with this flight controller so I set a timer to count down from 2.5 minutes when the throttle is engaged.  This gives me audible warnings so I don't damage the batteries or loose the aircraft.

Photo: Betaflight 3.0.0 PIDS and gains(click to enlarge)

Following is a video I made with the new pids.  It feels very locked in to me.  Almost like flying a 250mm quad but not quite as much power. 



A big benefit of these micro-quads is that they can be flown almost anywhere.  The modified aircraft comes in just under 40g without the battery.  So although the aircraft flies fast (40-50mph) danger to people and property is very low.

Monday, September 26, 2016

REVIEW: Eachine QX90 Mini-Drone

This is my review of the Eachine QX90 micro-quadcopter.  The point that attracted me to this particular quad initially was that the receiver is directly compatible with the Frsky receivers and 5.8Ghz FPV Goggles like Fatshark.  The idea is that if you have this equipment and race drones already you can purchase one of these QX90's, bind, and your ready to fly.  I will start out by describing features I don't like or where the product falls short of my expectations and I will end on a positive note.

Box Crushed
The drone is shipped in a foam cutout the shape of the aircraft.  The block of foam is pretty tough to withstand bumps during shipment.  The plastic wrapping around the case was not durable at all and falling off by the time I received my drone.  Luckily the drone was fine and protected by the foam.  I'm not really into unboxings.  I don't care about the box something ships in so long as I receive the contents undamaged.

Poor Quality Flight Video
Like many inexpensive cameras the one on the QX90 does not handle significant changes in contrast well.  For example, fly into the sun and all the darker parts of the picture turn black.  As long as light levels within the field of view remain constant, which is almost never the case outdoors, the video is acceptable.

Motors Under-Powered
Hover is at 50% - 75% throttle depending upon how fresh the batteries are.

Batteries Too Heavy
My drone shipped with 680mah batteries which is too big for the aircraft.  The aircraft will fly with them but it's lumbering and not very agile.


Difficult to Tune
I have some difficulty tuning at first.

Now let's switch over the postive.  There's a lot of great things to like about this aircraft.  These are the product features I like the best.

$60 price point
Need I say more?  $60 is a very low price point.  The fact is you can pay more for a single component on a 250mm class racing drone than this entire drone.  If your on a budget, this aircraft is a serious consideration.  However, if your on a budget why would you have an Frsky transmitter and 5.8Ghz goggles?  I like the price point but it's a marketing play to experienced pilots with a fatter checkbook.  While I appreciate the low price point I will gladly pay double or triple the price for a high quality 90mm drone that fulfills all my exceptions.  May I'm too picky.

SP Racing F3 EVO (Brushed) FC
I really like that you get an F3 you can flash with Cleanflight.  In fact, you can also flash this flight controller with Betaflight as well with good results.  Using familiar components and software is helpful for tuning.  Although I noticed that Betaflight was really difficult to tune.  To get the aircraft stable requires far higher Integral and Derivative settings than I usually see on any my 250 quads.  I'm not sure if this is extra error is introduced by the motors, poor quality board, or perhaps it's an artifact of a short 90mm frame size.

Frsky Receiver
Super cool.  The Frsky compatible receiver is also an SBUS version.  SBUS is a little snapper than CPPM.

Voltage Levels Unknown
I really have no idea when I'm flying the QX90 what the voltage levels are and when it's time to land.  There is no buzzer on the aircraft.  I didn't notice a second antenna on the receiver so I don't believe there is any telemetry.  Probably the best to know when it's time to land, once you get your aircraft configured the way you like it, is to set a timer on your transmitter. Set an audible alert after 3-4 minutes of flight time or whatever is appropriate for your aircraft.

Ok, I have covered some of the features I don't like and some I like.  Now I'm going to tell you what I'm doing to improve my QX90.

3-Bladed Props
The aircraft ships with 2-bladed props and it's quite fast but a little difficult to control.  First I'm ditching the 2-bladed props and moving to 3-blades.  I don't care about the top end speed.  This drone has plenty of speed.  I'm looking for more precision flying in tight spaces.  On my 250mm acro drone I fly with 6-blades and it fly's like a Ninja, very locked in feeling and precise.

New Camera
Next, I'm ditching the stock camera and replacing it with a better after market integrated camera and transmitter.

Motors
I'm ditching the stock motors and going with a set of 3.3v 17k rpm high output motors.  It could be this is not necessary and upgrading the props will improve flying.

Battery
I'm thinking the 680mah batteries are too big.  I had some extra 380mha batteries around that I used for a Hubsan X4.  When I tried these smaller batteries it seemed to fly better.  I'm not firm on this for now.  I still need to do more testing.

Purchasing the extra gear blows the $60 manufacturer price point.  From my perspective I'm more interested in tuning this platform to make it fly more like my larger drones than saving a few dollar.  I like the idea of flying inside the house, super cool.  However, I really want something super precise I can fly easily down a narrow hallway or fly super slow and stable.  After I get my new gear I will post an update to let everyone know if this fly's better.

Friday, August 19, 2016

Impressions of Furious FPV True-D Diversity Receiver

One word, awesome!  You can grab yours on the companies web site, furiousfpv.com.  The gear in this hobby is changing so fast it's difficult to keep pace.  Not long ago I wrote an article describing a diversity receiver build out I did using a small camera bag, Portable FPV Microwave Ground Station in a Backpack.  Now the Furious FPV module that fits into my Fatshark goggles eliminates the need for a backpack full of electronics.  I'm still carrying the backpack but now I can haul more batteries, water, etc.

I like having diversity with the lightweight benefits of google flying.  The quality is similar to my ImmersionRC Duo 5800 V4 without all the extra bulk.  I don't have any of the diversity video to share just yet just sharing a few initial impressions.

REVIEW: SJCAM S5000+ Sports Action Camera

This is going to be a no nonsense review of the SJCAM S5000+ for suitability and use for photography on your multirotor aircraft.  After my S5000+ arrived I strapped it my Vortex 250 Pro and went to the field to try it out.  Here are my initial impressions from my flight.

Camera Size
The camera is a ever so slightly larger than it's earlier predecessors.  While it is a small camera, especially without it's protective case, it must be small and lightweight for use on multirotor aircraft.  There's not that much room on a 250mm aircraft.  Once I have the camera strapped into the optional GoPro cradle there is about 1/4" between the front rotor blades and the camera.  During a hard landing the camera shifted slightly.  After repositioning the aircraft for another flight I neglected to check the camera position.  One of my blades was trapped against the camera and would not turn.  I'm lucky I didn't burn my motor out.  It's not a big problem now that I know about it but another line item for me to check on my pre-flights.

Another concern is that there is only I place to put a wrap strap to hold the camera the mount and sometimes it presses the button while in flight turning off the camera.  I missed recording several flights that day due to the camera being shutoff.  At first I thought it was a configuration feature to shutoff the camera to save battery power.  After some careful review of the options this turned out not to the case.

Finally, I had the battery door on the bottom of the camera fall off on several flights.  In all fairness, flight is a demanding environment and it's easy to pull a lot of G's in an aircraft, corning, loops, crashing, etc.  I never lost the door but it popped off and I had to put it back on.  No doubt SJCAM could make some improvements designing cameras specifically for mutirotor rotor but I think ImmersionRC could also improve it's cradle design.  The design of the camera cradle on the vortex is two vertical carbon fiber plates that fit into the top of the aircraft.  The vertical plates are stabilized by a carbon backing plate.  A piece of neoprene with adhesive on the back keeps the camera from sliding.  A slightly better design in my opinion would also include a small carbon plate across the bottom with another piece of neoprene.  The extra neoprene would keep the camera from sliding and door would not pop off the bottom.

Image Quality
Amazing!  I took a quick video while setting some PIDs on my Vortex 250 Pro.  The camera will do 4k.  However, I am mostly interested in 1080P.  I set the camera to record 1080P at 30fps which it can do very easily.  No lagging or ripping, jello lines, etc.  I'm really REALLY happy with the quality of the video.



Better Auto White Balance
Almost forgot, previously one of the problems with small cameras is that when flying into the sun many cameras will darken the entire picture to compensate for the bright sun.  The result of this balancing war is that the sun looks great while everything else in the picture is so dark it cannot be seen.  Of course this is not very practical for sharing video with your friends and family.  You don't need to worry about this with the 5000+.  The camera adjusts to the light perfectly without adjustment.  When flying into the sun the camera reduces intensity of the sun while keeping foreground and background the same.  Watching my old camera raise and lower the intensity of the screen is annoying.  This feature alone is worth the price of a new camera.

Price
The camera goes for about $139US.  Price is outstanding.  There is some competition in the space.  I can't say how well these compare.  GoPro is definitely the gold standard for sports cameras but cameras like SJCAM are edging in.  At least with 1080P, the difference in quality between the 5000+ and the GoPro Hero 4 Black is marginal - my opinion.  Especially, if your a new pilot and crash a lot.  ;o)

Enjoy!

