Monday, September 11, 2017

I'm back. My weekend Flying Session and Multi-Rotor Sale notification

I bet you all thought I disappeared from the scene :) No, I'm still alive. It's just that "life" has been getting in the way of my hobby :) It's been really crazy around here lately.

I think I got to go fly my quads like maybe 4 times all spring/summer. FPV racers are mostly just sitting on shelf and custom-build 550 hexa is still waiting for assembly. I have been maintaining my LiPos though (or, at least checking them).

It was kinda windy this past weekend, but the weather was otherwise good, so I decided to take the Nova over to the middle-school for a little flying.

I flipped though the switches and tried to remember how to do a proper pre-flight and actually fly this thing. It took a while (updating GPS almanac ?) but I finally got a good GPS lock (sats and hdop). I thought it would be OK if I use Manual-Mode sparingly ... instead relying on Loiter, Altitude-Hold, etc.

I had a couple of scary moments, but I flew-thru 2 batteries without any major damage. I also practiced my landings. I was glad to see I could just go fly the Nova after all these months. Not just with my rusty skills, but also that the Nova was still calibrated.

Hey, one of my primary affiliate partners just notified me of a limited time sale on multi-rotors and drones. This link and code are only good for a limited time, but if you need something, you might want to check it out.


Extra 20% Off For Multi-Rotor & Accessories DealsUp To 71% Off For FPV Racer and Multi Rotor Parts.



Saturday, January 7, 2017

Electronics and computerized machines fail early due to Lead-Free Solder and RoHS techniques

Have you ever wondered why electronics and computers from the 90's and early 2000's seem to last forever? But now, you are lucky to get 3 years from just about anything computerized. Now-days, just about any machine (including vehicles, boats, and airplanes) have some kind of logic-board or PCB installed. From coffee-makers to TVs, computers, and spacecraft ... and everything in-between.

So now we know why. It's not just a conspiracy theory. Are the machines you purchased any less valuable or important that the ones that NASA's and the US-Military bought?

Lead-Free Solder - A Train Wreck in the Making

The advisory concludes with this guidance: “Though there are many alternative solder alloys available to replace traditional tin-lead, none of them has passed the reliability testing required of aerospace-quality hardware.”

Killed (Not Saved) By A Whisker - Why Our Electronic Gizmos Inevitably Die

The military and NASA were especially concerned, so they generally operate under waivers allowing lead solder in the gear on which our space program or national security supposedly depend.

Military Suppliers Confront the RoHS Challenge

The two main reliability concerns with lead-free processes are solder joint reliability and tin whiskering ... Most discrete parts and parts with leads can withstand the temperatures of both leaded and lead-free processes, but BGAs can be a problem.

Hazardous Materials Statement - Amphenol Aerospace (PDF)

Due to stringent customer and Mil-Spec requirements, AIAO will not pursue 100% transition to Pb-free products ... MS and other mil-specification part-numbers requiring leaded solder will not be made available in lead-free or RoHS compliant versions until the applicable MilSpec is revised to incorporate this change.

I think this inferior unleaded solder is killing more electronics than we realize. Reliability and longevity seems to have fallen from around 95% to around 60% ... on just about everything. I used to think it was just poorly executed Lead-Free solder techniques, the additional solder-heat weakening the soldered components, and more recently ... a large industry shift to BGA components. While I'm sure all this is still true, now we know the main reason. It only takes one bad/cold solder connection or weak PCB multi-layer-via to render a machine inoperable. Your thoughts?

Wednesday, December 14, 2016

CERT warns of Netgear NightHawk Router Security Flaw, VU 582384 Vulnerability - Exploit News and Fix

I occasionally write about general computer tech that affects our hobby. Well, I think our broadband Internet connections and routers definitely fall into that category. I will add more details to this post as they become available about this exploit.

