Yawing issues?? Maybe not! (DX8 woes)

I'm writing this article in hopes of sparing someone from sending back thier radio to be maintenanced.  Also, I would be interested in knowing what RX/TX systems people are having the most luck with.  

For those of you that just want the skinny...  Scroll down to "the Skinny".   If you want the story, just continue reading...

I was convinced that I was having issues setting up the Vector FC.  I contacted Eagle Tree Systems and told them the sytoms.  

• Tail wag on descent.
• RX rudder travel creeping (a clue)

I thought that it was because I had adjusted the rudder trim the first time I flew my Y6.  After reading the setup procedure several more times, I realized I should not touch the rudder's trim.  Not needed.  Reasons why will be discussed when I write my next article about the Vector setup.

ET's response included the following. 

• all motors must be level for FC to work properly.  (mine were slightly off 2-3 degrees)
• Always recalibrate TX after making trim adjustments through the either the windows software or the stick menus via OSD.

I had a couple coincidences that made diagnosing the true issue difficult.  

• creating a new model and going back through a new setup in Vector seemed to fix the issue.  At least for 5-6 flights.
• Only seemed to be an issue with my Y6

It was not until I had wasted an hour driving to a field, setting up all my equipment, then not being able to arm my Y6, that I was frustrated to the point that I had to do something to perminately fix it.

I called Horrizon Hobby.  They said to download the latest software.  That it would most likely NOT fix the issue.  If it didn't, send the unit back.  By this time, resetting the model on the TX did nothing to the travel of the rudder.  It was always coming up 18+.

Before I shipped off my TX to IL and not fly for 3-4 weeks, I took it to the local hobby shop to get thier insight.  

To their suggestion, I split apart the radio and looked for loose wires.  I didn't find anything loose.  So, I unplugged the wire harness for the rudder gimbal and plugged it back in.  Testing proved that I had reduced the travel.  Then it dawned on me.  Gimbal = potentiometer = variable resistance which means the TX is interpretting the bad connections as travel.  I repeated the unplugging and plugging back in several more times and got the travel to read only 1-2+.  Way better than the 38-40 I had at the field. 

This proved to only be a temporary fix.  Next day the TX was acting up again.  This prompted me to do some more research and I found a product called Deoxit that Radio Shack carries.  You'll need some patients with this stuff.  Somehow, it doesn't work 100% immediately.  (you know, we Americans want instant gratification! )

The DX8 has 6 screws on the back of the transmitter that need to be removed to split the case.  I highly recommend removing the battery to prevent accidental shorts.

 Two screws at top. (seen below)

Screws underneith finger grips

And two from the bottom corners.

Then I unplugged the pico board connector and sprayed both the male and female pieces with the Deoxit.  I plugged and unplugged them several times to work the Deoxit into the plug.  I also attempted to spray the potentiometer itself. Although, it looked like it was pretty well sealed up.  Pretty sure nothing would have gotten into the inner workings.

the Skinny

In short, my rudder gimbal had one (a short, if you missed that)

How to recognize the sytoms.

1. Radio trims are centered.

2. Subtrims are centered.

3. Travel is off when sticks are centered.  Note: the RUD is at 20%.  Not good.

To fix, look for loose connections. In my case, my gimbals had been replaced by Horrizon Hobby due to a defect.  So, I was fairly certain that they were actually ok still.  After a quick inspection of the gimbal, I focused on the connection of the plug to the board.  Unplugging and plugging back in improved the connections significantly. Repeating improved.  

Used Deoxit.  Improved a little immediately.  TX was still rudder travel is still 7%.  Much better than 20% and floating around.  Within 24 hours, the radio was in complete working order!! Finally!

I am considering the FrSky Taranis X9D + for it's S-Bus and RSSI signal.  Any other suggestions?

UPDATE:

Well, it happened again! Of course it's never at a convenient time either. This last time, I was flying my QAV250 LOS (line of sight) practicing some tight figure eight movements, trying to build some muscle memory doing coordinated turns for when I am flying FPV (I want to clean up my turns so there's no gain/loss in altitude).

Anyway, my quad got squirrelly. I had to jump over it while it skipped across the driveway under my feet where it came to a rest at the bottom of my old Oak tree!! Stopped it dead in its tracks, it did!

Next morning, I picked up my radio, turned it on and rolled down to the monitor menu to see where my rudder values were (this has now become a habit). 24%!!! 

I took the radio apart for what seemed like the 100th time. More Deoxit...more unplugging...plugging back in.... But this time, those tricks that worked so well the last time had no effect.   

Determined to fix it, I started pulling apart the gimbal assembly to get to the potentiometer in hopes that I could work some Deoxit magic down inside of it. It was while I was tumbling the impenetrable meter in my fingers that I noticed one of the leads didn't look like it had been soldered. I took a closer look under a magnifying glass. Nope. The wire was stuck through the board but not soldered!! Of course, I heated up the iron and made a proper solder connection.

It's not that my DX8 came from the factory floor like this. I had sent it back for a warranty issue with the gimbals (I lost my Trex 600 when the roll went crazy just as I was lifting off). 

I guess the moral of the story is, double check other people's work!

As for the apparent fix I had with the Deoxit, my guess is the Deoxit did in fact deoxidize the connection that was not soldered. In the 2 months since, it oxidized again, causing the short that looked to the RX/TX as a 24% rudder move!