Monday, August 15, 2016

Patching ImmersionRC and Starting PIDS for Success

I wanted to follow-up with this post for some of you that may be considering or in the process of patching software on the ImmersionRC Vortex 250 Pro.  Please, stop and read FIRST.

ImmersionRC has made a very admirable attempt to make flying racing drones much easier.  And believe me, I have built two racing drones before purchasing my Vortex and it's an amazing piece of gear.  So don't wring your fists at ImmersionRC just yet.  It's likely if your having problems during patching there there's probably nothing wrong with your Vortex hardware.  I think where ImmersionRC falls a little short is in their documentation and even they have admitted as much.  Let's get on with it.

To be successful there are a few pieces of information need to learn that are not covered in the manual.  It may be the type of stuff that's obvious to expert multi-rotor builders but it's worth covering for everyone.

The Problem
Like me, you may have built a few drones on your own.  Because the Vortex has many of the components in common with DIY drones like flight control software and hardware you may be inclined to patch the Vortex like you patch your other DIY aircraft - but don't.  I have spent the last 2-days frustratingly trying to patch my Vortex like other aircraft I have that use Cleanflight and Betaflight.  In my case, I would patch the flight controller with Betaflight in the usual way.  Next, I would fire up the Betaflight Configurator desktop tool, configure my transmitter, my PIDS, etc.  Each time I changed some options on the configurator tool I would switch screens in the app to another screen and then return to the screen I made my changes.  Usually this is a good test to see if the values have been saved in the flight controller.  Each time you change a value, move to another screen, and return to the screen where you changed the settings it forces the configurator to load the settings from the flight controller again.  If the values are the same as you changed them then they have been "burned" into the flight controller.  In my case on the Vortex, I would apply some changes, check them in the way I describe.  To be extra sure my values where saved I power cycled the aircraft (unplug/replug battery).  When I returned to the configurator and open the last settings screen where I made changes I discovered the original values where stored not the values I set.  Essentially, the aircraft would forget all my settings if powered off.  I was not sure what to do.  I thought the problem was bad versions of flight control software or differences in the configurator vs. flight controller version.  None of this was a problem, it turns out the Vortex is trying to make setup easy for us and in the process of making things easy during setup if you attempt to configure flight control software as usual you will be frustrated.  I realize I have probably covered the problem in a little too much mind numbing detail but I wanted to put some bread crumbs out on the Internet in the event others are suffering from similar problems.

The Procedure
Keep in mind, this works well around the time of this article but ImmersionRC could improve the hardware in the future.  To restore all software on my Vortex 250 Pro to a known working state I did the following.

1)  Install ImmersionRCTools on your desktop
2)  Install the latest version Baseflight Configurator.  I have only used Baseflight.  I hear some on the forums using Cleanflight but I have not used it on the Vortex myself.
3) Install Baseflight/IRCFUSION3 flight control software on the Vortex.  Don't bother configuring it yet.
4) Install OSD on the Vortex using ImmersionRCTools
5) Use your FPV goggles to configure your Vortex with your transmitter.
6) Power cycle the aircraft
7) Now you can enter the Betaflight Configurator to set, flight modes, PIDS, etc.  Keep in mind, not all settings can be changed since they are configured by ImmersionRC's OSD.

Known working software versions include
Vortex 250 Pro (Ummagawd edition)
Bootloader 1.0.0.3
Firmware 1.0.0.29*
Betaflight/IRCFUSION3 2.7.1
Betaflight Configurator (latest version)
*Version 29 is not available publicly on ImmersionRC's website - to my knowledge.

Vortex 250 Pro
Bootloader 1.0.0.3
Firmware 1.0.0.28
Betaflight/IRCFUSION3 2.8.1
Betaflight Configurator (latest version)

Starting PIDS
I'm not sure why good Vortex 250 PIDs are so damn hard to find.  Ordinarily sharing PIDs does not make too much sense for DIY racing drones.  Everyone's builds are just different enough so someone else's PIDs will probably not be too helpful anyone but that's not the case with Vortex.  We all have the same aircraft with some minor differences.  The weights of antenna's, differences in batteries, etc.  However, if your going with ImmersionRC's recommended setup of a 1300mAh battery then these PIDs will be a good starting point.  Please, please, please, write down any settings before you change them so you can easily roll back.

       P    I    D    RC Rate   Rate   Expo
Role   31   21   16   1.0       .47    .75
Pitch  36   23   16             .47    .75
Yaw    91   45                  .47    .75

I developed these over time and it seems reasonable it will work for others.  My son has the Ummgawd edition and it was a good starting point for him so it works for both 5x4x3 and 5x4x4 props.  You may need to touch up the settings slightly for flying 2-bladed props so iron out the vibration but it should fly ok.  I do fly with a little more expo on Taranis transmitter so if the aircraft if a bit sluggish you may want to bump up the expo to your liking.  Enjoy!

Tuesday, June 21, 2016

New FAA Part 107 Rules

Original FAA Post, Summary of Small Unmanned Aircraft Rule (Part 107)

Operational Limitations 

  • Unmanned aircraft must weigh less than 55 lbs. (25 kg).
  • Visual line-of-sight (VLOS) only; the unmanned aircraft must remain within VLOS of the remote pilot in command and the person manipulating the flight controls of the small UAS [comment#1]
  • Alternatively, the unmanned aircraft must remain within VLOS of the visual observer.
  • At all times the small unmanned aircraft must remain close enough to the remote pilot in command and the person manipulating the flight controls of the small UAS for those people to be capable of seeing the aircraft with vision unaided by any device other than corrective lenses.
  • Small unmanned aircraft may not operate over any persons not directly participating in the operation, not under a covered structure, and not inside a covered stationary vehicle [comment#2].
  • Daylight-only operations, or civil twilight (30 minutes before official sunrise to 30 minutes after official sunset, local time) with appropriate anti-collision lighting.
  • Must yield right of way to other aircraft.
  • May use visual observer (VO) but not required.
  • First-person view camera cannot satisfy “see-and-avoid” requirement but can be used as long as requirement is satisfied in other ways.
  • Maximum groundspeed of 100 mph (87 knots). [comment#3]
  • Maximum altitude of 400 feet above ground level (AGL) or, if higher than 400 feet AGL, remain within 400 feet of a structure.
  • Minimum weather visibility of 3 miles from control station.
  • Operations in Class B, C, D and E airspace are allowed with the required ATC permission.
  • Operations in Class G airspace are allowed without ATC permission.
  • No person may act as a remote pilot in command or VO for more than one unmanned aircraft operation at one time.
  • No operations from a moving aircraft.
  • No operations from a moving vehicle unless the operation is over a sparsely populated area.
  • No careless or reckless operations.
  • No carriage of hazardous materials.
  • Requires preflight inspection by the remote pilot in command.
  • A person may not operate a small unmanned aircraft if he or she knows or has reason to know of any physical or mental condition that would interfere with the safe operation of a small UAS.
  • Foreign-registered small unmanned aircraft are allowed to operate under part 107 if they satisfy the requirements of part 375.
  • External load operations are allowed if the object being carried by the unmanned aircraft is securely attached and does not adversely affect the flight characteristics or controllability of the aircraft.
  • Transportation of property for compensation or hire allowed provided that
    • The aircraft, including its attached systems, payload and cargo weigh less than 55 pounds total;
    • The flight is conducted within visual line of sight and not from a moving vehicle or aircraft; and
    • The flight occurs wholly within the bounds of a State and does not involve transport between (1) Hawaii and another place in Hawaii through airspace outside Hawaii; (2) the District of Columbia and another place in the District of Columbia; or (3) a territory or possession of the United States and another place in the same territory or possession.
  • Most of the restrictions discussed above are waivable if the applicant demonstrates that his or her operation can safely be conducted under the terms of a certificate of waiver.
Remote Pilot in Command Certification and Responsibilities 
  • Establishes a remote pilot in command position.
  • A person operating a small UAS must either hold a remote pilot airman certificate with a small UAS rating or be under the direct supervision of a person who does hold a remote pilot certificate (remote pilot in command).
  • To qualify for a remote pilot certificate, a person must:
    • Demonstrate aeronautical knowledge by either:
      • Passing an initial aeronautical knowledge test at an FAA-approved knowledge testing center; or
      • Hold a part 61 pilot certificate other than student pilot, complete a flight review within the previous 24 months, and complete a small UAS online training course provided by the FAA.
      • Be vetted by the Transportation Security Administration.
      • Be at least 16 years old.
  • Part 61 pilot certificate holders may obtain a temporary remote pilot certificate immediately upon submission of their application for a permanent certificate. Other applicants will obtain a temporary remote pilot certificate upon successful completion of TSA security vetting. The FAA anticipates that it will be able to issue a temporary remote pilot certificate within 10 business days after receiving a completed remote pilot certificate application.
  • Until international standards are developed, foreign- certificated UAS pilots will be required to obtain an FAA issued remote pilot certificate with a small UAS rating. A remote pilot in command must:
    •  Make available to the FAA, upon request, the small UAS for inspection or testing, and any associated documents/records required to be kept under the rule.
    • Report to the FAA within 10 days of any operation that results in at least serious injury, loss of consciousness, or property damage of at least $500.
    • Conduct a preflight inspection, to include specific aircraft and control station systems checks, to ensure the small UAS is in a condition for safe operation.
    • Ensure that the small unmanned aircraft complies with the existing registration requirements specified in § 91.203(a)(2).
  • A remote pilot in command may deviate from the requirements of this rule in response to an in-flight emergency.
Aircraft Requirements
  • FAA airworthiness certification is not required. However, the remote pilot in command must conduct a preflight check of the small UAS to ensure that it is in a condition for safe operation.
Model Aircraft
  • Part 107 does not apply to model aircraft that satisfy all of the criteria specified in section 336 of Public Law 112-95.
  • The rule codifies the FAA’s enforcement authority in part 101 by prohibiting model aircraft operators from endangering the safety of the NAS.
Comment #1, 2 roles.  Remote Pilot and Pilot.  Pilot can be wearing FPV gear
Comment #2, "not under a covered structure", first time FAA has controlled the airspace inside a structure
Comment #3, no restriction on dive speed

