Tips & Techniques
Helpful Articles From Georgia Jets Members
Bench Testing Your Gyros For Thermal Drift
I recently began flying a brand new BobCat that seemed to exhibit very unstable tendencies when going slow with the gear down late in flight. It
snapped on me twice and flying it slow was akin to balancing a ping pong ball atop a pin head. I knew something odd was going on since everybody else was so happy with their BobCat’s rock solid flying characteristics
especially in slow flight. Subsequently, upon returning to the pits following a flight, I noticed that the rudders were deflected significantly. On the next two flights, I ensured that the rudders were lined up perfectly
prior to take off, however, upon landing they were deflected between 1/4" and 3/4"! Uh Oh....
This BobCat was setup with a gyro that was only active with the gear down and the gain was then adjustable with one
of the transmitter’s knobs. I noticed that the off-center rudder deflection was proportional to the gyro gain so I immediately suspected a problem with my gyro. I did some testing and found that my BobCat's gyro was
defective with a phenomena called "thermal drift". Basically, after you turn the gyro on, it heats up internally which can cause the gyro's neutral point to drift.
I will hence forth be checking all
of my gyros out for this phenomena.
I was flying a JR 450 Gyro that I purchased from Horizon so I called Horizon Technical Support and got a testing procedure for detecting thermal
drift. The following procedure assumes that the gyro is connected to the rudders, however, it will work for any other control surface as well.
First, power the receiver (and gyro) up and
verify that the rudders are at neutral trim. Then, leave the radio electronics powered up for 10 minutes to simulate the time elapsed during a flight. Set the gyro’s gain to 100% and look for any rudder
deflection after the 10 minutes has elapsed. You should not observe any rudder deflection. If the rudder goes out of trim at all over time then the gyro has thermal drift and is defective.
JR tests their gyros by also applying heat to the gyro for a few seconds to simulate worse-case in-flight heat buildup and Horizon recommended that I do that as well with a hair dryer. (Do not use a heat gun - Its
too hot!) So, after the 10 minute time period in Test 1, apply heat from a hair dryer for 15 to 20 seconds and note any rudder deflection with the gyro set to 100%. You should not observe more than a very slight
deflection with the use of the hair dryer. If the rudder goes out of trim more than 1/32” then the gyro could be defective.
Test Results With The Defective Gyro
Here are the results for the
defective gyro that was in my BobCat:
Test 1) During the 10 minute time period, the rudders gradually deflected to approximately 3/8" out of trim.
Test 2) Following the 10 minute time period in
Test 1, heat was applied from the hair dryer and the gyro then displaced the rudders over 3/4" past neutral. In addition, after heating the gyro with the hair dryer, the gyro started making the rudder/NG servos oscillate
wildly; Sorta vibrate real fast; Shake, rattle and roll if you know what I mean. Unnerving...
Also, the displacement in both tests were proportional to the gain setting. In other words, at zero gain, the rudders line up
perfectly and then at 100% gain the rudders were skewed to their maximum out-of trim deflection.
Test Results WIth A Gyro In Good Working Order
I then tested a spare JR 450 gyro that I had from my flight box. My
spare gyro didn't deflect at all away from neutral trim after 30 minutes in operation. The hair dryer caused less than 1/32" of deflection and it didn't produce any oscillations or vibrations. So, I installed this
spare gyro into my BobCat which now flies great.
This is a very dangerous phenomena since everything looks perfect when you taxi out to take off but becomes so extremely out of trim
thereafter. I am certain that my defective gyro’s thermal drift is what caused my BobCat to be so unstable. Just imagine flying an airplane around slow with the rudders out of trim between 3/8" and 3/4" while
possibly in some sort of wild oscillation as well. Ugh.....