CERT: Vulnerability Note VU#582384 - Multiple Netgear routers are vulnerable

Netgear's Response to Acew0rm's discovery (also links to updated firmwares)

Naked Security: Netgear router remote control bug – what you need to know

Kalypto: NetGear Vulnerability Expanded

Affected Routers
As at 2016-12-14 ... Netgear listed the following models: R6250, R6400, R6700, R6900, R7000, R7100LG, R7300DST, R7300, R7900, R8000, D6220, D7000. IIRC, all these routers are in the NightHawk series, but watch for this list to possibly change a bit. Don't let this exploit change your opinion of these routers. For consumer-class routers, these are a good value. They are some of the fastest, feature loaded, and most dependable routers available right now (IMO, especially the R6400).

Mainstream News Articles

Wired: A Ton of Popular Netgear Routers Are Exposed—With No Easy Fix

Fortune: Unplug Your Easily Hijacked Netgear Routers Pronto

LifeHacker: More Netgear Routers Found To Be Vulnerable To Super Easy Exploit


My Current Situation
I purchased and installed my Netgear R6400 (AC1750) router in March 2016. An easy choice since I liked my old Netgear WiFi-N router (circa 2010). The (dual-core) Netgear R6400 even has the new NightHawk technology without the extra cost (I did some pre-sales research).

I like to think I use adequate encryption on the WiFi (and Guest Network is OFF). I also use the "wired part" heavily (I also have an 8-port Gigabit Ethernet switch attached). This is because I have a few too many wired devices than the R6400 has ports for, but the R6400 is still the "gate-keeper" for the whole network or LAN. I left the uPnP ON (a first for me, but sure is nice not having to setup Firewall PinHoles or Port-Forwards). I have always kept the "Remote Management" feature OFF. I'm using a DHCP-range to hand-out dynamic IPs (and a few of those IPs are Reserved Addresses) as well as some Fixed IPs beyond that range (for Printers, TVs, etc.). The SPI-class firewall is always ON (because there is no way to accidentally disable it) ... which I like. This SPI Firewall must be on or your network is basically unprotected and completely exposed. All my networking equipment is on an APC UPS unit.

Verifying this Exploit Exists
My current firmware is v1.0.1.6. I used the "Can I test my own router?" exploit test from the Naked Security link above.
I visited this URL in the browser : http://routerlogin.net/cgi-bin/;uname
I got a whole browser page of code (depending on browser, it might look differently in some), but "Linux" is always the last word at the very bottom. That is what they are talking about (the command ran). So with this older firmware, my router definitely has the vulnerability (as expected).

Updating Firmware
These are the steps I followed. I will do this again when final (non-beta) version is released (and continue to keep the firmware updated). Once firmware is "Released as Final" you will be able to upgrade it within the router's Admin Console like normal.
1. Get everyone off the network and the Internet so you can work on it for a few minutes. Better to use a computer connected to network with a real Ethernet wire (not wireless or WiFi).
2. Downloaded updated firmware for my R6400 to my local computer. Unblock file after download.
3a. In router's Admin Console Interface, I backed-up my settings to a local file (just in case - Plan-A).
3b. Save any non-default Router Settings to hard copy (Plan-B). You can write them down, take pics, or save screen-shots.
4. In Advanced, Router Update, Browse to new firmware file and select it. Then click Upload.
5. It will show progress bars of both Updating and its Reboot.
6. Worked fine for me. Router came back online, so I logged back in. The router's Status Page and other settings I viewed still look correct (no loss of data). Logs will be reset. Internet access is working (at least at this one machine).
7. Do not be in a hurry to do next step (you must be absolutely sure firmware is done and router is completely rebooted and working again). Turn off all networking gear (modem, router) for 15 seconds, and then turn back on. This will re-initialize the modem and test broadband for proper new connection negotiation. But mainly, it cold-boots the router now that it's running a new firmware ... this seems to be required on these NightHawk modems after any firmware upgrade.
8. Re-Test machines for network connection, WiFi, printing, Internet access, etc. All mine work fine.