Monday, March 28, 2016

S.1608 Consumer Drone Safety Act

S.1608 Consumer Drone Safety Act.  US bill proposed bill by Sen Feinstein (CA-D)

Interestingly the bill does not seem to cover racing drones built from scratch by individual drone enthusiast.  Consider the following section of the bill...

(b) Applicability.—The final rule required by subsection (a) shall apply to consumer drones that—

We see from the bill applicability is in only relation to "consumer drones".  When we dig further to understand what consumer drones are at the end of the bill it says this...

(4) CONSUMER DRONE.—
(A) IN GENERAL.—The term “consumer drone” means a civil unmanned aircraft or a civil unmanned aircraft system that—
(i) is—
(I) equipped with an automatic stabilization system; or
(II) capable of providing a video signal allowing operation beyond the visual line of sight of the operator;
(ii) is manufactured and intended for commercial distribution; and
(iii) weighs 55 pounds or less or is certified as described in section 336(a)(3) of the FAA Modernization and Reform Act of 2012 (Public Law 112–9549 U.S.C. 40101 note).
(B) INCLUSION.—The term “consumer drone” includes a kit of component parts that would be a consumer drone if the kit were assembled prior to commercial distribution.
If we look at 4(B) we see the term "consumer drone" includes kits of components parts that can be assembled into a drone.  The bill includes commercial drones like DJI phantom also racing drones like the Vortex.  It also includes Ready To Fly(RTF) kits.  I'm not a lawyer but my interpretation is that the bill does not include, custom non-RTF DIY drones hobbiest build themselves from parts gathered from individual suppliers.  The reason is that the parts are sold individually from suppliers and individual parts do not comprise a kit.  An example, is a racing drone like my ZMR250 quad.

If your a lawyer or read this proposed bill different please comment and share your thoughts.

Tuesday, March 8, 2016

REVIEW: ImmersionRC Vortex 250 Pro

Updated on March 9, 2016

I recently purchased an ImmersionRC Vortex 250 Pro and thought I would provide some thoughts for anyone considering a purchase.  Five words, you will not be disappointed.

Love it!
- Best looking racing drone around!  Great finished look.  Less like someone's research project.
- Arms sit higher than most racing drones for a lower center of gravity.  Ancillary benefit, you'll be cutting less grass.
- OSD includes integrated current sensing.  Tough to add on your own for <250mm multi-rotors.  Sensors tend to be a little bulky for small platform aircraft.
- Includes S.BUS support, very cool.
- Extra arm space around the motors protects motors in crash.
- Support to run non-stock motors.  Even the pro-PIDs include configuration for motors other non-stock motors.  I have been wondering what a set of T-Motors new F-40's would be like on the Vortex.
- Set configuration via transmitter and OSD.  No need to bring laptop to airfield or add Bluetooth LE support.  The menu system is intelligently designed.  No complicated movements and easy to learn.
- Vortex supports/anticipates customization.  Flight controller is compatible with Cleanflight, support for non-stock motors, etc.
- Comes shipped with pro-pilot PIDS.  Like to see more PIDs for the stock motors.
- Pimp your Vortex.  After market kits are available to change the aircraft color to green, white, etc.

Needs improvement
- Cleanflight customized for hardware.  No way to set acc_lpf_factor and some of the other settings look different.  Advantage, easy to set pro-pilot PIDs, set PIDs and other features via transmitter, and more.  Disadvantage, not sure what I loose at the moment if I install Cleanflight from Cleanflight project.  Still investigating.
- CMOS flight camera could be better quality picture.  Virtually no adjustments available.  Upgrading mine to a CCD.
- Flight camera attached to HD camera jello platform.  Advantage, jello free flight video.  Easy to remove without disassembling model.  Disadvantage, camera can be ripped off in a crash.
- Several different size screws used throughout the aircraft.  Advantage, saves weight, less cost to produce.  Disadvantage, complicates crash repair or airfield maintenance since more tools required.
- Carbon fiber thicker than necessary for arms and likely thinner than required in center of body (e.g., next to crash plate).  I'm not an aeronautical engineer so I make my observation based upon experience working with other racing drones, crash survivability of my other models, etc.
- GoPro camera mount.  Desirable to be able to install without having to take the quad apart.

Ideas for future improvements
- Option to run batteries up to 6s.  New motors like T-Motors F-60/F-80 series racing motors include support for 6s batteries.  Other ImmersionRC products like their video transmitter support 6s batteries.
- Option for integrated GPS.  Hard to add after the fact.
- Slightly longer receiver pad to better accommodate ezUHF receivers.  ezUHF does fit but 1cm additional receiver pad length would be helpful.
- Neoprene landing pads for softer landings on hard surfaces.
- 600mW video transmitter.  350mW is ok but 600mW would be better.  Options to replace module with other frequencies like 1.3Ghz.
- Ditch the stock props and work a deal with DAL.  DAL's 5x4.5 are much better than the stock props and come pre-balanced.  My stock props included with the Vortex were out of balance and produced loads of vibration.  BTW, DAL makes matching red props.
- More flexible video transmitter antenna option.  Removable spring loaded rubber.

Radio Settings
Throttle expo=15%, ailerons=25%, elevators=25%

PID Settings
Role, P=3.1, I=.021, D=16
Pitch, P=3.6, I=.023, D=16
Yaw, P=0.1, I=.045, D=10
Level, S(a)=9.0, S(h)=0.10, T(h)=100
Roll Rate=.46
Pitch Rate=.46
Yaw Rate=.46
TPA=0
TPA Break=1500
* I offer the previous PIDs as a starting point for the stock motors and DAL 5x4.5 props.  I will post future improvements as I make them.  Still working on this PIDs but they should get you flying without too much vibration.

ImmersionRC is not advertising the Vortex as a beginner aircraft but even so my experience was better than I was expecting.  It's tough to build a racing drone anyone can fly directly out of the box but ImmersionRC does a great job.  The Vortex 250 Pro is 90% of the way to what I consider a flyable racing drone.  Contrary to ImmersionRC's advice, if you are a beginner you probably want a Vortex.  Trying to build a racing drone on your own is far more complicated.  If your a complete beginner, I suggest starting out with a low cost Hubsan X4 series drone and start watching some drone racing videos.  Once you have some experience under your belt, understand how the sticks work, step up to the big boys club and grab a Vortex.  I'm happy to have a Vortex in my collection.

Following is my horrible maiden flight.  Ouch, but I share the good and the bad.  Hard to see on the video but the maiden flight had so much vibration I felt more comfortable flying in angle mode instead of rate mode.  I returned home, removed the stock props and mounted DAL props, improved the PIDs, and I follow up the maiden flight video with a much better flight video demonstrating the amazing power of this aircraft.  No fancy flying but it provides fair idea of what your experience may be like.


Video: Vortex 250 Pro Maiden w/improved follow-up flight.  No music.  In my haste to fly, I temporarily mounted camera to jello mount platform using velcro.  Using the included mounts provides better video results and aims the camera a few degrees up so your not looking at the ground so much.

Wednesday, March 2, 2016

REVIEW: Lumenier Neoprene Landing Pads

Have a small(<300mm) racing or acro quad and want a landing gear solution?  I found these Neoprene Landing Pads by Lumenier online.  The back of the pad comes with plastic tape covering an adhesive service.  Pull off the tape and stick to the underside of your quads arms.  That's it.

Landing carbon fiber on any kind of hard service can loosen bolts so absorbing the shock is important.  This is one of the reasons I never like the carbon fiber clip type landing gear shipped in the quad kits.  I purchased several neoprene landing kits from GetFPV.com in the event one I loose a pad.  So far none have fallen off.  I think these will be popular with all except the most hardcore racers.  Another benefit is that the pads are very low profile and unlikely to get snagged on objects in proximity flights, very important.  The pads are sold in a 4-pack for around $4US and weigh around 1g for 4.  A great low cost upgrade anyone can do.  Thumbs up!

* Image from GetFPV.com.

