On Wednesday, October 29, 2014 9:51:00 AM UTC-7, jfitch wrote:
On Tuesday, October 28, 2014 9:24:00 PM UTC-7, Andy Blackburn wrote:
On Tuesday, October 28, 2014 5:28:19 PM UTC-7, jfitch wrote:
You are missing my point entirely. A horizontal gust causes actual, real, measurable, and "feelable" vertical acceleration. Ignoring the vario entirely, how can you differentiate it from that acceleration caused by a vertical gust? You cannot without additional information - vertical acceleration is vertical acceleration.

No.

There are transient versus sustained effects that are different for horizontal versus vertical shears (gust versus thermal).

Saying that because a horizontal gust generates lift that it is the same as a thermal that accelerates the glider's frame of reference in a sustained vertical direction is simply incorrect.

9B

Saying that would be simply incorrect - but that is not what I said. A glider is never accelerated in a sustained way. All accelerations the glider experiences are transient, whether induced by a horizontal or vertical gust (excepting turning flight). Once the glider reaches its new velocity, vertical acceleration is zero, regardless of steady state climb rate. This is high school physics. The transient effect is acceleration, this is what you feel. The sustained effect is climb rate, this is what you hope for. But climb rate cannot be felt, only acceleration. When you feel that acceleration, you have about 2 or 3 seconds to determine its cause and react appropriately.

A transient horizontal gust (say ramping quickly from 0 to 10, then back to zero) will be felt as an upward acceleration, followed by a downward acceleration - a bump. But a sustained gust will be felt as an upward acceleration (and an airspeed increase, and a very slight angle of attack reduction, and a lagging variometer up deflection). In nice smooth well behaved air, you might be able to use the more subtle clues to differentiate that from a vertical gust, which will also cause an upward acceleration (and a smaller airspeed increase, a greater angle of attack increase, and perhaps a small momentary lagging downward variometer deflection). In rougher air (mostly what I fly in) sorting this from the noise is practically impossible most of the time. Remembering also that most gusts are neither perfectly vertical nor horizontal, but some random angle in-between.

Of those transient effects, the angle of attack change is probably the easiest to measure, which makes me wonder why this hasn't been pursued more for variometer use. But that signal has a lot of noise in it too.

That's closer to my understanding though I would quibble about some of the details of how an aircraft responds to a horizontal gust.

Assume, for illustration, the gust is 10 knots, and follows the classic "one minus cosine" profile over a second or two. You would see a 10 knot increase in airspeed and if you kept the controls fixed it would activate a modest phugoid response but then be reversed on the back side of the gust. Presuming the glider is flying with the c.g. forward of the center of pressure you should get some onset of upward pitch, but not a lot of immediate g-force as the phugoid is generally a much longer time constant that the short period (AOA) mode. You should also experience some deceleration against the direction of flight from the higher form drag and induced drag due to the change in airspeed, though I suspect this would be harder to pick up than the airspeed change.

With a thermal entry the glider is entering an airmass with vertical velocity that is altered. Again presume 10 knots and in this case also assume it has a rapid onset like the horizontal gust (my experience is that most thermals actually build over a longer time period and are more sustained than horizontal gusts from turbulence but lets make it as similar as possible to tease out the pure differences). The glider experiences two things - a direct vertical acceleration as its inertial reference changes from still air to rising air and it starts to go up directly - this happens pretty quickly, but in the transition it also experiences an increase in angle of attack which activates the short-period longitudinal mode. Given the geometry you can imagine that a vertical air movement has much more of an effect on AOA than a horizontal gust of similar velocity so the sort-period response should be much more energetic.

The other difference is that horizontal gusts tend to look like a "one minus cosine" profile (ramp up and back down) whereas thermal ramp up but don't really ramp back down until you fly out of them several seconds later.

Of course vertical gusts that are not associated with thermals look more like thermals in everything except this symmetric versus asymmetric profile so if the big surge you feel isn't reversed immediately it's more likely a thermal.

If you are familiar with concepts of aircraft dynamics and control theory this article is somewhat informative:

http://scialert.net/fulltext/?doi=srj.2008.17.28

I think to have a vario filter out horizontal gusts you would need to have a dynamic model for the glider and both accelerometers and angular rate gyros plus air data. A simple Kalman filter could then solve for airmass movement and generate a three-dimensional airmass vector in real time. You're only really interested in the Z component so you'd discard the other info unless you were curious about decoding what your body was telling you.

Whether this is the approach vario designers are taking, whether the varios have the sensors to measure all the linear and angular rates and accelerations and whether the effects are pronounced enough to measure clearly amidst all the noise, control inputs and measurement errors and lags I couldn't really say.

9B