Re-Testing for Exploit (with fixed firmware)
My new current (still in beta) firmware is v1.0.1.18_1.0.15 . Using the above exploit test again, I now only get the router's Admin Login Prompt (waiting for credentials) or a "401 Unauthorized" message in the browser. It could NOT bypass the Admin Login and run a command like before. Looks like the fix is working. All computers and devices are still connecting and Internet access is working fine.

Other Thoughts
As security experts know, any time you choose easy-of-use (or freedoms) over security, you sacrifice some security (protections). In a way, this is what Netgear has done by invoking routerlogin.net (instead of users having to know their router's actual local IP address ... for use on the "safe side" of the router ... aka, their local network). However, since you ARE on the safe-side, I'm not sure that your router's IP address was ever meant to be a true security mechanism.

In general, I have always suggested that router owners (Netgear brand or otherwise) should keep their "Remote Management" options turned OFF. This restricts any router Administration to your local network (and to some extent, possible hacks). Before WiFi, this meant hackers had to be inside the structure with a physical wire connection to your network (but now, they can be next-door with a phone). Everyone should have strong passwords set ... not only on their encrypted WiFi SSID-AccessPoints, but also on their router's Administration Logins.

Finally, I think any router owner (and brand) can use the above URL to test for this exploit. However, if it's not a recent Netgear router, it would be something like http://192.168.1.1/cgi-bin/;uname , where 192.168.1.1 is the IP address of your particular router. Many router manufacturers use the same open-source Linux as a code-base for their routers as well, but its unclear whether the un-patched exploit is in this base-code, or in Netgear's branch.

Updates 
01-17-2017: Looks like R6400's previous beta firmware v1.0.1.18_1.0.15 has been changed to "Release Version Status" without any changes. There is no immediate need to reflash since the files digitally match. At this point, my R6400 is already running the current release version. 

Thursday, October 6, 2016

Nova repairs and updates

Here are the repairs and updates done to the Nova #2. The post will be updated until this aircraft is returned-to-flight.

Current known issues:
1. Repair or replace bad motor and/or ESC in bad set
   - Done here ... it was the ESC.
2. Repair cracked puck and do epoxy mod to reinforce.
   - Done like this
3. Check GPS-module cable. I lost GPS for 5 mins during field testing (and it didn't appear to be an GPS-Almanac update).
    - I lightly filed and cleaned up black corrosion from pins on both the compass and GPS-Module (like this). Re-used existing cables, but did clean them with (plastic-safe) contact cleaner.
4. My old Turnigy 2700mah Lipos have puffed-up to much too fit inside any more.
    - While I had it open, I raised PDB from 2mm to 6mm. I also had to move some things to keep them just barely away from top white body-shell.



Here is the final internal config (right before closing it up). Examine closely to see how I lowered the profile. Obviously, the GPS-Module was moved to be closer to tallest part of dome. Also, I had to remove part of USB cable wire insulation so it would not touch side of shell inside. APM_252 FC should still be "floating" on Zeal (which was the main objective).



Full Calibration results:
No problems with Radio, ESCs, or compass (offsets are now -31, -74, -41).

Having trouble with Accelerometer Calibration in MissionPlanner v1.3.39 and v1.3.41 . I'm currently investigating issue with the more knowledgeable forum members. At first I thought it was MP, but I recently found out that it works fine with a PixHawk v2.4.8 ... so it must be something with this APM_v252 FC or APM in general. It does seem to work thru Terminal though, which is even more puzzling.


Wednesday, September 7, 2016

ESC and/or Motor set failure and troubleshooting - Quanum Nova

As I mentioned in my previous post ... it took two Nova's and 2 years of flights, but it appears I have finally experienced my first (stock) ESC or Motor failure. The post will update as I fumble-through the troubleshooting for the first time :-)



Here are the symptoms the Quanum Nova is displaying with a failed ESC or motor:
  • On power-up, the Nova's power-up sequence of Motor & ESC twitching/ beeping is different and abnormal. It lasts a few seconds longer than normal, but finally stops.
      