Sunday, January 31, 2016

Parallel LiPo Battery Charging

I have been charging my batteries sequentially until recently.  I saw some discussion on the Hardcore FPV Racing thread on parallel charging.  Parallel charging is where several batteries can be charged simultaneously.  There's a great video on YouTube that gave me the idea.

The gear featured in the video is the following.

Turnigy Accucel-6 50W 6-cell balance charger, from HobbyKing
RMRC Parallel Charging Board - 35A XT-60, from RMRC

If you purchase these components they will connect up no problems.  There are a couple of rules for charging.

  • You can only charge a single type of battery per charger.  In my example photo, I could add 2 more 4s batteries.  Don't mix 3s and 4s batteries on the same charger.  Of course, you can run a 2nd charger if your power supply support it and charge 3s on that charger.
  • Make sure each battery is discharged to around the same level.  Don't connect a fresh and a dead battery or something strange like that.  The Paraboard has thermal protection but it's best if you observe this rule
  • To charge at 1C multiply the number of batteries charged by the mA rating.  In my example photo, my batteries are 1800mA and I am charging 2 batteries so I set my charger for 1800mA x 2 = 3600mA, charging 4 is 1800 x 4, etc.  Of course, make sure your power supply and charger can source the current.

For my project, and unlike the video I provided, I had a iCharge 208B on hand.   The iCharger 208B is great charger and supports up to 8s batteries.  Unfortunately, RMRC's Paraboard does not plug directly into my charger out of the box.  To get it working you have to pop the pins out of the 6-cell Molex connector into a the 8s connector that was supplied with the iCharge 208B.  To pop them out slide a small screw driver under the Molex pins as shown and give the wire behind a tug.  It should come out easily.  Plug each wire from the Paraboard into the corresponding numbered cell on the iCharge 208B side.  When your done there will be two pins unused (cell 7 and cell 8).

Before you begin charging, plug you batteries balance port into the Paraboard.  You should see the voltage readings on your charger and a red LED on the Paraboard blinks.  If you don't something is up with your wiring.  Always supervise your batteries while charging.



Wednesday, January 20, 2016

Challenges With RTF FPV Racing Drones for the Public

My background is software application security and programming with a side interest in electronics.  I consider myself a "technical" person.  More than a few times I thought perhaps I over extended myself and my capabilities when I began building my first drone.  Since then, I have built a couple different aircraft (featured on my site) and finished up a microwave ground station in a backpack.  I have a little more than a year experience and I'm still learning.  Drones are capturing the interest of the public across the globe but small platform racing drones have been slow to catch on due to a number of issues.

Software & Electronics Expertise
Building your own racing drone requires expertise across the a variety of technology disciplines like soldering, knowledge of electricity, reading digital multi-meters, RC model building, etc.  While much of the best software is open source and free, it takes some understanding to apply the patches.  Patching some components requires building custom cables.   There is much to learn if you don't come from an RC modeling background.

Configuration
I include configuration separate from software since configuration deserves it's own section.  In fact, it's a bit of an art form.  A successful build is useless without proper configuration.  The type of configuration you need depends upon the components and brand of components you use.  Most flight controllers require some configuration for smooth flight.  For example, typical small racing drone design places camera on the top at the front of the aircraft and battery on the rear.  This means it takes slightly more force for the flight controller and motors to make corrections to the front and back of the aircraft than it does side to side.  These differences must be accounted in configuration.  Since there is no standard build there is no standard configuration.  My settings are specific for my aircraft.  Another example, Electronic Speed Controllers(ESC) are driven by the flight controller and often require software different software since may of these ESC's are intended for model aircraft.  Flying multi-rotor is a far more demanding environment than fixed wing since multi-rotor aircraft have no control surfaces and thrust vectors are controlled entirely by differences in motor RPM.  Most of us do not run the stock manufactures ESC code but instead apply custom open source software like SimonK or BLHeli.  Flight and FPV cameras also require configuration for optimum performance.

Air Worthiness
Experience is a painful teacher.  I learned early on through trial and error the type of acceptable construction and configuration short through needless crashes that could have been avoided.  I have had shorts in carbon fiber bodies that resulted in frame warping.  I have had screws that were checked before flight unfasten themselves in flight.  Building a reliable aircraft can be done but it takes some education.  My very limited experience provides me genuine appreciation for Boeing's safety record.  I don't know, but I can imagine, the additional challenges flying aircraft reliably at 45,000ft or more.

Expense
There is no standard with expenses and it's changing all the time.  When I first started there was nobody in my area building racing drones.  At least, any that I know personally.  It cost me a lot more money than it should since I made a number of mistakes.

Government
The facts, flying radio controlled models has been around for 50 years.  Not a single individual has been killed by a hobbyists drone.  No hobbyists aircraft has ever been used to deliver an explosive payload or even in the commission of a crime (so far as I know).  It's difficult to understand the origin of disproportional concern from governments around drones.  Perhaps the fear is historical, the term "drone" has been used mostly to describe UAV aircraft governments use for surgical strikes over the past 20 years.  Along the way the term was unwittingly co-opted to multi-rotor and hobby aircraft equipped with FPV gear.  Now the hobby community is surprised and frustrated fighting an uphill battle to improve sentiment.  Government concern around drones may discourage public interests but it definitely slows commercial developments like Amazon.

Crash Physics
Most of the racing drones frames are made from carbon fiber.  Even so, a crash from a significant altitude and speed does significant damage to an aircraft.  The problem with learning to fly is that the ground is hard.  A challenge to any Ready To Fly(RTF) platform is to develop easy to replace plug and play components, limit ground time, and UPS delivery wait times.

Everyone's RTF Different
To the consumer RTF means everything you need to put an aircraft up in the sky and have some fun.  Potentially this is a lot of gear like: aircraft(frame, motors, flight controller, ESC's, flight camera, FPV camera, receiver, video trx, OSD, SD Cards, wire and misc, etc), transmitter, LiPO batteries, battery charger, FPV goggles or TFT, antennas, and more.  Making the consumer successful at flight requires simple integration, reduced configuration, packaging options, and low-cost effective replacement options.  Flying racing drones is not like flying camera platforms.  There's going to be some crashes.  Retailers should work with manufactures and look at this as an opportunity to build an ongoing relationship with the public.

Bringing success to the public is one of the hallmarks of DJI and the Phantom platform.  The Phantom is larger camera platform and not a racing drone but making flight simple and customers successful has increased public interest and also interest that spills over into racing drones.  Commoditizing the electronics for racing drones, reducing configuration complexity, improved RTF component packages, easy access to individual low-cost replacement parts, will go a long way to put racing drone technology into the hands of public would be fliers.  I will continue to fly my research projects and enjoy them but the addition of simple RTF options are a welcome improvement.


Tuesday, January 19, 2016

Portable FPV Microwave Ground Station in a Backpack

A ground station is an enhancement to improve your video signal from your FPV aircraft.  There are a number of ways to build a ground stations depending upon your goals.  Two styles are popular, backpack builds and larger antenna trackers on tripods.  In my case, I wanted something portable, light weight, with everything I need for a few flights.  If you have built a multi-rotor or two you undoubtably stumbled upon Youtube videos about man packable ground stations.  This is a very simple to build project.  I doubt it offers the range of larger setups but it's great for my needs - park flying.  Quick list of the benefit if your not sure why such a pack could be helpful.

Benefits
- Very portable: holds 3 flight batteries, 1 battery to electronics in pack, transmitter, aircraft, tools.
- Better FatShark FPV goggle battery life: I have the V1 which only includes a 1000mAh 2S battery.  After the 2nd flight my goggles are beeping low battery.  Moving to external battery reduces anxiety.
- Improved FPV reception: Duo Diversity receiver has two antennas and switches between the strongest signal for added safety.  I chose a high gain patch antenna for flying in front and a omni directional when I pass behind the pack.
- Handy access: plenty of room to carry transmitter, batteries, aircraft, FPV gear, extra props, hex drivers, etc.  Including place to carry tripod on the side, if you wish.  Hang pack from hook on tripod for extra distance.
Click to expand

Parts List
- Lowepro Transite Sling 250 AW (Camera backpack, BestBuy)
- 8" x 6" 1/4" Plywood
- ImmersionRC Duo 5800 V4 Diversity Receiver Race Edition (ReadyMadeRC.com)
- IRC 5.8GHz SpiroNET Omni-directional SMA Antenna Set (RHCP) (ReadyMadeRC.com)
- IRC 5.8GHz SpiroNET Patch Directional CP Antenna 13dBi (RHCP) (ReadyMadeRC.com)
- SMA Female to SMA Male RG316 Extension (15cm) (ReadyMadeRC.com)
- Misc: HTX battery connector, wire, heat-shrink tubing, solder, velcro Command Strips, 2 velcro 12" velcro straps, M3 screws, 2A circuit breaker/fuse(larger if TFT used), etc

