I have a problem with a T-38. There is uncommanded roll to the left when the aircraft flies at 250 knots. But as the aircraft speed goes up around 500 knots, the uncommanded roll is now to the right. During 500 knots we could control the plane with 15-20 aileron trims.

What may be the problem?

We have replaced a lot of systems and cylinders. And we changed the travel limits of aileron, flap, rudder, and horizontal stabilizer. But no change has been observed.

  • 8
    $\begingroup$ It would probably help a lot if you indicated what aircraft you were referring to. Obviously, at 500 kt IAS, you're probably not talking about a Cessna 150, but there are a lot of aircraft that will hit 500kt. $\endgroup$ – FreeMan May 28 '19 at 12:08
  • $\begingroup$ Have you inspected the geometry of the wings? $\endgroup$ – ratchet freak May 28 '19 at 12:19
  • 3
    $\begingroup$ You may wish to edit to clarify this -- as posted, it reads that you have left at low speed, then left at high speed. If, as I suspect, they're opposite, this is different from the case where they're the same direction. $\endgroup$ – Zeiss Ikon May 28 '19 at 12:21
  • $\begingroup$ Have you checked the aileron actuating system and associated hydraulic lines? $\endgroup$ – Juan Jimenez May 28 '19 at 13:06
  • 2
    $\begingroup$ So both at 250 and at 500 knots the uncommanded roll is to the left? $\endgroup$ – Koyovis May 28 '19 at 13:24

I once went on-site to troubleshoot an airliner with an uncommanded roll problem. The operator had rigged this and rigged that and were in the process of ordering new aileron power control units, pretty much out of desperation. I asked about the trim actuator, which they had replaced with a known good unit off another aircraft, making them think that couldn't be the problem. I centered the trim and went to look at the position of the rig pin holes in the aileron control circuit. They were mis-aligned.

The trim actuator on this airplane was an electric linear type with an internal Linear Variable Displacement Transducer (LVDT), that supplied the position signal to the cockpit trim indicator (or EICAS in this case), that was "dithering", that is, the signal was drifting around. When you thought the trim was centered, it was actually offset, and this error would randomly change as the actuator was moved. On the original airplane the actuator was installed on, the dithering wasn't enough to cause a snag to be raised by the flight crew so its internal problem went undetected.

The variations in roll you saw at speed may be some other phenomenon, or it just could have been coincidental variations in the trim indication on that flight.

So, if it was me, the trim system itself would be an item to cross off the list before proceeding to more desperate measures. Do a thorough functional test of the trim system and its indication, and also look for things like backlash within the actuator itself or in the linkage.

| improve this answer | |
  • $\begingroup$ It sounds like roll trim is a function of speed, not a random dither. $\endgroup$ – Koyovis May 28 '19 at 13:26
  • 6
    $\begingroup$ Maybe, maybe not. You would have to go out and do a carefully documented series of tests to establish what is going on precisely. Was the shift with speed a direct, repeatable condition or just a one time coincidence. We don't have all the information and are just spitballing here. Golden rule of troubleshooting: eliminate the easy/cheap things first. $\endgroup$ – John K May 28 '19 at 13:38

If this were a model aircraft with a wide speed range (like, say, a hand launch glider), I'd expect this to be a cross-trim problem. Generally, the rudder will have more authority at lower speeds, while aileron will gain authority as speed increases.

Based on that, I'd suggest verifying the rigging of the airframe; look for a fin, rudder, or rudder trim that's generating yaw in the direction of your lower speed uncommanded roll (yaw will produce roll due to whatever mechanism provides roll stability). The aileron trim that offsets that uncommanded yaw will overpower it as speed increases, producing your uncommanded roll in the opposite direction.

| improve this answer | |
  • $\begingroup$ Good answer but if it were a model, I'd suspect that something was broken or loose and moving under the increased aerodynamic load of high speed. Or an incorrect thrust line. $\endgroup$ – Robin Bennett May 29 '19 at 8:40
  • $\begingroup$ @RobinBennett A thrust line problem is tied to power setting; it'll happen during takeoff/launch even more than at high speed (because less aero forces on the surfaces at low speed and high power). Broken or loose will be erratic. $\endgroup$ – Zeiss Ikon May 29 '19 at 10:55

If the flight control is a hydraulic system, have you changed the fluid recently, if yes you need to purge the system for air bubbles, otherwise you need to check the fluid quality for impurity.

| improve this answer | |


This presentation was given in AIAA Houston Section Annual Technical Symposium (ATS) at NASA Johnson Space Center, Gilruth Center. A new discovered root cause for uncommanded roll, pitch and yaw was given.

A technical paper was presented and a live demonstration was shown in the 2019 AIAA AVIATION FORUM and EXPOSITION. A Youtube video shows this phenomenon.

Hope this may help you to understand the uncommanded roll. Actually, it was not "uncommanded". It was indirectly commanded.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.