  • It still allows you to Arm the aircraft. If you remember to bump the throttle on/off a few times (to make sure all props/motors spin on command) you will likely find one prop not responding properly. If you forget to test and attempt to take-off anyway, the multi-rotor will likely flip and crash.
So I start with the trouble-shooting. The is the newer Nova with the stock v1.6 ESCs. The bad esc/motor set is the Green-LED back-left (aft-port) one.

Here is something I noticed and might further help in troubleshooting.

If you disconnect only the PWM signal cable (the Dupont connector at the FC) of a "suspected bad" esc/motor-set (but leave power going to the ESC), and then power-up the Nova ...
  • The known-BAD set continues to twitch/beep constantly.
     
  • A known-GOOD set does not twitch/beep constantly. It does for it's initial test, then goes quiet like normal.
So, an indication of something. Obviously, the ESC is complaining about something. Only catch with his test is that a motor must be connected to hear the twitch/beeps but you don't actually have to power-up the motor manually (beyond what the ESC does during self-test). I also now see how shorted motors can fry and good ESC on simple power-up-and-test.

The ESC soldering, wiring, and PCB look fine. I see no blown chips or bad soldering.

I opened the motor, and the windings look good. My new LC/LCR meter (a gift) is not working properly for motor coil measurements in uH/mH, so I have another one on order. I will update this post later.

Update 1:
Ok, I got a working LC-Meter. This motor appears to be good because I am getting the following between the 3 motor pairs: 35.8, 35.6, 36.2 uH (aka mH). I took the readings with the magnet-bell removed and they were as they should be (within 10% of each other). Also, windings and magnets physically look good (nothing melted, burnt, or out-of-place).

I installed a used/working Nova v2.4 green-led-ESC. Seems motor solder-pads are in a different orientation than old v1.6 ESCs. I say that because after soldering to match the bad ESC, I still had to swap a motor wire-pair to get this motor to spin in the proper direction. It was no big-deal, but note worthy. It behaves like a good ESC/Motor set now (see above ... both tests ... connected and disconnected). I needed to program the ESCs (all-at-once) and re-do all calibrations.

Toward the end of ESC-Calibration, I observed that all motors now spin properly and turn in the proper directions. There is also a Motor Test in MissionPlanner. I like to do it from Terminal.

More repairs and updates documented here.

Saturday, September 3, 2016

Flight Session 51 - Flying my Novas

My last few flight sessions have been with the new DT-180 drone. Some general LoS flying, but also trying to practice my FPV. Today, I thought I would take my Novas over to the school-field for some flights. Partly cloudy, 90f, and 5mph winds.

Super-Nova (newer Nova #2):

As I get to the field and get everything setup and ready for take-off ... I feel a few sprinkles. I look off in the distance, and a new dark thunder-cloud is approaching. I decide to go for it.

On power-up, I noticed that the Nova's power-up sequence of motor/esc twitch/beeping was different and abnormal. I tried again and got the same. I didn't think much more of it since it Armed ok. I likely bumped the throttle a couple of times to test motor spin, (as I usual do) but not sure. I felt a few more sprinkles so it's now or never. On take-off, it flipped over. Compass puck got cracked. Nova will still Arm, but back/left (aft-port) motor/esc set will not spin prop. Super-Nova definitely not living-up to its name. It never rained.

It took two Nova's and 2 years of flights, but it appears I have finally experienced a stock ESC or Motor failure. Caught me totally off-guard. The Nova was trying to tell me something (a few extended seconds of twitching/beeping of the ultimately non-functioning esc/motor-set on power-up) ... but I didn't understand.

Strange thing about it ... this is the newer Nova (with the v1.6 ESCs), and it 's never been crashed (well, unless you count just now when it flipped on take-off due to a non-spinning motor). Whatever component went bad here, it appears to have happened while Nova was sitting on garage shelf. Either that or it happened on initial take-off/throttle-up/amp-up.

In light of this event, I'm gonna have to add a couple of steps to my main pre-flight checklist.