I looked at several backpacks but settled on the Lowepro due to it's small size, adjustable configuration of the compartments, the pack sits flat on ground without falling over, and BestBuy had a deal at the time.  Some considerations, the pack must be room enough to safely stow a FrySky Taranis transmitter, 4 batteries (3 for flight, 1 FPV gear/receiver), room at top for receiver and antennas.  When I began I wanted to build a pack with a small 3ft telescoping mast for the patch antenna.  Instead I choose simplicity, in the current design I mount the patch antenna directly to a plywood board within the pack and leave the front flap open when I fly.  I think there is a joke here but ah never mind.The BuildThe build is ridiculously simple and no frills.  First I cut a piece of plywood 8"x6"x1/4".  I purchased a piece of Poplar wood in a hobby store display box or the local hardware.  On this board you will mount the receiver and patch antenna.  Next, cut some holes in on the right side of the board for your patch antenna.  To get correct hole alignment and sizes I taped a piece of paper on the back of the patch antenna and rubbed the holes with a pencil.  Once you have the correct layout, tape the template on the board and drill holes as necessary.  Use self-adhesive velcro Command Strips to securely attach the patch antenna in addition to screwing the antenna to the board.  You will need M3 screws long enough to pass through the board and mount to the antenna.  Next, mount the receiver with the velcro straps. More to do...I need to change the FPV backpack power connector from XT-60 to HTX style.  The reason is that the FatShark goggles support a maximum of 3s battery pack.  All my flight batteries are 4s with XT-60 style connectors.  I know one of these days, when I'm tired, I will plug in a 4s battery and their will be a mushroom cloud in the my backpack.  To make sure I don't make that mistake, I will use HTX style connectors for the backpack power supply and XT-60 style connectors for my aircraft.  I also plan to sew loops on the front of my pack as velcro points to secure my aircraft.  I don't want to carry the aircraft if I'm hiking.  As a safety measure, I need to add a fuse or circuit breaker to the main backpack power supply, 2A-3A should be adequate.  The antennas range is probably attenuated due to proximity to the ground.  I will include future tests to see how this configuration compares to FatShark mounted antenna.  The pack includes an exterior pocket and strap for carrying a small tripod.  I was thinking it would be interesting to hang the pack using the top strap on a tripod to get the patch antenna off the ground.  The interesting testing scenarios I'm considering, open to suggestion, are 1) FatShark mounted antenna, 2) pack on ground, 3) pack suspended via strap on tripod.  I have also considered adding a 5" TFT display so others on the ground can see what I see as I fly.  It's a good way to share the experience with everyone on the ground.  If you add this to your pack you should quickly review the current draw to make sure the circuit breaker\fuse will accommodate the increased load.  I will update this article in the future but hopefully this is enough information to get everyone started.  Happy flying!

When Will Delivery Drones Hit US Skies?

No date is determined but commercial use of a drones will not be permitted in US airspace before 2018.  NASA/FAA presents the following test plan for the public,

UTM TCL 1: Concluded field testing in August 2015/
ongoing testing at FAA site. Addressed rural UAS
operations for agriculture, firefighting and infrastructure
monitoring. In this TCL, the UAS ground pilot reserved
the airspace and adjusted the flight plan if notified of
a conflict.

• UTM TCL 2: Tests in October 2016 to address beyondvisual
line-of-sight operations in sparsely populated
areas, and provide flight procedures and traffic rules for
longer-range applications.

• UTM TCL 3: Tests in January 2018 to include cooperative
and uncooperative UAS tracking capabilities to ensure
collective safety of manned and unmanned operations
over moderately populated areas.

• UTM TCL 4: Test dates to be determined. Would involve
UAS operations in higher-density urban areas for tasks
such as news gathering and package delivery, and largescale
contingency mitigation.

There are always some exceptions like the FAA's Section 333 exemption permitting limited commercial use of drones for some approximately 3000 applicants.

Tuesday, January 12, 2016

ZMR250 Build Log

Updated on March 7, 2016
This is the second installment in my drone building saga.  For those interested, check out my hexcopter build, Blackout Mini Hexcopter.  I wanted to try the ZMR250 quad since it looks like they are lighter and fly more precisely.  At least once the aircraft is configured appropriately.  If you are entirely new to flying I recommend starting with a Hubsan X4 quad since crashing larger drones is costly and requires significant effort to repair.  Make your mistakes on lower cost aircraft.

This is an advanced project.  Anyone uncomfortable with computers, software, and using a soldering iron should probably stay far away from this hobby.

Parts List, ZMR250 Quad Multi-Rotor Aircraft (current build)
ZMR250 Carbon Fiber Airframe:  UBAD Kunai PCB,Tiger T-Motor MN2206 2350kv (x4) (banggood.com frame, motors anywhere you can get them)
ESC: KISS 20A v1.4 (x4) *KISS ESC's are awsome!  No firmware flashing required (or supported).  Support for both PWM@500hz and Oneshot125.  I'm running Oneshot125 it's great!
Battery: RMRC Orange Series1300mah 4S 60C Lipo (readymaderc.com)
Prop: DAL 5x4.5 v2  CWx2/CCWx2, 1 set shipped with each unit, what a concept (rmrc.com)
FC: Abusemark Naze32 Acro Version running Cleanflight v 1.11.0
Receiver: FrSky D4R-11 4-ch 2.4 Ghz/ flashed to 27ms 8-CH (hobbyking.com) [4]
FPV Camera:  600TVL Sony Super HAD CCD D-WDR Color Board Camera with OSD Menu DNR
FPV TX: 5.8 GHz 600mW FPV Race Band from ImmersionRC
DVR Camera: SJCAM SJ4000 (GoPro clone, bh.com)
Standoffs & Misc: Bluetooth Mini LE module (Blackbear Labs) required for Apple BT support, antenna: 5.8Ghz SpiroNet RHCP antenna from ImmersionRC, battery straps: Scorpion Lipoly Lock Strap 205mm (1 ea x 3 Small), 3D printed vented Uni-Spacer, blue 1ea (eBay or Shapeways), 3D printed 10 degree motor mounts (droneplastics.com).

I learned a few things from the first build.  My original 30A ESC where way too big and didn't fit on the aircraft frame properly.  I wanted something smaller so decided to go with smaller and popular KISS ESC's.  I eliminated the OSD.  The OSD is mixes the telemetry with your video signal to provide helpful information throughout your flight.  The flight time on these small aircraft is around 6-10min and they fly far too fast to be watching the daisy's.  I found I was not paying attention to the OSD anyway.  I also improved the video and moved to 5.8Ghz system.  Race drone technology is always improving and since I started about 1 year ago the video systems have improved.  The race band transmitter allows up to 32 aircraft to fly together.  Since I dropped the OSD I downgraded to the Naze32 Acro edition.  An unintended consequence is that the Full Naze32 has extra memory for the Blackbox feature.  So even if you don't care about the extra telemetry you may care about using Blackbox.  An ancillary benefit of the video frequency change is that the antennas are much smaller using 5.8Ghz.  ImmersionRC makes a great transmitter as well with built-in filtering and voltage regulation.  Unfortunately, it's also hugely overpriced just like their OSD.

Photo 1: 1st aircraft build (click to expand)
Photo 1, is the more or less the finished quad in my first build.  I have serval hover tests and tried 5" and 6" propellers, 3s and 4s batteries, etc.  I recommend running this aircraft with 6" props since it hovers around 1/4 throttle with 4s batteries.  The 5" props hover around 3/4 throttle.  To run 5" on my motors I would probably need 3-bladed props.  I didn't realize when I began the project the D4R-II receiver had to be reflashed to support 8-channels.  You will need to grab a cable to do that so do your homework if your purchasing one of these transmitters.  In the first build it was really hard to accommodate all the components in the 250mm frame. I still have improvements to make.  The battery cable comes dangerously close to the large 6" props.  I need to lock these cables down better.  Next I have the RC receiver antenna coming off the rear of the aircraft(wrong polarization angle).  I need to reorient this somewhere else, presumably the front of the aircraft sticking up offset from one another 45 degree angle like the old style rabbit ears.  It looks funny but there's not a lot of choice.
Photo 2: UBAD Kunai PCB board
Configuration of the aircraft involves plugging in a USB cable into the aircraft and running Cleanflight configurator on my Windows 10 computer.  I had an unfortunate incident where the back of my flight camera contacted the carbon fiber frame and exploded in a ball of flame.  Since I needed to reinstall the camera I decided to clean up some other stuff I didn't like.  I decided to return back to the PCB and purchased a UBAD Kunai PCB, Photo 2.  The PCB eliminated tons of wiring and really cleaned up the build.  The UBAD board adds a little bulk to the bottom of the board with a voltage regulator.  Since I needed some space on the bottom I decided to run my wires on the bottom and purchased a vented uni-spacer (shown in photo 4).
Photo 3: Top view
Photo 3 is the same stage of construction as photo 2 except looking at the top.  The small board on the top is the Naze32 flight controller.  Notice there is no wiring to the flight controller, it's all in the PCB.  The blue tape on the legs is to keep the ESC's from touching the frame until I lock down the heatshrink tubing to insulate them.
Photo 4: Bottom view, almost finished
Photo 4 is the bottom view almost finished.  I have added a 5vdc regulator, shown in black heatshrink.  The Bluetooth module is shown just above.   A warning on the bluetooth module.  I made a mistake and did not purchase the Bluetooth LE version.  The mistake means I only have support on my Android tablet.  If you want to use your Apple gear to configure your aircraft in the field you will need to purchase a Bluetooth LE module.
Photo 5: Configuration of flight control software
Photo 5 show the aircraft at configuration time.  The blue tape on the top deck is to hold the camera platform stable while my epoxy dries.  The blue table on the motors is so I can see which way the motors rotate.  Unless I put something on the motor it's tough to see which way they are rotating.
Photo 6: Finished Aircraft


Photo 6 is the finished aircraft with LEDs.  The LEDs are super bright.  I tried a quick hover test just before sunset but my Fatshark goggles ran out of charge.  I don't really need them for hover testing anyway.  The aircraft flew rough with high speed oscillations.  The heavier the thrust the more severe the oscillations.  I'm thinking the PIDs configuration settings, specifically the P values, are too high.  I am pretty optimistic about this aircraft build. A small word about Finished caption on Photo 6.  When you get into this hobby you will find you are never really finished.  There is always some improvement to make.  Perhaps a better description is flight worthy.  I plan to post some updates in the future.  Like my maiden flight.

Video 1 is the maiden flight.  I had a number of configuration issues along the way.  Finally got the major bugs worked out.
Video 1: ZMR250 quad maiden flight
Video 2 is a little more interesting.  I few more flights under my belt.  My PID configuration needs some slight adjustment since I change the props from 6x4.5 to 5x4.5x3, a tri-blade design.  Anytime the power to weight ratio changes it messes with the configuration.
Video 2: ZMR250, a more interesting flight

Blackout Super Mini Hex Build Log

Prior to building or even buying a multi-rotor craft you need to learn something about the hobby.  There is a lot to know that could influence your purchase decisions.  This article covers some of my experiences along the way, hobby basics, skills necessary, and what you need to get started.
First thing is first, before you do anything what are your goals for your aircraft?  Do you want an aerial platform?  If so, then you need a top-notch camera platform.  You will need an aircraft with a long loiter time.  Stability is important, so you will need a gimbal mount for your camera and an aircraft large enough to carry everything.  Are you interested in autonomous flight?  With proper navigation systems and software it's possible to build an aircraft that fly's autonomously along a set of waypoints to a destination you specify.
I'm interested in a racing platform for my first multi-rotor and it's what I will be covering.  I need low latency video to fly so I don't hit anything, higher quality 1080p video suitable replaying aerials on my big screen, in flight electronics to help keep the aircraft stable in winds, electronics to help navigate since it's easy to get lost when flying out of view.  Of course, powerful motors since it's got to be fast.  What is fast?  About 100mph or 160kmh to be exact.  Following is my parts list for this project.

Parts List

Blackout Super Mini Hex Spider Multi-Rotor Aircraft
[1] Have not testing this yet but looking forward to trying.
[2] Tried imitations and they don't work as well.
[3] Clone purchased not to save cost but on the chance quality of the hardware will improve.  Had problems with hardware quality of boards made by Abusemark.
    Equipment Tried, Removed, and Why
    • GPS: Whitespy “house” uBlox 6M GPS.  GPS frequency is close to my video transmitter frequency.  I tried various filters to get this working and was unsuccessful.  Others have succeeded so I think I could find a better combination of filters, transmitter, antenna but I decided to pull the GPS.  It's pretty useless on a racing craft anyway since batteries limit range.
    • FC: Naze32 FunFly (Full Version)  I am giving up on a Naze32 hardware made by Abusemark.  I have had two and had strange problems that are difficult to diagnose.  I like the architecture but don't like the implementation.  **May 9, 2015 CORRECTION:  My problem turned out to be power not Abusemark's Naze32.  
    • FPV Camera:  Sony 960H CCD Effio-V 800TVL Ultra WDR Camera One would assume since this camera has higher resolution that it's better for flying but it's not.  The 600TVL model is older but better since it adjust contrast super quick.  For example, remove the lens cap and the camera adjusts the white balance almost instantaneously.  This is important if you fly into the sun momentarily and then turn into the shade.  3-4 seconds to readjust the white balance is unacceptable.  Also the contrast is much better, if you do face the sun then it's still possible to see objects in the background.  The sun does not overwhelm the camera.  The image processing is also super fast and does not skip.  Older is sometimes better.
    Ground Support
    • Spektrum DX8 DSMX Transmitter Mode 2 (amazon.com)
    • FPV RX: Fatshark Dominator v2 Headset System (getfpv.com) 
    • FPV RX Module: 1.3Ghz Fatshark Dom. Module (laserbgc.com)  1280mhz, 1320mhz, 1360mhz 
    • FPV Goggle Power: Fatshark 7.4v 2S Battery (getfpv.com)
    • iCharger 208B 350W (hobbyking.com) 
    • PRC350 Power Supply (progressiverc.com)
    • Parallel (6x) HXT 4mm Bullet Charge Cable (progressiverc.com)  
    Misc
    • Liquid electrical tape
    • Du-Bro 499 Tru-Spin Prop Balancer
    • 3M double sided sticky tape
    • Antenna SMA Panel Mount: SMA male plug to SMA female jack (qty 3)
    • Bulkhead right angle RG316 10cm pigtail
    • SMA Male to Female Right Angle 90-Deg Adapter w/ Gold Plated Contacts (RF-M103-2 2 pack) 
    • Blue painters tape
    • 1/4” and 1/2’ heat shrink tubing
    • Heat gun (or use mini blow torch)
    • helping hands, clips holders 
    • Rosin paste
    • 12awg wire for power
    • 28awg rc trx\rx wires
    • USB ASP, programmer for ATMEGA328, miniOSD (Amazon.com) 
    • FTDI connector, USB programmer Naze32, (Amazon.com)
    • Servo Extension Lead Wire Cable (10ct)
    • Scale 6kg max (Amazon.com) 
    • Mobius Audi/Video Out Cable (ebay) 
    • Mini 3A DC-DC Converter Adj Step Down Power Supply Module ( qty 10)
    • Generic Break Away Right Angle Headers 0.1” 3x40 Pin Gold Plated (pack of 5) 
    • Lester Pocket Pack Solder 60/40 0.031 0.50 oz Tube 
    • RF coaxial coax adapter SMA female to SMA female bulkhead (pk of 2) 
    • 1/2 inch PET Expandable Braided Sleeving - 10ft - Black 
    • HXT 4mm Connectors (2 pairs) - Red 
    Features & Characteristics
    Aircraft weight w/3300mah bat, 1069g
    Flight Time: around 7 mins Hover: approx 40% throttle
    Sensors & Control
    Altimeter\Barometer (altitude from sea-level)
    Magnetometer (compass heading)
    Gyroscope\3-axis (accelleration in each dimension)
    Flight Controller w\GPS (assisted flight, position, altitude, navigate home) DVR x2; dvr 1, records 1080p flight video.  dvr2 records flight cam w/telemetry data