- Listen for extra (abnormal) beeps/twitching from ESC/Motor sets during initial lipo connection and aircraft power-up (even if it stops after a few seconds).

- After Arming, bump the throttle on/off a few times and make sure all props/motors spin on command.

Because apparently, you can have a failure in a ESC/Motor set and it will still pass pre-arm tests and allow Arming ( basically goes un-detected). Without the pilot performing the above pre-flight observations ... there is no way to prevent a flip/crash/damage when you punch the throttle to take-off.

Current known issues:
1. Repair or replace bad motor and/or ESC in bad set
2. Repair cracked puck and do epoxy mod to reinforce.
3. Check GPS-module cable. I lost GPS for 5 mins during field testing (and it didn't appear to be an GPS-Almanac update).

Nova #1:

Only thing here is that both of my Turnigy-2700 Lipos have now puffed-up again and are apparently too large to fit into the old Nova any more (even with PDB raised-up 2mm). I used the Vant-3000 (that I usually run in the Super-Nova) instead.

The Nova flew great. All modes and Loiter. Quanum Retracts are still working fine. Flew around fast and stable. Fun flying and seems more casual than FPV. Just had the one battery that fit, but got over 12 minutes flight-time on it. GPS working OK for a 6-series. Barometer altitude reporting acceptably today on this quad.

Tuesday, August 30, 2016

Two uBlox GPS & Compass combo-modules for Pixhawk compared

Both of these are advertised as being "GPS combo-modules" with uBlox NEO-M8N GPS and a HMC5883L compass. They are both pinned with Molex PicoBlade for easy PixHawk use. The content of this post started as a rant/complaint, but I've decided to present it as a Comparison, so I can offer a suggested alternative. Ok, so first the bad ...

Beitian GPS & Compass combo-module (white):



Looks pretty on the outside right? Well, looks can be deceiving.

For starters, it only works intermittently. Initial testing revealed it gets more erratic as it warms-up. Inside, we first find a conventional uBlox chipset missing. However, if you look through the hole, you see a chip with uBlox printed on it ... so who knows if it's all real.

Then I noticed that instead of using the available (dependable and industry-standard) Molex PicoBlade connector, they have instead decided to hard-wire solder the tiny cable wires directly to the module's PCB. Seems this critical miss-step allowed the real problem (below) to even manifest itself in the first place.



But the real surprise inside is the soldering (zoom in or "open in new tab/window"). The wires are frayed and barely attached. Not only that, but all the connections are poor, dull, and obvious "cold solder" connections that aren't always making good contact. Everyone knows what happens if your GPS or compass becomes disconnected or glitched in flight ... right, the aircraft flies-away or crashes. This poor soldering is why the module works intermittently, worse as it warms-up, clearly un-reliably, and completely unfit for its purpose.

Finally, there is no obvious indication or arrow pointing forward. Neither on the module's outer case or inner PCB. The compass chip is not visible either.

Lets see what the seller has to say, and I will report back.


GR-BD uBlox GPS & Compass combo-module (black):

http://www.banggood.com/Ublox-NEO-M8N-Flight-Controller-GPS-with-Protective-Shell-for-PIX-PX4-Pixhawk-p-1005394.html?p=NR1603976533201412HJ 

I had this NEO-M8N unit, out on porch getting "warmed-up" for a recent test flight. I was getting SATS:15 & HDOP:0.8 . So, good performance and as expected ... much better than my uBlox 6-series units in same environment.

  


This GR-BD GPS module is much better built. Inside you find a more standard looking uBlox chipset. They also use a properly crimped cable connector to match the one provided on the PCB. This provides a much better connection to vital sensors like these (compared to the white Beitian unit) . In hours of testing, I have never seen this GR-BD module fail. While the proper direction is already clearly marked, the compass chip is also visible. I like the visible blue status led also.

In summary, I guess the real advantage of this black GR-BD one, over the white Beitian one ... is that it actually works :-)