    Lesson Learned

    I will update these as I go along.  No particular order of importance.
    1. Not a project for a beginners.  Make sure you feel comfortable with soldering iron, working with electronics, flashing hardware, etc.  If you have these skills you should do well.  There are some ready to fly racing drones.  I would avoid these since your almost certainly going to crash when learning.  Building your own offers some advantages when it comes time to rebuild, you know what each and every wire does.  Starting with someone's finished product make make it difficult and expensive to service.  
    2. Carbon fiber is conductive.  Don't place bare PCB's or electrical conductors against carbon fiber bodies.  Makes sense since resistors are made from carbon but may not be immediately obvious.
    3. Different propellers for different directions.  On multirotors, half the motors spin in different directions (Clockwise or Counter-Clockwise) to balance rotation forces.  You need to order the right propellers depending on the spin direction.  You can't flip a propeller upside down.  Some propellers also fit better than others, even with the plastic adapters they ship.
    4. Propeller adaptors are made for different directions.  These are the metal pieces that bolt on top of the motor.  A shaft sicks out and the propeller fits on this shaft.  I'm using T-Motors and I noticed T-Motor makes prop adaptors for CW or CCW depending on your needs.  It's best to use the right adapter for the job since these parts are under a lot of forces.
    5. Don't believe everything you read.  Some people give bad advice like running 4S batteries on 3S rated gear.  If the manufacturer says 3S then your taking a chance if you run 4S.  Fires are no joke.
    6. Carbon fiber is super hard.  Steel screws don't snug well against carbon fiber since it's so hard.  You need washers and Blue Locktite (don't use Red) to keep screws tight under stresses of flight.
    7. 1.3ghz video signals interfere with GPS.  Bleed over from the video transmitter interferes with my GPS so I cannot get satellite lock.  1.3ghz for video was a good choice but it requires a couple of improvements to work with GPS on my small multirotor.  I'm purchasing a lower power 300mw transmitter and retiring my 800mw unit.  Next, I ordered a 1.4ghz low pass microwave filter.  I'm hopeful the two changes will reduce noise and allow my GPS to lock.
    8. RC transmitter.  My first choice for a transmitter was FrSky Taranus.  I did my research and I should have gone with this choice.  The Taranus offered a number of out of the box improvements like, module bay for moving to UHF.  My current transmitter supports UHF as a box you strap onto the transmitter, very clunky.  Next, Taranus offers more channels, it also offers RSSI which is received transmitter signal strength.  Taranus also provides options to encode and assign the RSSI to servo channel, really handy.  RSSI is like receive strength bars on cell phones and really handy for flying if your getting out of range.  My current radio does not support this so I need to purchase an extra electronic module (e.g., DORA) for RSSI.  Live and learn.  Once I get some flight time I may turn around and sell my current radio.  Not sure I want to commit to this technology.  Another cool thing about Taranus is transmitter code is open source.  How cool is that?  It would be fun to see if I can hack the transmitter code.  More on that later.   
    9. SMD soldering tip.  When soldering SMD components don't use flux.  Components are exceedingly tiny.  If you flux pads on both sides of a solder connection the solder will usually travel down the pin to the pad on the other side and build up a huge blob somewhere along the way.  When soldering one of these slathered flux joints it appeared like all my solder was disappearing but in fact it was building up in a big blob in the middle of a right angle connector that was very difficult to desolder.  I would avoid flux and tinning altogether with SMD.  An iron with a super small tip would be helpful but you can do the job with a normal tip if your careful.
    10. Transmitter ON!  Always turn transmitter on prior to applying power to any aircraft.  Make sure motors are Disarmed, assuming you have disarm assigned to a switch.  Make sure the throttle stick is in the low\off position.  I have seen the motors come on under Cleanflight when power is applied to the aircraft and the transmitter is off and when the transmitter is turned off while the aircraft is on.  I don't trust some of this flight controller software just yet.  I want to make sure it's receiving zero throttle and motor disable signals from the transmitter unless I'm ready to fly.
    11. Flight data handy.  Recording your telemetry video (flight video+OSD) is handy to diagnose crashes.  See if you can figure out what happened here,  VIDEO.
    12. Mini H Better First Build.  A hexcopter is cool but in retrospect a 250-300mm class mini-H type frame would be better for a few reasons, 1) much more agile.  Hex's have other advantages but they are not as agile and weigh more out of the gate.  2) everything is more expensive.  A hex has two more motors, two more ESC's, everything costs a little more.
    13. Nothing wrong with offshore purchase.  Almost every part of this aircraft is made in another country.  I have had few problems purchasing products overseas with the exception of longer shipping times than I would like.  Don't worry about purchasing something overseas.  Enjoy some savings yourself and buy something from China or another country directly.

    Aircraft Assembly

    A word of caution before I get started.  To build your own multirotor you must be comfortable with software, programming, and a soldering iron.  There are a number of people who have done this but this project is not a kit.  There are no step-by-step instruction manuals.  It's easy to buy the wrong parts or damage parts you do have.   If you don't have a few friends close by that have done this then plan on a month long project.  You will must also plan some time for project oversights, which turn into more parts you need, shipping delays, and stretch out your build. 
    Photo 1: parts, parts, and more parts.  All parts not shown
    So this is what you get when you get started, bags, and bags of parts.  The UPS, FedEX, and the PostOffice will think your starting your Christmas shopping early.  I recommend you keep a list so you check off parts you order and when they arrive.  If anything is missing you will need to work with vendors to track down your order or reship.  I know this seems like a pedantic detail but as you can see from the parts list this is a big project so your up-front organization will save you some time down the road.  A parts list is also useful if you want to write build log someday and share your experiences.
    Photo 2: airframe with signal wires
    I began my project by doing some rough assembly of the carbon fiber airframe.  It's a hexcopter so it has 6 arms with a motor on each arm.  The frame is assembled as a sequence of flat plates with breakaway plastic standoffs (like a computer but longer) bolted together.  Sandwiched between is custom PCB to carry current between major components and cut down on wire weight.  For example, Electronic Speed Controllers (ESC) mounted on each leg are soldered to large pads on the PCB located by each arm.

    "A hot flame jet erupted from the FET about the size of a pencil.  Let's just say, it got my attention."

    Photo 3: Blown Electronic Speed Controller
    A word of caution, carbon fiber aircraft models are the state of the art but keep in mind carbon fiber is conductive.  How conductive your model is depends upon where electrical contact is made, the type of carbon fiber, as well as resins and coating applied during manufacture.  I found an interesting video on Youtube of meter tests on various carbon fiber weaves.  Keep in mind, there are no resins or coatings on the weaves in the tests but it's good information to file away for the future.  In my case, I had an uninsulated ESC strapped the leg of the craft catch on fire at my desk during configuration of the flight controller.  A hot flame jet erupted from the FET about the size of a pencil.  Let's just say, it got my attention.  The burned out FET shown on the lower left in Photo 3.  I checked the ESC for wiring problems but found none.  One further research and reflection is likely due to either, 1) runing 4S batteries on 3S ESCs and motors, 2) or the carbon fiber frame.  I checked the frame resistance with my multimeter and results where all over the place.  Some spots showed resistance of a few ohms and many spots much higher.  Ether way, I have more ESC's on the way and i'm switching to 3S batteries for safety.  Points to note, carbon fiber is conductive.  Don't attach electrical gear directly to carbon fiber bodies.   If the documentation for your ESC's and motors say they support 3S stick with manufacturers recommendations.  Don't do what other knuckleheads on the Internet are doing.  I found lots of really bad advice in my search for information.

    Video Systems

    Photo 4: flight video system.  Not all components shown
    There are two discrete video systems supporting the aircraft.  The first system is used to fly the aircraft.  The second system is used to produce a better quality recording, 1080p to share with family and friends.  I will discuss the system used to fly the aircraft first.  In photo 4 you can see some components of the flight video system, Fatshark goggles, 1.3ghz video transmitter, the control module for configuring the video camera, and my power supply so I don't have to run on batteries inside the house for testing.
    Photo 5: Testing Fatsharks and Flight Camera
    Photo 5 shows a goofy looking guy with a circularly polarized antenna hanging out the side of his head, yup that's me.  Notice the photo is grainy.  It's a still image made from the flight video.  Generally, flight video systems don't produce the best quality images but their video is low latency and high contrast to show sticks and twigs, moss hanging from trees, or other debris that can obstruct your flight.  Following is a great video comparison why two cameras are necessary.  The video at this point is only the raw video feed and does not include any telemetry information from the flight controller.  The edges are also curled up since I replaced the stock lens with a 2.8mm wide-angle lens.  The 2.8mm lens provides a slightly larger angle of view for flying without too much distortion.  Of course, everyone in the house is curious what I'm building.  My girls liked the video system.  They discovered you can put on the Fatsharks and look at the back of your hair to see if your having a bad hair day.  They came up with ideas for other practical uses as well like mounting the camera on your back or shoulders to see who's checking you from behind while looking forward.  They are beginning to think like security people.
    Photo 6: Fry's electronics store
    Photo 6 is the Fry's electronics store.  If your ever missing a part.  Need a temperature controlled soldering station or a duel-trace oscilloscope perhaps?  Fry's is a the place to go.  I'm generally not disappointed when I need a part fast.  Unfortunately, I went to 2 Fry's stores and could not find a SMA female to SMA female through bulkhead connector.  I also looked in several hobby shops for more ESC's and nobody had what I needed.  The point being, some parts are even too nerdy for Silicon Valley.  Imagine Woz trying to assemble the first Apple computer part by part before the Internet.  Phew, no small task to get all the right parts I'm sure.

    Flight Controller 

    Photo 7: Naze32 Flight Controller
    The flight controller is the brains of the operation.  I'm using the Naze32, shown in Photo 7, but there are many others.  Like all components, you need to carefully consider your objectives when making choices and flight controllers are no exception.  The Naze32 is very sophisticated for it's size but it's not necessarily the most powerful.  I choose the Naze32 mostly because of it's small form factor, power consumption, and low cost.  The Naze32 cannot fly fully autonomously.  For example, it cannot guide the aircraft along a set of user defined waypoints to a destination.  Some controllers can do this.  Naze32 runs two flight controller software packages, Baseflight and Cleanflight.  I went with Cleanflight since there seems to be more support.  Cleanflight also offers some advanced features like a screen with sliders that allow manual control of the motors for bench testing.  Manual testing of the motors is useful to ensure soldering joins are secure and the motors are wired in proper direction.  Depending on your model and flight controller, your motors will rotate in different directions to better balance rotational forces.
    Photo 8: Naze32, GPS, OSD, and radio Receiver
    After setting up the flight controller, Photo 8, it's time to get the Ublox GPS setup.  Originally, I ran into some problems with the GPS and On Screen Display (OSD).  UARTS on the Naze32 are a limited resource.  If I plugged in the computer USB cable to configure the Naze32 at the same time the GPS and OSD is plugged in then the GPS and OSD power lights turn off.  The solution here is that you cannot configure all of them from the convenience of your USB cable or at least I couldn't figure out how to do it.  The Naze32 is configured from USB cable.  The OSD is configured via a FTDI connector.  The FTDI board connects between your computer and the OSD.  The FTDI board breaks out your USB cable connection into serial lines compatible with the OSD.  Like the flight controller, different software is available for the OSD.  Mochaboy RC has some great Youtube videos on the subject.  I ran into some difficulties flashing the OSD with KVOSD software.  I noticed on Mochaboy's video he was using version 1.0.5 of the Arduino compiler tool.  You will need to download the older Arduino version to compile KVOSD or you will receive compile errors.  On my OSD I'm planning to display the following flight data: GPS coordinates (for recovery), GPS direction to home (where aircraft disarmed before takeoff), magnetometer, barometric altitude, battery voltage and warnings.  There is a lot of information you can display but it's best to keep your screen uncluttered and display only the essentials.  More advanced flight controllers like the OpenPilot Revo allow telemetry data to be saved and replayed after the flight.  It's kind of a cool option if your trying to tune your aircraft.  The Revo is not much larger than the Naze32 and I considered it earlier on but availability was scarce.

    At this point I still don't have the hexcopter in the air.  However, because of the motor tests I'm confident I'm on the home stretch and my aircraft will fly.   Remaining work, I need to pull the ESC's off the aircraft arms.  Then rewrap heatshrink tubing over each ESC and zip tie them to the aircraft arms.  The heatshrink will insulate the electronics from the carbon fiber.  Remember don't mount uninsulated ESC's or wiring directly to carbon fiber.  I need to mount the Mobius 1080p camerafor sharing high quality movies with family and friends.   After basic assembly and rough configuration is complete I should be able to make the first flight.  I'm going to start small, do a hover test and fly a few meters off the ground.  Best to take it easy until the aircraft and batteries have been broken in.

    Ok, so what's all this got to do with security?  Nothing just yet.  But I figure the best way to understand aerial systems, software, and challenges is to build a multirotor.  In the future, I may look at radios, GPS, or perform a security review of opensource flight controller code.  It might also make a cool pentest platform by mounting my Wifi Pineapple aboard.  I'm not a pentester but I can see how others would be interested in an aerial pentesting platform for some work.  Another idea would be to mount my ADS-B receiver to the aircraft, pi in the sky.  More on the security angle at a later date, small steps, got to get this thing in the air first.  Remember use your security powers for good and have fun.  No bothering or spying on people.  I will post more in the future.

      March 26, 2015 Update
    Photo 9: Flight Camera (click to expand)
    Photo 9 is the view from the flight camera without the GPS information.  The aircraft is not actually flying (disarmed).  Still waiting for a few more parts but as you can see the avionics are working.  Thought some might want to see what the displays look like when your flying.  If you want to check out the open source OSD software that does the telemetry overlay see the project page, rush-osd-development.  The project page provides the full list of capabilities.
    March 26, 2015 Update
    Photo 10: Mostly finished aircraft
    Switch out the prop adapters, put on the props, and ready to fly.  I flew yesterday before sunset.  No videos since I pushed the wrong button on the camera nothing recorded.  The flight was really short, around 10 minutes.  I was surprised how smooth it flew for  the level of wind.  The flight controller works great and the aircraft was completely level.  I will post some video soon.
    Success!!!
    Photo 11: Blackbox data recorder
    Also found some really cool software for a black box flight recorder.  Take a look at the following screen shot and video link.  It's not my aircraft but I thought it's an interesting, Photo 11 and video.  I don't think it's useful on my platform but it may be worthwhile for larger platforms.
    Video 1: MaidenFlight from Milton Smith on Vimeo
    Ok, the maiden flight video as promised, Video 1.  The good news is I got the aircraft up in the sky.  The bad news is I crashed a couple of times. Probably not very smart but I was flying in a 20+mph wind.  The pros fly in windy conditions all the time but noobs like me should wait for better weather.  Still the damage for the day was not too bad, a couple of props, a couple of bullets, some nylon standoffs, a few nylon bolts, and a few Molex connectors.  I can put Humpty Dumpty back together again but I'm definitely holding out for better conditions next time.  I'm sure I will get the hang of it.  I fly my Hubsan X4 pretty wells so I'm sure it's a matter of tuning and getting use to the aircraft  All in all, a fun day.  The lesson here if your a noob your going to crash but be smart and be safe.

    April 8, 2015 Update Received lower power FPV transmitter and 1.3GHz low pass microwave filter.  Hooked everything up, enabled GPS on the Naze32 board, a quick test confirms I now have satellite lock.  Previously, I was running a much higher power transmitter with no filter and it interfered with the GPS preventing satellite lock.  All better now.  Ready to use flight features like GPS hold and Return Home.  Although I'm not sure I trust those features.  I'm mostly interested in the OSD feature to help pilots find home.  The OSD displays an arrow in the display that always points to the spot where the motors where Armed (usually located near pilot).  When your flying around up high it's easy to get lost since terrain looks different higher up so it's handy to find your way back.  I also picked up a DORA module for RSSI support.  RSSI is Received Signal Strength Indicator.  Think 5 bars on your cell phone, RSSI is the same thing.  When flying a signal strength indicator is displayed in my Fatshark goggles by the OSD.  The percentage indicator provides me information how my transmitter signal is being received by the aircraft.  If you receive a weak strength indicator you can turn around and fly back closer to the transmitter or avoid the area if there is interference.  Due to the quirks around radio transmission and reception it's not 100% effective.  It's possible signal strength could go from 100% to 0% and radio loss before you can turn the aircraft around but it's helpful to prevent loosing an aircraft and a good overall safety measure.

    April 13, 2015 Update Not the maiden flight but I'm including it since it's more interesting to watch than my maiden flight.  In my case, building and software was easy part of the project but flying is a challenge.  Moving a craft in 3 dimensions is challenge.  Also configuring the equipment for best control is challenging and small changes in settings make a big different on aircraft control.  A few days ago I figured out how transmitter Expo settings work.  Expo is very similar to variable resolution mouse on a computer.  With Expo it's easy to create a center zone on the transmitter sticks for fine level control while making the edges of sticks more sensitive.  The Expo settings are great for precise maneuvers like coordinated turns and soft landings.  Video

    April 17, 2015 Update Apply nits to the parts list.  Latest flight video and crash for those interested.  Video

    April 24, 2015 Update All repaired from the crash.  Looks like the cause of the crash was either a hardware issue with the Naze32 controller or software issue with Cleanflight (version 1.8.1).  To attempt to minimize the software risk I am switching to a different package Baseflight.  Baseflight is a little less featured and been around longer.  The software is more mature and likely more reliable (my impression).  For now, no change to the flight controller board, Naze32.  Although I did discover there are Naze32 clone boards you can get like the Mullet.  Some claim the clone boards are better.

    May 4, 2015 Update I have flown a couple of times since installing Baseflight.  I noticed while flying "Disarmed" flashing in the compass heading area in my Fatsharks.  This is proof positive my problems are a result of a bad Naze32 flight controller board.  I ordered a Naze32 close, the Mullet32 by Armattan.  I'm purchasing the clone not for cost savings but it's an attempt at finding some better quality hardware.  Also I have been thinking of reflashing my SimonK BS ESC's with BLHeli and Oneshot125 w/active breaking (damping light).  The real motivator for me is watching the following video.

     

    I would really like to upgrade my motors to MN2208's, KISS 30A ESC's flashed w\BLHeli running 6S batteries.  At this point, I just want to get up in the air.  I feel like for every 1/2hr in the air I spend 2 weeks waiting for a FedEx or UPS truck.

    May 9th, 2015 Update I'm back up in the air.  I noticed after I watched the video's from my flight that the problem with "Disarmed" occurred again and it happened about the time the video glitches.  No crash this time.  Since I have changed the software twice, Cleanflight to Baseflight, as well as two different flight controllers.  This is not a software or hardware issue.  This must be a power issue.  Perhaps if I noticed the timing when the flight controller go flakey was timed with the glitchy video it would have occurred to me.  A couple of people on Reddit multi-rotor group mentioned power but I didn't believe it at the time.  I am hoping it's a bad battery and not a cracked wire or PCB.  I think it's the battery.  I remember watching the battery voltage fluctuate under load almost a volt.  My other batteries usually only vary about .1-.2 of a volt or so.  I think I will fly with the other batteries for awhile and see how it goes. I think this project could never be over.  There is always some cool hardware or software improvement to make.  However, I think this is a good time to close out this build log since I'm flying.  Of course, if I make some cool improvements, crashes, or otherwise I will post to my site but as a new post.  This was an incredibly fun project and the technology is amazing.  This is an entirely new security domain and highly relevant in the future.  Drones will be ubiquitous in the future.

    THIS BUILD LOG IS CLOSED, SUCCESS!  Click the "Projects" tag in the tag cloud for more information about my drone adventures.  My closing video.

    Share It!